Merge pull request #1745 from dexidp/remove-vendor

Remove vendor
This commit is contained in:
Márk Sági-Kazár 2020-06-30 21:50:39 +02:00 committed by GitHub
commit eef8c0a60d
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2223 changed files with 4 additions and 1088070 deletions

1
.gitignore vendored
View file

@ -2,3 +2,4 @@ bin
dist
_output
.idea
/vendor/

View file

@ -1,7 +1,4 @@
run:
skip-dirs:
- vendor
timeout: 2m
linters-settings:

View file

@ -41,8 +41,6 @@ $ protoc --go_out=import_path=dexapi:. api.proto
Client programs can then be written using the generated code. A Go client which uses dex's internally generated code might look like the following:
__NOTE:__ Because dex has the `google.golang.org/grpc` package in its `vendor` directory, gRPC code in `github.com/dexidp/dex/api` refers to the vendored copy, not copies in a developers GOPATH. Clients must either regenerate the gRPC Go code or vendor dex and remove its `vendor` directory to run this program.
```
package main

View file

@ -2,7 +2,7 @@
## Go modules
Dex uses [Go modules][go-modules] to manage its [`vendor` directory][go-vendor]. Go 1.11 or higher is recommended. While Go 1.12 is expected to finalize the Go modules feature, with Go 1.11 you should [activate the Go modules feature][go-modules-activate] before interacting with Go modules.
Dex uses [Go modules][go-modules] to manage its dependencies. Go 1.11 or higher is recommended. While Go 1.12 is expected to finalize the Go modules feature, with Go 1.11 you should [activate the Go modules feature][go-modules-activate] before interacting with Go modules.
Here is one way to activate the Go modules feature with Go 1.11.
@ -21,22 +21,9 @@ To add a new dependency to dex or update an existing one:
* Run `go mod tidy -v`. This is a good option if you do not wish to immediately pin to a specific Semantic Version or commit.
* Run, for example, `go get <package-name>@<commit-hash>`. This is a good option when you want to immediately pin to a specific Semantic Version or commit.
* Manually update `go.mod`. If one of the two options above doesn't suit you, do this -- but very carefully.
3. Create a git commit to reflect your code (not vendor) changes. See below for tips on composing commits.
4. Once `go.mod` describes the desired state and you've create a commit for that change, run `make revendor` to update `go.mod`, `go.sum` and `vendor`. This calls `go mod tidy -v`, `go mod vendor -v` and `go mod verify`.
5. Create a git commit to reflect the changes made by `make revendor`. Again, see below for tips on composing commits.
3. Create a git commit to reflect your code changes.
## Composing commits
When composing commits make sure that updates to `vendor` are in a separate commit from the main changes. GitHub's UI makes commits with a large number of changes unreviewable.
Commit histories should look like the following:
```
connector/ldap: add a LDAP connector
vendor: revendor
```
[go-modules]: https://github.com/golang/go/wiki/Modules
[go-modules-activate]: https://github.com/golang/go/wiki/Modules#how-to-install-and-activate-module-support
[go-vendor]: https://golang.org/cmd/go/#hdr-Vendor_Directories
[module-aware-go-get]: https://tip.golang.org/cmd/go/#hdr-Module_aware_go_get

View file

@ -35,12 +35,6 @@ bin/grpc-client:
release-binary:
@go build -o /go/bin/dex -v -ldflags $(LD_FLAGS) $(REPO_PATH)/cmd/dex
.PHONY: revendor
revendor:
@go mod tidy -v
@go mod vendor -v
@go mod verify
test: bin/test/kube-apiserver bin/test/etcd
@go test -v ./...
@ -91,7 +85,7 @@ bin/protoc: scripts/get-protoc
@./scripts/get-protoc bin/protoc
bin/protoc-gen-go:
@go install -v $(REPO_PATH)/vendor/github.com/golang/protobuf/protoc-gen-go
@go install -v github.com/golang/protobuf/protoc-gen-go
clean:
@rm -rf bin/

View file

@ -1,14 +0,0 @@
// +build vendor
package main
// This file exists to trick "go mod vendor" to include "main" packages.
// It is not expected to build, the build tag above is only to prevent this
// file from being included in builds.
import (
_ "github.com/golang/protobuf/protoc-gen-go"
_ "golang.org/x/lint/golint"
)
func main() {}

202
vendor/cloud.google.com/go/LICENSE generated vendored
View file

@ -1,202 +0,0 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

View file

@ -1,513 +0,0 @@
// Copyright 2014 Google LLC
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
// Package metadata provides access to Google Compute Engine (GCE)
// metadata and API service accounts.
//
// This package is a wrapper around the GCE metadata service,
// as documented at https://developers.google.com/compute/docs/metadata.
package metadata // import "cloud.google.com/go/compute/metadata"
import (
"context"
"encoding/json"
"fmt"
"io/ioutil"
"net"
"net/http"
"net/url"
"os"
"runtime"
"strings"
"sync"
"time"
)
const (
// metadataIP is the documented metadata server IP address.
metadataIP = "169.254.169.254"
// metadataHostEnv is the environment variable specifying the
// GCE metadata hostname. If empty, the default value of
// metadataIP ("169.254.169.254") is used instead.
// This is variable name is not defined by any spec, as far as
// I know; it was made up for the Go package.
metadataHostEnv = "GCE_METADATA_HOST"
userAgent = "gcloud-golang/0.1"
)
type cachedValue struct {
k string
trim bool
mu sync.Mutex
v string
}
var (
projID = &cachedValue{k: "project/project-id", trim: true}
projNum = &cachedValue{k: "project/numeric-project-id", trim: true}
instID = &cachedValue{k: "instance/id", trim: true}
)
var (
defaultClient = &Client{hc: &http.Client{
Transport: &http.Transport{
Dial: (&net.Dialer{
Timeout: 2 * time.Second,
KeepAlive: 30 * time.Second,
}).Dial,
ResponseHeaderTimeout: 2 * time.Second,
},
}}
subscribeClient = &Client{hc: &http.Client{
Transport: &http.Transport{
Dial: (&net.Dialer{
Timeout: 2 * time.Second,
KeepAlive: 30 * time.Second,
}).Dial,
},
}}
)
// NotDefinedError is returned when requested metadata is not defined.
//
// The underlying string is the suffix after "/computeMetadata/v1/".
//
// This error is not returned if the value is defined to be the empty
// string.
type NotDefinedError string
func (suffix NotDefinedError) Error() string {
return fmt.Sprintf("metadata: GCE metadata %q not defined", string(suffix))
}
func (c *cachedValue) get(cl *Client) (v string, err error) {
defer c.mu.Unlock()
c.mu.Lock()
if c.v != "" {
return c.v, nil
}
if c.trim {
v, err = cl.getTrimmed(c.k)
} else {
v, err = cl.Get(c.k)
}
if err == nil {
c.v = v
}
return
}
var (
onGCEOnce sync.Once
onGCE bool
)
// OnGCE reports whether this process is running on Google Compute Engine.
func OnGCE() bool {
onGCEOnce.Do(initOnGCE)
return onGCE
}
func initOnGCE() {
onGCE = testOnGCE()
}
func testOnGCE() bool {
// The user explicitly said they're on GCE, so trust them.
if os.Getenv(metadataHostEnv) != "" {
return true
}
ctx, cancel := context.WithCancel(context.Background())
defer cancel()
resc := make(chan bool, 2)
// Try two strategies in parallel.
// See https://github.com/googleapis/google-cloud-go/issues/194
go func() {
req, _ := http.NewRequest("GET", "http://"+metadataIP, nil)
req.Header.Set("User-Agent", userAgent)
res, err := defaultClient.hc.Do(req.WithContext(ctx))
if err != nil {
resc <- false
return
}
defer res.Body.Close()
resc <- res.Header.Get("Metadata-Flavor") == "Google"
}()
go func() {
addrs, err := net.LookupHost("metadata.google.internal")
if err != nil || len(addrs) == 0 {
resc <- false
return
}
resc <- strsContains(addrs, metadataIP)
}()
tryHarder := systemInfoSuggestsGCE()
if tryHarder {
res := <-resc
if res {
// The first strategy succeeded, so let's use it.
return true
}
// Wait for either the DNS or metadata server probe to
// contradict the other one and say we are running on
// GCE. Give it a lot of time to do so, since the system
// info already suggests we're running on a GCE BIOS.
timer := time.NewTimer(5 * time.Second)
defer timer.Stop()
select {
case res = <-resc:
return res
case <-timer.C:
// Too slow. Who knows what this system is.
return false
}
}
// There's no hint from the system info that we're running on
// GCE, so use the first probe's result as truth, whether it's
// true or false. The goal here is to optimize for speed for
// users who are NOT running on GCE. We can't assume that
// either a DNS lookup or an HTTP request to a blackholed IP
// address is fast. Worst case this should return when the
// metaClient's Transport.ResponseHeaderTimeout or
// Transport.Dial.Timeout fires (in two seconds).
return <-resc
}
// systemInfoSuggestsGCE reports whether the local system (without
// doing network requests) suggests that we're running on GCE. If this
// returns true, testOnGCE tries a bit harder to reach its metadata
// server.
func systemInfoSuggestsGCE() bool {
if runtime.GOOS != "linux" {
// We don't have any non-Linux clues available, at least yet.
return false
}
slurp, _ := ioutil.ReadFile("/sys/class/dmi/id/product_name")
name := strings.TrimSpace(string(slurp))
return name == "Google" || name == "Google Compute Engine"
}
// Subscribe calls Client.Subscribe on a client designed for subscribing (one with no
// ResponseHeaderTimeout).
func Subscribe(suffix string, fn func(v string, ok bool) error) error {
return subscribeClient.Subscribe(suffix, fn)
}
// Get calls Client.Get on the default client.
func Get(suffix string) (string, error) { return defaultClient.Get(suffix) }
// ProjectID returns the current instance's project ID string.
func ProjectID() (string, error) { return defaultClient.ProjectID() }
// NumericProjectID returns the current instance's numeric project ID.
func NumericProjectID() (string, error) { return defaultClient.NumericProjectID() }
// InternalIP returns the instance's primary internal IP address.
func InternalIP() (string, error) { return defaultClient.InternalIP() }
// ExternalIP returns the instance's primary external (public) IP address.
func ExternalIP() (string, error) { return defaultClient.ExternalIP() }
// Hostname returns the instance's hostname. This will be of the form
// "<instanceID>.c.<projID>.internal".
func Hostname() (string, error) { return defaultClient.Hostname() }
// InstanceTags returns the list of user-defined instance tags,
// assigned when initially creating a GCE instance.
func InstanceTags() ([]string, error) { return defaultClient.InstanceTags() }
// InstanceID returns the current VM's numeric instance ID.
func InstanceID() (string, error) { return defaultClient.InstanceID() }
// InstanceName returns the current VM's instance ID string.
func InstanceName() (string, error) { return defaultClient.InstanceName() }
// Zone returns the current VM's zone, such as "us-central1-b".
func Zone() (string, error) { return defaultClient.Zone() }
// InstanceAttributes calls Client.InstanceAttributes on the default client.
func InstanceAttributes() ([]string, error) { return defaultClient.InstanceAttributes() }
// ProjectAttributes calls Client.ProjectAttributes on the default client.
func ProjectAttributes() ([]string, error) { return defaultClient.ProjectAttributes() }
// InstanceAttributeValue calls Client.InstanceAttributeValue on the default client.
func InstanceAttributeValue(attr string) (string, error) {
return defaultClient.InstanceAttributeValue(attr)
}
// ProjectAttributeValue calls Client.ProjectAttributeValue on the default client.
func ProjectAttributeValue(attr string) (string, error) {
return defaultClient.ProjectAttributeValue(attr)
}
// Scopes calls Client.Scopes on the default client.
func Scopes(serviceAccount string) ([]string, error) { return defaultClient.Scopes(serviceAccount) }
func strsContains(ss []string, s string) bool {
for _, v := range ss {
if v == s {
return true
}
}
return false
}
// A Client provides metadata.
type Client struct {
hc *http.Client
}
// NewClient returns a Client that can be used to fetch metadata. All HTTP requests
// will use the given http.Client instead of the default client.
func NewClient(c *http.Client) *Client {
return &Client{hc: c}
}
// getETag returns a value from the metadata service as well as the associated ETag.
// This func is otherwise equivalent to Get.
func (c *Client) getETag(suffix string) (value, etag string, err error) {
// Using a fixed IP makes it very difficult to spoof the metadata service in
// a container, which is an important use-case for local testing of cloud
// deployments. To enable spoofing of the metadata service, the environment
// variable GCE_METADATA_HOST is first inspected to decide where metadata
// requests shall go.
host := os.Getenv(metadataHostEnv)
if host == "" {
// Using 169.254.169.254 instead of "metadata" here because Go
// binaries built with the "netgo" tag and without cgo won't
// know the search suffix for "metadata" is
// ".google.internal", and this IP address is documented as
// being stable anyway.
host = metadataIP
}
u := "http://" + host + "/computeMetadata/v1/" + suffix
req, _ := http.NewRequest("GET", u, nil)
req.Header.Set("Metadata-Flavor", "Google")
req.Header.Set("User-Agent", userAgent)
res, err := c.hc.Do(req)
if err != nil {
return "", "", err
}
defer res.Body.Close()
if res.StatusCode == http.StatusNotFound {
return "", "", NotDefinedError(suffix)
}
all, err := ioutil.ReadAll(res.Body)
if err != nil {
return "", "", err
}
if res.StatusCode != 200 {
return "", "", &Error{Code: res.StatusCode, Message: string(all)}
}
return string(all), res.Header.Get("Etag"), nil
}
// Get returns a value from the metadata service.
// The suffix is appended to "http://${GCE_METADATA_HOST}/computeMetadata/v1/".
//
// If the GCE_METADATA_HOST environment variable is not defined, a default of
// 169.254.169.254 will be used instead.
//
// If the requested metadata is not defined, the returned error will
// be of type NotDefinedError.
func (c *Client) Get(suffix string) (string, error) {
val, _, err := c.getETag(suffix)
return val, err
}
func (c *Client) getTrimmed(suffix string) (s string, err error) {
s, err = c.Get(suffix)
s = strings.TrimSpace(s)
return
}
func (c *Client) lines(suffix string) ([]string, error) {
j, err := c.Get(suffix)
if err != nil {
return nil, err
}
s := strings.Split(strings.TrimSpace(j), "\n")
for i := range s {
s[i] = strings.TrimSpace(s[i])
}
return s, nil
}
// ProjectID returns the current instance's project ID string.
func (c *Client) ProjectID() (string, error) { return projID.get(c) }
// NumericProjectID returns the current instance's numeric project ID.
func (c *Client) NumericProjectID() (string, error) { return projNum.get(c) }
// InstanceID returns the current VM's numeric instance ID.
func (c *Client) InstanceID() (string, error) { return instID.get(c) }
// InternalIP returns the instance's primary internal IP address.
func (c *Client) InternalIP() (string, error) {
return c.getTrimmed("instance/network-interfaces/0/ip")
}
// ExternalIP returns the instance's primary external (public) IP address.
func (c *Client) ExternalIP() (string, error) {
return c.getTrimmed("instance/network-interfaces/0/access-configs/0/external-ip")
}
// Hostname returns the instance's hostname. This will be of the form
// "<instanceID>.c.<projID>.internal".
func (c *Client) Hostname() (string, error) {
return c.getTrimmed("instance/hostname")
}
// InstanceTags returns the list of user-defined instance tags,
// assigned when initially creating a GCE instance.
func (c *Client) InstanceTags() ([]string, error) {
var s []string
j, err := c.Get("instance/tags")
if err != nil {
return nil, err
}
if err := json.NewDecoder(strings.NewReader(j)).Decode(&s); err != nil {
return nil, err
}
return s, nil
}
// InstanceName returns the current VM's instance ID string.
func (c *Client) InstanceName() (string, error) {
host, err := c.Hostname()
if err != nil {
return "", err
}
return strings.Split(host, ".")[0], nil
}
// Zone returns the current VM's zone, such as "us-central1-b".
func (c *Client) Zone() (string, error) {
zone, err := c.getTrimmed("instance/zone")
// zone is of the form "projects/<projNum>/zones/<zoneName>".
if err != nil {
return "", err
}
return zone[strings.LastIndex(zone, "/")+1:], nil
}
// InstanceAttributes returns the list of user-defined attributes,
// assigned when initially creating a GCE VM instance. The value of an
// attribute can be obtained with InstanceAttributeValue.
func (c *Client) InstanceAttributes() ([]string, error) { return c.lines("instance/attributes/") }
// ProjectAttributes returns the list of user-defined attributes
// applying to the project as a whole, not just this VM. The value of
// an attribute can be obtained with ProjectAttributeValue.
func (c *Client) ProjectAttributes() ([]string, error) { return c.lines("project/attributes/") }
// InstanceAttributeValue returns the value of the provided VM
// instance attribute.
//
// If the requested attribute is not defined, the returned error will
// be of type NotDefinedError.
//
// InstanceAttributeValue may return ("", nil) if the attribute was
// defined to be the empty string.
func (c *Client) InstanceAttributeValue(attr string) (string, error) {
return c.Get("instance/attributes/" + attr)
}
// ProjectAttributeValue returns the value of the provided
// project attribute.
//
// If the requested attribute is not defined, the returned error will
// be of type NotDefinedError.
//
// ProjectAttributeValue may return ("", nil) if the attribute was
// defined to be the empty string.
func (c *Client) ProjectAttributeValue(attr string) (string, error) {
return c.Get("project/attributes/" + attr)
}
// Scopes returns the service account scopes for the given account.
// The account may be empty or the string "default" to use the instance's
// main account.
func (c *Client) Scopes(serviceAccount string) ([]string, error) {
if serviceAccount == "" {
serviceAccount = "default"
}
return c.lines("instance/service-accounts/" + serviceAccount + "/scopes")
}
// Subscribe subscribes to a value from the metadata service.
// The suffix is appended to "http://${GCE_METADATA_HOST}/computeMetadata/v1/".
// The suffix may contain query parameters.
//
// Subscribe calls fn with the latest metadata value indicated by the provided
// suffix. If the metadata value is deleted, fn is called with the empty string
// and ok false. Subscribe blocks until fn returns a non-nil error or the value
// is deleted. Subscribe returns the error value returned from the last call to
// fn, which may be nil when ok == false.
func (c *Client) Subscribe(suffix string, fn func(v string, ok bool) error) error {
const failedSubscribeSleep = time.Second * 5
// First check to see if the metadata value exists at all.
val, lastETag, err := c.getETag(suffix)
if err != nil {
return err
}
if err := fn(val, true); err != nil {
return err
}
ok := true
if strings.ContainsRune(suffix, '?') {
suffix += "&wait_for_change=true&last_etag="
} else {
suffix += "?wait_for_change=true&last_etag="
}
for {
val, etag, err := c.getETag(suffix + url.QueryEscape(lastETag))
if err != nil {
if _, deleted := err.(NotDefinedError); !deleted {
time.Sleep(failedSubscribeSleep)
continue // Retry on other errors.
}
ok = false
}
lastETag = etag
if err := fn(val, ok); err != nil || !ok {
return err
}
}
}
// Error contains an error response from the server.
type Error struct {
// Code is the HTTP response status code.
Code int
// Message is the server response message.
Message string
}
func (e *Error) Error() string {
return fmt.Sprintf("compute: Received %d `%s`", e.Code, e.Message)
}

View file

@ -1 +0,0 @@
*.exe

View file

@ -1,22 +0,0 @@
The MIT License (MIT)
Copyright (c) 2015 Microsoft
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

View file

@ -1,22 +0,0 @@
# go-winio
This repository contains utilities for efficiently performing Win32 IO operations in
Go. Currently, this is focused on accessing named pipes and other file handles, and
for using named pipes as a net transport.
This code relies on IO completion ports to avoid blocking IO on system threads, allowing Go
to reuse the thread to schedule another goroutine. This limits support to Windows Vista and
newer operating systems. This is similar to the implementation of network sockets in Go's net
package.
Please see the LICENSE file for licensing information.
This project has adopted the [Microsoft Open Source Code of
Conduct](https://opensource.microsoft.com/codeofconduct/). For more information
see the [Code of Conduct
FAQ](https://opensource.microsoft.com/codeofconduct/faq/) or contact
[opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional
questions or comments.
Thanks to natefinch for the inspiration for this library. See https://github.com/natefinch/npipe
for another named pipe implementation.

View file

@ -1,280 +0,0 @@
// +build windows
package winio
import (
"encoding/binary"
"errors"
"fmt"
"io"
"io/ioutil"
"os"
"runtime"
"syscall"
"unicode/utf16"
)
//sys backupRead(h syscall.Handle, b []byte, bytesRead *uint32, abort bool, processSecurity bool, context *uintptr) (err error) = BackupRead
//sys backupWrite(h syscall.Handle, b []byte, bytesWritten *uint32, abort bool, processSecurity bool, context *uintptr) (err error) = BackupWrite
const (
BackupData = uint32(iota + 1)
BackupEaData
BackupSecurity
BackupAlternateData
BackupLink
BackupPropertyData
BackupObjectId
BackupReparseData
BackupSparseBlock
BackupTxfsData
)
const (
StreamSparseAttributes = uint32(8)
)
const (
WRITE_DAC = 0x40000
WRITE_OWNER = 0x80000
ACCESS_SYSTEM_SECURITY = 0x1000000
)
// BackupHeader represents a backup stream of a file.
type BackupHeader struct {
Id uint32 // The backup stream ID
Attributes uint32 // Stream attributes
Size int64 // The size of the stream in bytes
Name string // The name of the stream (for BackupAlternateData only).
Offset int64 // The offset of the stream in the file (for BackupSparseBlock only).
}
type win32StreamId struct {
StreamId uint32
Attributes uint32
Size uint64
NameSize uint32
}
// BackupStreamReader reads from a stream produced by the BackupRead Win32 API and produces a series
// of BackupHeader values.
type BackupStreamReader struct {
r io.Reader
bytesLeft int64
}
// NewBackupStreamReader produces a BackupStreamReader from any io.Reader.
func NewBackupStreamReader(r io.Reader) *BackupStreamReader {
return &BackupStreamReader{r, 0}
}
// Next returns the next backup stream and prepares for calls to Read(). It skips the remainder of the current stream if
// it was not completely read.
func (r *BackupStreamReader) Next() (*BackupHeader, error) {
if r.bytesLeft > 0 {
if s, ok := r.r.(io.Seeker); ok {
// Make sure Seek on io.SeekCurrent sometimes succeeds
// before trying the actual seek.
if _, err := s.Seek(0, io.SeekCurrent); err == nil {
if _, err = s.Seek(r.bytesLeft, io.SeekCurrent); err != nil {
return nil, err
}
r.bytesLeft = 0
}
}
if _, err := io.Copy(ioutil.Discard, r); err != nil {
return nil, err
}
}
var wsi win32StreamId
if err := binary.Read(r.r, binary.LittleEndian, &wsi); err != nil {
return nil, err
}
hdr := &BackupHeader{
Id: wsi.StreamId,
Attributes: wsi.Attributes,
Size: int64(wsi.Size),
}
if wsi.NameSize != 0 {
name := make([]uint16, int(wsi.NameSize/2))
if err := binary.Read(r.r, binary.LittleEndian, name); err != nil {
return nil, err
}
hdr.Name = syscall.UTF16ToString(name)
}
if wsi.StreamId == BackupSparseBlock {
if err := binary.Read(r.r, binary.LittleEndian, &hdr.Offset); err != nil {
return nil, err
}
hdr.Size -= 8
}
r.bytesLeft = hdr.Size
return hdr, nil
}
// Read reads from the current backup stream.
func (r *BackupStreamReader) Read(b []byte) (int, error) {
if r.bytesLeft == 0 {
return 0, io.EOF
}
if int64(len(b)) > r.bytesLeft {
b = b[:r.bytesLeft]
}
n, err := r.r.Read(b)
r.bytesLeft -= int64(n)
if err == io.EOF {
err = io.ErrUnexpectedEOF
} else if r.bytesLeft == 0 && err == nil {
err = io.EOF
}
return n, err
}
// BackupStreamWriter writes a stream compatible with the BackupWrite Win32 API.
type BackupStreamWriter struct {
w io.Writer
bytesLeft int64
}
// NewBackupStreamWriter produces a BackupStreamWriter on top of an io.Writer.
func NewBackupStreamWriter(w io.Writer) *BackupStreamWriter {
return &BackupStreamWriter{w, 0}
}
// WriteHeader writes the next backup stream header and prepares for calls to Write().
func (w *BackupStreamWriter) WriteHeader(hdr *BackupHeader) error {
if w.bytesLeft != 0 {
return fmt.Errorf("missing %d bytes", w.bytesLeft)
}
name := utf16.Encode([]rune(hdr.Name))
wsi := win32StreamId{
StreamId: hdr.Id,
Attributes: hdr.Attributes,
Size: uint64(hdr.Size),
NameSize: uint32(len(name) * 2),
}
if hdr.Id == BackupSparseBlock {
// Include space for the int64 block offset
wsi.Size += 8
}
if err := binary.Write(w.w, binary.LittleEndian, &wsi); err != nil {
return err
}
if len(name) != 0 {
if err := binary.Write(w.w, binary.LittleEndian, name); err != nil {
return err
}
}
if hdr.Id == BackupSparseBlock {
if err := binary.Write(w.w, binary.LittleEndian, hdr.Offset); err != nil {
return err
}
}
w.bytesLeft = hdr.Size
return nil
}
// Write writes to the current backup stream.
func (w *BackupStreamWriter) Write(b []byte) (int, error) {
if w.bytesLeft < int64(len(b)) {
return 0, fmt.Errorf("too many bytes by %d", int64(len(b))-w.bytesLeft)
}
n, err := w.w.Write(b)
w.bytesLeft -= int64(n)
return n, err
}
// BackupFileReader provides an io.ReadCloser interface on top of the BackupRead Win32 API.
type BackupFileReader struct {
f *os.File
includeSecurity bool
ctx uintptr
}
// NewBackupFileReader returns a new BackupFileReader from a file handle. If includeSecurity is true,
// Read will attempt to read the security descriptor of the file.
func NewBackupFileReader(f *os.File, includeSecurity bool) *BackupFileReader {
r := &BackupFileReader{f, includeSecurity, 0}
return r
}
// Read reads a backup stream from the file by calling the Win32 API BackupRead().
func (r *BackupFileReader) Read(b []byte) (int, error) {
var bytesRead uint32
err := backupRead(syscall.Handle(r.f.Fd()), b, &bytesRead, false, r.includeSecurity, &r.ctx)
if err != nil {
return 0, &os.PathError{"BackupRead", r.f.Name(), err}
}
runtime.KeepAlive(r.f)
if bytesRead == 0 {
return 0, io.EOF
}
return int(bytesRead), nil
}
// Close frees Win32 resources associated with the BackupFileReader. It does not close
// the underlying file.
func (r *BackupFileReader) Close() error {
if r.ctx != 0 {
backupRead(syscall.Handle(r.f.Fd()), nil, nil, true, false, &r.ctx)
runtime.KeepAlive(r.f)
r.ctx = 0
}
return nil
}
// BackupFileWriter provides an io.WriteCloser interface on top of the BackupWrite Win32 API.
type BackupFileWriter struct {
f *os.File
includeSecurity bool
ctx uintptr
}
// NewBackupFileWriter returns a new BackupFileWriter from a file handle. If includeSecurity is true,
// Write() will attempt to restore the security descriptor from the stream.
func NewBackupFileWriter(f *os.File, includeSecurity bool) *BackupFileWriter {
w := &BackupFileWriter{f, includeSecurity, 0}
return w
}
// Write restores a portion of the file using the provided backup stream.
func (w *BackupFileWriter) Write(b []byte) (int, error) {
var bytesWritten uint32
err := backupWrite(syscall.Handle(w.f.Fd()), b, &bytesWritten, false, w.includeSecurity, &w.ctx)
if err != nil {
return 0, &os.PathError{"BackupWrite", w.f.Name(), err}
}
runtime.KeepAlive(w.f)
if int(bytesWritten) != len(b) {
return int(bytesWritten), errors.New("not all bytes could be written")
}
return len(b), nil
}
// Close frees Win32 resources associated with the BackupFileWriter. It does not
// close the underlying file.
func (w *BackupFileWriter) Close() error {
if w.ctx != 0 {
backupWrite(syscall.Handle(w.f.Fd()), nil, nil, true, false, &w.ctx)
runtime.KeepAlive(w.f)
w.ctx = 0
}
return nil
}
// OpenForBackup opens a file or directory, potentially skipping access checks if the backup
// or restore privileges have been acquired.
//
// If the file opened was a directory, it cannot be used with Readdir().
func OpenForBackup(path string, access uint32, share uint32, createmode uint32) (*os.File, error) {
winPath, err := syscall.UTF16FromString(path)
if err != nil {
return nil, err
}
h, err := syscall.CreateFile(&winPath[0], access, share, nil, createmode, syscall.FILE_FLAG_BACKUP_SEMANTICS|syscall.FILE_FLAG_OPEN_REPARSE_POINT, 0)
if err != nil {
err = &os.PathError{Op: "open", Path: path, Err: err}
return nil, err
}
return os.NewFile(uintptr(h), path), nil
}

View file

@ -1,137 +0,0 @@
package winio
import (
"bytes"
"encoding/binary"
"errors"
)
type fileFullEaInformation struct {
NextEntryOffset uint32
Flags uint8
NameLength uint8
ValueLength uint16
}
var (
fileFullEaInformationSize = binary.Size(&fileFullEaInformation{})
errInvalidEaBuffer = errors.New("invalid extended attribute buffer")
errEaNameTooLarge = errors.New("extended attribute name too large")
errEaValueTooLarge = errors.New("extended attribute value too large")
)
// ExtendedAttribute represents a single Windows EA.
type ExtendedAttribute struct {
Name string
Value []byte
Flags uint8
}
func parseEa(b []byte) (ea ExtendedAttribute, nb []byte, err error) {
var info fileFullEaInformation
err = binary.Read(bytes.NewReader(b), binary.LittleEndian, &info)
if err != nil {
err = errInvalidEaBuffer
return
}
nameOffset := fileFullEaInformationSize
nameLen := int(info.NameLength)
valueOffset := nameOffset + int(info.NameLength) + 1
valueLen := int(info.ValueLength)
nextOffset := int(info.NextEntryOffset)
if valueLen+valueOffset > len(b) || nextOffset < 0 || nextOffset > len(b) {
err = errInvalidEaBuffer
return
}
ea.Name = string(b[nameOffset : nameOffset+nameLen])
ea.Value = b[valueOffset : valueOffset+valueLen]
ea.Flags = info.Flags
if info.NextEntryOffset != 0 {
nb = b[info.NextEntryOffset:]
}
return
}
// DecodeExtendedAttributes decodes a list of EAs from a FILE_FULL_EA_INFORMATION
// buffer retrieved from BackupRead, ZwQueryEaFile, etc.
func DecodeExtendedAttributes(b []byte) (eas []ExtendedAttribute, err error) {
for len(b) != 0 {
ea, nb, err := parseEa(b)
if err != nil {
return nil, err
}
eas = append(eas, ea)
b = nb
}
return
}
func writeEa(buf *bytes.Buffer, ea *ExtendedAttribute, last bool) error {
if int(uint8(len(ea.Name))) != len(ea.Name) {
return errEaNameTooLarge
}
if int(uint16(len(ea.Value))) != len(ea.Value) {
return errEaValueTooLarge
}
entrySize := uint32(fileFullEaInformationSize + len(ea.Name) + 1 + len(ea.Value))
withPadding := (entrySize + 3) &^ 3
nextOffset := uint32(0)
if !last {
nextOffset = withPadding
}
info := fileFullEaInformation{
NextEntryOffset: nextOffset,
Flags: ea.Flags,
NameLength: uint8(len(ea.Name)),
ValueLength: uint16(len(ea.Value)),
}
err := binary.Write(buf, binary.LittleEndian, &info)
if err != nil {
return err
}
_, err = buf.Write([]byte(ea.Name))
if err != nil {
return err
}
err = buf.WriteByte(0)
if err != nil {
return err
}
_, err = buf.Write(ea.Value)
if err != nil {
return err
}
_, err = buf.Write([]byte{0, 0, 0}[0 : withPadding-entrySize])
if err != nil {
return err
}
return nil
}
// EncodeExtendedAttributes encodes a list of EAs into a FILE_FULL_EA_INFORMATION
// buffer for use with BackupWrite, ZwSetEaFile, etc.
func EncodeExtendedAttributes(eas []ExtendedAttribute) ([]byte, error) {
var buf bytes.Buffer
for i := range eas {
last := false
if i == len(eas)-1 {
last = true
}
err := writeEa(&buf, &eas[i], last)
if err != nil {
return nil, err
}
}
return buf.Bytes(), nil
}

View file

@ -1,323 +0,0 @@
// +build windows
package winio
import (
"errors"
"io"
"runtime"
"sync"
"sync/atomic"
"syscall"
"time"
)
//sys cancelIoEx(file syscall.Handle, o *syscall.Overlapped) (err error) = CancelIoEx
//sys createIoCompletionPort(file syscall.Handle, port syscall.Handle, key uintptr, threadCount uint32) (newport syscall.Handle, err error) = CreateIoCompletionPort
//sys getQueuedCompletionStatus(port syscall.Handle, bytes *uint32, key *uintptr, o **ioOperation, timeout uint32) (err error) = GetQueuedCompletionStatus
//sys setFileCompletionNotificationModes(h syscall.Handle, flags uint8) (err error) = SetFileCompletionNotificationModes
//sys wsaGetOverlappedResult(h syscall.Handle, o *syscall.Overlapped, bytes *uint32, wait bool, flags *uint32) (err error) = ws2_32.WSAGetOverlappedResult
type atomicBool int32
func (b *atomicBool) isSet() bool { return atomic.LoadInt32((*int32)(b)) != 0 }
func (b *atomicBool) setFalse() { atomic.StoreInt32((*int32)(b), 0) }
func (b *atomicBool) setTrue() { atomic.StoreInt32((*int32)(b), 1) }
func (b *atomicBool) swap(new bool) bool {
var newInt int32
if new {
newInt = 1
}
return atomic.SwapInt32((*int32)(b), newInt) == 1
}
const (
cFILE_SKIP_COMPLETION_PORT_ON_SUCCESS = 1
cFILE_SKIP_SET_EVENT_ON_HANDLE = 2
)
var (
ErrFileClosed = errors.New("file has already been closed")
ErrTimeout = &timeoutError{}
)
type timeoutError struct{}
func (e *timeoutError) Error() string { return "i/o timeout" }
func (e *timeoutError) Timeout() bool { return true }
func (e *timeoutError) Temporary() bool { return true }
type timeoutChan chan struct{}
var ioInitOnce sync.Once
var ioCompletionPort syscall.Handle
// ioResult contains the result of an asynchronous IO operation
type ioResult struct {
bytes uint32
err error
}
// ioOperation represents an outstanding asynchronous Win32 IO
type ioOperation struct {
o syscall.Overlapped
ch chan ioResult
}
func initIo() {
h, err := createIoCompletionPort(syscall.InvalidHandle, 0, 0, 0xffffffff)
if err != nil {
panic(err)
}
ioCompletionPort = h
go ioCompletionProcessor(h)
}
// win32File implements Reader, Writer, and Closer on a Win32 handle without blocking in a syscall.
// It takes ownership of this handle and will close it if it is garbage collected.
type win32File struct {
handle syscall.Handle
wg sync.WaitGroup
wgLock sync.RWMutex
closing atomicBool
socket bool
readDeadline deadlineHandler
writeDeadline deadlineHandler
}
type deadlineHandler struct {
setLock sync.Mutex
channel timeoutChan
channelLock sync.RWMutex
timer *time.Timer
timedout atomicBool
}
// makeWin32File makes a new win32File from an existing file handle
func makeWin32File(h syscall.Handle) (*win32File, error) {
f := &win32File{handle: h}
ioInitOnce.Do(initIo)
_, err := createIoCompletionPort(h, ioCompletionPort, 0, 0xffffffff)
if err != nil {
return nil, err
}
err = setFileCompletionNotificationModes(h, cFILE_SKIP_COMPLETION_PORT_ON_SUCCESS|cFILE_SKIP_SET_EVENT_ON_HANDLE)
if err != nil {
return nil, err
}
f.readDeadline.channel = make(timeoutChan)
f.writeDeadline.channel = make(timeoutChan)
return f, nil
}
func MakeOpenFile(h syscall.Handle) (io.ReadWriteCloser, error) {
// If we return the result of makeWin32File directly, it can result in an
// interface-wrapped nil, rather than a nil interface value.
f, err := makeWin32File(h)
if err != nil {
return nil, err
}
return f, nil
}
// closeHandle closes the resources associated with a Win32 handle
func (f *win32File) closeHandle() {
f.wgLock.Lock()
// Atomically set that we are closing, releasing the resources only once.
if !f.closing.swap(true) {
f.wgLock.Unlock()
// cancel all IO and wait for it to complete
cancelIoEx(f.handle, nil)
f.wg.Wait()
// at this point, no new IO can start
syscall.Close(f.handle)
f.handle = 0
} else {
f.wgLock.Unlock()
}
}
// Close closes a win32File.
func (f *win32File) Close() error {
f.closeHandle()
return nil
}
// prepareIo prepares for a new IO operation.
// The caller must call f.wg.Done() when the IO is finished, prior to Close() returning.
func (f *win32File) prepareIo() (*ioOperation, error) {
f.wgLock.RLock()
if f.closing.isSet() {
f.wgLock.RUnlock()
return nil, ErrFileClosed
}
f.wg.Add(1)
f.wgLock.RUnlock()
c := &ioOperation{}
c.ch = make(chan ioResult)
return c, nil
}
// ioCompletionProcessor processes completed async IOs forever
func ioCompletionProcessor(h syscall.Handle) {
for {
var bytes uint32
var key uintptr
var op *ioOperation
err := getQueuedCompletionStatus(h, &bytes, &key, &op, syscall.INFINITE)
if op == nil {
panic(err)
}
op.ch <- ioResult{bytes, err}
}
}
// asyncIo processes the return value from ReadFile or WriteFile, blocking until
// the operation has actually completed.
func (f *win32File) asyncIo(c *ioOperation, d *deadlineHandler, bytes uint32, err error) (int, error) {
if err != syscall.ERROR_IO_PENDING {
return int(bytes), err
}
if f.closing.isSet() {
cancelIoEx(f.handle, &c.o)
}
var timeout timeoutChan
if d != nil {
d.channelLock.Lock()
timeout = d.channel
d.channelLock.Unlock()
}
var r ioResult
select {
case r = <-c.ch:
err = r.err
if err == syscall.ERROR_OPERATION_ABORTED {
if f.closing.isSet() {
err = ErrFileClosed
}
} else if err != nil && f.socket {
// err is from Win32. Query the overlapped structure to get the winsock error.
var bytes, flags uint32
err = wsaGetOverlappedResult(f.handle, &c.o, &bytes, false, &flags)
}
case <-timeout:
cancelIoEx(f.handle, &c.o)
r = <-c.ch
err = r.err
if err == syscall.ERROR_OPERATION_ABORTED {
err = ErrTimeout
}
}
// runtime.KeepAlive is needed, as c is passed via native
// code to ioCompletionProcessor, c must remain alive
// until the channel read is complete.
runtime.KeepAlive(c)
return int(r.bytes), err
}
// Read reads from a file handle.
func (f *win32File) Read(b []byte) (int, error) {
c, err := f.prepareIo()
if err != nil {
return 0, err
}
defer f.wg.Done()
if f.readDeadline.timedout.isSet() {
return 0, ErrTimeout
}
var bytes uint32
err = syscall.ReadFile(f.handle, b, &bytes, &c.o)
n, err := f.asyncIo(c, &f.readDeadline, bytes, err)
runtime.KeepAlive(b)
// Handle EOF conditions.
if err == nil && n == 0 && len(b) != 0 {
return 0, io.EOF
} else if err == syscall.ERROR_BROKEN_PIPE {
return 0, io.EOF
} else {
return n, err
}
}
// Write writes to a file handle.
func (f *win32File) Write(b []byte) (int, error) {
c, err := f.prepareIo()
if err != nil {
return 0, err
}
defer f.wg.Done()
if f.writeDeadline.timedout.isSet() {
return 0, ErrTimeout
}
var bytes uint32
err = syscall.WriteFile(f.handle, b, &bytes, &c.o)
n, err := f.asyncIo(c, &f.writeDeadline, bytes, err)
runtime.KeepAlive(b)
return n, err
}
func (f *win32File) SetReadDeadline(deadline time.Time) error {
return f.readDeadline.set(deadline)
}
func (f *win32File) SetWriteDeadline(deadline time.Time) error {
return f.writeDeadline.set(deadline)
}
func (f *win32File) Flush() error {
return syscall.FlushFileBuffers(f.handle)
}
func (f *win32File) Fd() uintptr {
return uintptr(f.handle)
}
func (d *deadlineHandler) set(deadline time.Time) error {
d.setLock.Lock()
defer d.setLock.Unlock()
if d.timer != nil {
if !d.timer.Stop() {
<-d.channel
}
d.timer = nil
}
d.timedout.setFalse()
select {
case <-d.channel:
d.channelLock.Lock()
d.channel = make(chan struct{})
d.channelLock.Unlock()
default:
}
if deadline.IsZero() {
return nil
}
timeoutIO := func() {
d.timedout.setTrue()
close(d.channel)
}
now := time.Now()
duration := deadline.Sub(now)
if deadline.After(now) {
// Deadline is in the future, set a timer to wait
d.timer = time.AfterFunc(duration, timeoutIO)
} else {
// Deadline is in the past. Cancel all pending IO now.
timeoutIO()
}
return nil
}

View file

@ -1,61 +0,0 @@
// +build windows
package winio
import (
"os"
"runtime"
"syscall"
"unsafe"
)
//sys getFileInformationByHandleEx(h syscall.Handle, class uint32, buffer *byte, size uint32) (err error) = GetFileInformationByHandleEx
//sys setFileInformationByHandle(h syscall.Handle, class uint32, buffer *byte, size uint32) (err error) = SetFileInformationByHandle
const (
fileBasicInfo = 0
fileIDInfo = 0x12
)
// FileBasicInfo contains file access time and file attributes information.
type FileBasicInfo struct {
CreationTime, LastAccessTime, LastWriteTime, ChangeTime syscall.Filetime
FileAttributes uint32
pad uint32 // padding
}
// GetFileBasicInfo retrieves times and attributes for a file.
func GetFileBasicInfo(f *os.File) (*FileBasicInfo, error) {
bi := &FileBasicInfo{}
if err := getFileInformationByHandleEx(syscall.Handle(f.Fd()), fileBasicInfo, (*byte)(unsafe.Pointer(bi)), uint32(unsafe.Sizeof(*bi))); err != nil {
return nil, &os.PathError{Op: "GetFileInformationByHandleEx", Path: f.Name(), Err: err}
}
runtime.KeepAlive(f)
return bi, nil
}
// SetFileBasicInfo sets times and attributes for a file.
func SetFileBasicInfo(f *os.File, bi *FileBasicInfo) error {
if err := setFileInformationByHandle(syscall.Handle(f.Fd()), fileBasicInfo, (*byte)(unsafe.Pointer(bi)), uint32(unsafe.Sizeof(*bi))); err != nil {
return &os.PathError{Op: "SetFileInformationByHandle", Path: f.Name(), Err: err}
}
runtime.KeepAlive(f)
return nil
}
// FileIDInfo contains the volume serial number and file ID for a file. This pair should be
// unique on a system.
type FileIDInfo struct {
VolumeSerialNumber uint64
FileID [16]byte
}
// GetFileID retrieves the unique (volume, file ID) pair for a file.
func GetFileID(f *os.File) (*FileIDInfo, error) {
fileID := &FileIDInfo{}
if err := getFileInformationByHandleEx(syscall.Handle(f.Fd()), fileIDInfo, (*byte)(unsafe.Pointer(fileID)), uint32(unsafe.Sizeof(*fileID))); err != nil {
return nil, &os.PathError{Op: "GetFileInformationByHandleEx", Path: f.Name(), Err: err}
}
runtime.KeepAlive(f)
return fileID, nil
}

View file

@ -1,9 +0,0 @@
module github.com/Microsoft/go-winio
go 1.12
require (
github.com/pkg/errors v0.8.1
github.com/sirupsen/logrus v1.4.1
golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3
)

View file

@ -1,18 +0,0 @@
github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
github.com/konsorten/go-windows-terminal-sequences v1.0.1 h1:mweAR1A6xJ3oS2pRaGiHgQ4OO8tzTaLawm8vnODuwDk=
github.com/konsorten/go-windows-terminal-sequences v1.0.1/go.mod h1:T0+1ngSBFLxvqU3pZ+m/2kptfBszLMUkC4ZK/EgS/cQ=
github.com/pkg/errors v0.8.1 h1:iURUrRGxPUNPdy5/HRSm+Yj6okJ6UtLINN0Q9M4+h3I=
github.com/pkg/errors v0.8.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
github.com/sirupsen/logrus v1.4.1 h1:GL2rEmy6nsikmW0r8opw9JIRScdMF5hA8cOYLH7In1k=
github.com/sirupsen/logrus v1.4.1/go.mod h1:ni0Sbl8bgC9z8RoU9G6nDWqqs/fq4eDPysMBDgk/93Q=
github.com/stretchr/objx v0.1.1/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
github.com/stretchr/testify v1.2.2 h1:bSDNvY7ZPG5RlJ8otE/7V6gMiyenm9RtJ7IUVIAoJ1w=
github.com/stretchr/testify v1.2.2/go.mod h1:a8OnRcib4nhh0OaRAV+Yts87kKdq0PP7pXfy6kDkUVs=
golang.org/x/sys v0.0.0-20180905080454-ebe1bf3edb33/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
golang.org/x/sys v0.0.0-20190507160741-ecd444e8653b h1:ag/x1USPSsqHud38I9BAC88qdNLDHHtQ4mlgQIZPPNA=
golang.org/x/sys v0.0.0-20190507160741-ecd444e8653b/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3 h1:7TYNF4UdlohbFwpNH04CoPMp1cHUZgO1Ebq5r2hIjfo=
golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=

View file

@ -1,305 +0,0 @@
package winio
import (
"fmt"
"io"
"net"
"os"
"syscall"
"time"
"unsafe"
"github.com/Microsoft/go-winio/pkg/guid"
)
//sys bind(s syscall.Handle, name unsafe.Pointer, namelen int32) (err error) [failretval==socketError] = ws2_32.bind
const (
afHvSock = 34 // AF_HYPERV
socketError = ^uintptr(0)
)
// An HvsockAddr is an address for a AF_HYPERV socket.
type HvsockAddr struct {
VMID guid.GUID
ServiceID guid.GUID
}
type rawHvsockAddr struct {
Family uint16
_ uint16
VMID guid.GUID
ServiceID guid.GUID
}
// Network returns the address's network name, "hvsock".
func (addr *HvsockAddr) Network() string {
return "hvsock"
}
func (addr *HvsockAddr) String() string {
return fmt.Sprintf("%s:%s", &addr.VMID, &addr.ServiceID)
}
// VsockServiceID returns an hvsock service ID corresponding to the specified AF_VSOCK port.
func VsockServiceID(port uint32) guid.GUID {
g, _ := guid.FromString("00000000-facb-11e6-bd58-64006a7986d3")
g.Data1 = port
return g
}
func (addr *HvsockAddr) raw() rawHvsockAddr {
return rawHvsockAddr{
Family: afHvSock,
VMID: addr.VMID,
ServiceID: addr.ServiceID,
}
}
func (addr *HvsockAddr) fromRaw(raw *rawHvsockAddr) {
addr.VMID = raw.VMID
addr.ServiceID = raw.ServiceID
}
// HvsockListener is a socket listener for the AF_HYPERV address family.
type HvsockListener struct {
sock *win32File
addr HvsockAddr
}
// HvsockConn is a connected socket of the AF_HYPERV address family.
type HvsockConn struct {
sock *win32File
local, remote HvsockAddr
}
func newHvSocket() (*win32File, error) {
fd, err := syscall.Socket(afHvSock, syscall.SOCK_STREAM, 1)
if err != nil {
return nil, os.NewSyscallError("socket", err)
}
f, err := makeWin32File(fd)
if err != nil {
syscall.Close(fd)
return nil, err
}
f.socket = true
return f, nil
}
// ListenHvsock listens for connections on the specified hvsock address.
func ListenHvsock(addr *HvsockAddr) (_ *HvsockListener, err error) {
l := &HvsockListener{addr: *addr}
sock, err := newHvSocket()
if err != nil {
return nil, l.opErr("listen", err)
}
sa := addr.raw()
err = bind(sock.handle, unsafe.Pointer(&sa), int32(unsafe.Sizeof(sa)))
if err != nil {
return nil, l.opErr("listen", os.NewSyscallError("socket", err))
}
err = syscall.Listen(sock.handle, 16)
if err != nil {
return nil, l.opErr("listen", os.NewSyscallError("listen", err))
}
return &HvsockListener{sock: sock, addr: *addr}, nil
}
func (l *HvsockListener) opErr(op string, err error) error {
return &net.OpError{Op: op, Net: "hvsock", Addr: &l.addr, Err: err}
}
// Addr returns the listener's network address.
func (l *HvsockListener) Addr() net.Addr {
return &l.addr
}
// Accept waits for the next connection and returns it.
func (l *HvsockListener) Accept() (_ net.Conn, err error) {
sock, err := newHvSocket()
if err != nil {
return nil, l.opErr("accept", err)
}
defer func() {
if sock != nil {
sock.Close()
}
}()
c, err := l.sock.prepareIo()
if err != nil {
return nil, l.opErr("accept", err)
}
defer l.sock.wg.Done()
// AcceptEx, per documentation, requires an extra 16 bytes per address.
const addrlen = uint32(16 + unsafe.Sizeof(rawHvsockAddr{}))
var addrbuf [addrlen * 2]byte
var bytes uint32
err = syscall.AcceptEx(l.sock.handle, sock.handle, &addrbuf[0], 0, addrlen, addrlen, &bytes, &c.o)
_, err = l.sock.asyncIo(c, nil, bytes, err)
if err != nil {
return nil, l.opErr("accept", os.NewSyscallError("acceptex", err))
}
conn := &HvsockConn{
sock: sock,
}
conn.local.fromRaw((*rawHvsockAddr)(unsafe.Pointer(&addrbuf[0])))
conn.remote.fromRaw((*rawHvsockAddr)(unsafe.Pointer(&addrbuf[addrlen])))
sock = nil
return conn, nil
}
// Close closes the listener, causing any pending Accept calls to fail.
func (l *HvsockListener) Close() error {
return l.sock.Close()
}
/* Need to finish ConnectEx handling
func DialHvsock(ctx context.Context, addr *HvsockAddr) (*HvsockConn, error) {
sock, err := newHvSocket()
if err != nil {
return nil, err
}
defer func() {
if sock != nil {
sock.Close()
}
}()
c, err := sock.prepareIo()
if err != nil {
return nil, err
}
defer sock.wg.Done()
var bytes uint32
err = windows.ConnectEx(windows.Handle(sock.handle), sa, nil, 0, &bytes, &c.o)
_, err = sock.asyncIo(ctx, c, nil, bytes, err)
if err != nil {
return nil, err
}
conn := &HvsockConn{
sock: sock,
remote: *addr,
}
sock = nil
return conn, nil
}
*/
func (conn *HvsockConn) opErr(op string, err error) error {
return &net.OpError{Op: op, Net: "hvsock", Source: &conn.local, Addr: &conn.remote, Err: err}
}
func (conn *HvsockConn) Read(b []byte) (int, error) {
c, err := conn.sock.prepareIo()
if err != nil {
return 0, conn.opErr("read", err)
}
defer conn.sock.wg.Done()
buf := syscall.WSABuf{Buf: &b[0], Len: uint32(len(b))}
var flags, bytes uint32
err = syscall.WSARecv(conn.sock.handle, &buf, 1, &bytes, &flags, &c.o, nil)
n, err := conn.sock.asyncIo(c, &conn.sock.readDeadline, bytes, err)
if err != nil {
if _, ok := err.(syscall.Errno); ok {
err = os.NewSyscallError("wsarecv", err)
}
return 0, conn.opErr("read", err)
} else if n == 0 {
err = io.EOF
}
return n, err
}
func (conn *HvsockConn) Write(b []byte) (int, error) {
t := 0
for len(b) != 0 {
n, err := conn.write(b)
if err != nil {
return t + n, err
}
t += n
b = b[n:]
}
return t, nil
}
func (conn *HvsockConn) write(b []byte) (int, error) {
c, err := conn.sock.prepareIo()
if err != nil {
return 0, conn.opErr("write", err)
}
defer conn.sock.wg.Done()
buf := syscall.WSABuf{Buf: &b[0], Len: uint32(len(b))}
var bytes uint32
err = syscall.WSASend(conn.sock.handle, &buf, 1, &bytes, 0, &c.o, nil)
n, err := conn.sock.asyncIo(c, &conn.sock.writeDeadline, bytes, err)
if err != nil {
if _, ok := err.(syscall.Errno); ok {
err = os.NewSyscallError("wsasend", err)
}
return 0, conn.opErr("write", err)
}
return n, err
}
// Close closes the socket connection, failing any pending read or write calls.
func (conn *HvsockConn) Close() error {
return conn.sock.Close()
}
func (conn *HvsockConn) shutdown(how int) error {
err := syscall.Shutdown(conn.sock.handle, syscall.SHUT_RD)
if err != nil {
return os.NewSyscallError("shutdown", err)
}
return nil
}
// CloseRead shuts down the read end of the socket.
func (conn *HvsockConn) CloseRead() error {
err := conn.shutdown(syscall.SHUT_RD)
if err != nil {
return conn.opErr("close", err)
}
return nil
}
// CloseWrite shuts down the write end of the socket, notifying the other endpoint that
// no more data will be written.
func (conn *HvsockConn) CloseWrite() error {
err := conn.shutdown(syscall.SHUT_WR)
if err != nil {
return conn.opErr("close", err)
}
return nil
}
// LocalAddr returns the local address of the connection.
func (conn *HvsockConn) LocalAddr() net.Addr {
return &conn.local
}
// RemoteAddr returns the remote address of the connection.
func (conn *HvsockConn) RemoteAddr() net.Addr {
return &conn.remote
}
// SetDeadline implements the net.Conn SetDeadline method.
func (conn *HvsockConn) SetDeadline(t time.Time) error {
conn.SetReadDeadline(t)
conn.SetWriteDeadline(t)
return nil
}
// SetReadDeadline implements the net.Conn SetReadDeadline method.
func (conn *HvsockConn) SetReadDeadline(t time.Time) error {
return conn.sock.SetReadDeadline(t)
}
// SetWriteDeadline implements the net.Conn SetWriteDeadline method.
func (conn *HvsockConn) SetWriteDeadline(t time.Time) error {
return conn.sock.SetWriteDeadline(t)
}

View file

@ -1,510 +0,0 @@
// +build windows
package winio
import (
"context"
"errors"
"fmt"
"io"
"net"
"os"
"runtime"
"syscall"
"time"
"unsafe"
)
//sys connectNamedPipe(pipe syscall.Handle, o *syscall.Overlapped) (err error) = ConnectNamedPipe
//sys createNamedPipe(name string, flags uint32, pipeMode uint32, maxInstances uint32, outSize uint32, inSize uint32, defaultTimeout uint32, sa *syscall.SecurityAttributes) (handle syscall.Handle, err error) [failretval==syscall.InvalidHandle] = CreateNamedPipeW
//sys createFile(name string, access uint32, mode uint32, sa *syscall.SecurityAttributes, createmode uint32, attrs uint32, templatefile syscall.Handle) (handle syscall.Handle, err error) [failretval==syscall.InvalidHandle] = CreateFileW
//sys getNamedPipeInfo(pipe syscall.Handle, flags *uint32, outSize *uint32, inSize *uint32, maxInstances *uint32) (err error) = GetNamedPipeInfo
//sys getNamedPipeHandleState(pipe syscall.Handle, state *uint32, curInstances *uint32, maxCollectionCount *uint32, collectDataTimeout *uint32, userName *uint16, maxUserNameSize uint32) (err error) = GetNamedPipeHandleStateW
//sys localAlloc(uFlags uint32, length uint32) (ptr uintptr) = LocalAlloc
//sys ntCreateNamedPipeFile(pipe *syscall.Handle, access uint32, oa *objectAttributes, iosb *ioStatusBlock, share uint32, disposition uint32, options uint32, typ uint32, readMode uint32, completionMode uint32, maxInstances uint32, inboundQuota uint32, outputQuota uint32, timeout *int64) (status ntstatus) = ntdll.NtCreateNamedPipeFile
//sys rtlNtStatusToDosError(status ntstatus) (winerr error) = ntdll.RtlNtStatusToDosErrorNoTeb
//sys rtlDosPathNameToNtPathName(name *uint16, ntName *unicodeString, filePart uintptr, reserved uintptr) (status ntstatus) = ntdll.RtlDosPathNameToNtPathName_U
//sys rtlDefaultNpAcl(dacl *uintptr) (status ntstatus) = ntdll.RtlDefaultNpAcl
type ioStatusBlock struct {
Status, Information uintptr
}
type objectAttributes struct {
Length uintptr
RootDirectory uintptr
ObjectName *unicodeString
Attributes uintptr
SecurityDescriptor *securityDescriptor
SecurityQoS uintptr
}
type unicodeString struct {
Length uint16
MaximumLength uint16
Buffer uintptr
}
type securityDescriptor struct {
Revision byte
Sbz1 byte
Control uint16
Owner uintptr
Group uintptr
Sacl uintptr
Dacl uintptr
}
type ntstatus int32
func (status ntstatus) Err() error {
if status >= 0 {
return nil
}
return rtlNtStatusToDosError(status)
}
const (
cERROR_PIPE_BUSY = syscall.Errno(231)
cERROR_NO_DATA = syscall.Errno(232)
cERROR_PIPE_CONNECTED = syscall.Errno(535)
cERROR_SEM_TIMEOUT = syscall.Errno(121)
cSECURITY_SQOS_PRESENT = 0x100000
cSECURITY_ANONYMOUS = 0
cPIPE_TYPE_MESSAGE = 4
cPIPE_READMODE_MESSAGE = 2
cFILE_OPEN = 1
cFILE_CREATE = 2
cFILE_PIPE_MESSAGE_TYPE = 1
cFILE_PIPE_REJECT_REMOTE_CLIENTS = 2
cSE_DACL_PRESENT = 4
)
var (
// ErrPipeListenerClosed is returned for pipe operations on listeners that have been closed.
// This error should match net.errClosing since docker takes a dependency on its text.
ErrPipeListenerClosed = errors.New("use of closed network connection")
errPipeWriteClosed = errors.New("pipe has been closed for write")
)
type win32Pipe struct {
*win32File
path string
}
type win32MessageBytePipe struct {
win32Pipe
writeClosed bool
readEOF bool
}
type pipeAddress string
func (f *win32Pipe) LocalAddr() net.Addr {
return pipeAddress(f.path)
}
func (f *win32Pipe) RemoteAddr() net.Addr {
return pipeAddress(f.path)
}
func (f *win32Pipe) SetDeadline(t time.Time) error {
f.SetReadDeadline(t)
f.SetWriteDeadline(t)
return nil
}
// CloseWrite closes the write side of a message pipe in byte mode.
func (f *win32MessageBytePipe) CloseWrite() error {
if f.writeClosed {
return errPipeWriteClosed
}
err := f.win32File.Flush()
if err != nil {
return err
}
_, err = f.win32File.Write(nil)
if err != nil {
return err
}
f.writeClosed = true
return nil
}
// Write writes bytes to a message pipe in byte mode. Zero-byte writes are ignored, since
// they are used to implement CloseWrite().
func (f *win32MessageBytePipe) Write(b []byte) (int, error) {
if f.writeClosed {
return 0, errPipeWriteClosed
}
if len(b) == 0 {
return 0, nil
}
return f.win32File.Write(b)
}
// Read reads bytes from a message pipe in byte mode. A read of a zero-byte message on a message
// mode pipe will return io.EOF, as will all subsequent reads.
func (f *win32MessageBytePipe) Read(b []byte) (int, error) {
if f.readEOF {
return 0, io.EOF
}
n, err := f.win32File.Read(b)
if err == io.EOF {
// If this was the result of a zero-byte read, then
// it is possible that the read was due to a zero-size
// message. Since we are simulating CloseWrite with a
// zero-byte message, ensure that all future Read() calls
// also return EOF.
f.readEOF = true
} else if err == syscall.ERROR_MORE_DATA {
// ERROR_MORE_DATA indicates that the pipe's read mode is message mode
// and the message still has more bytes. Treat this as a success, since
// this package presents all named pipes as byte streams.
err = nil
}
return n, err
}
func (s pipeAddress) Network() string {
return "pipe"
}
func (s pipeAddress) String() string {
return string(s)
}
// tryDialPipe attempts to dial the pipe at `path` until `ctx` cancellation or timeout.
func tryDialPipe(ctx context.Context, path *string) (syscall.Handle, error) {
for {
select {
case <-ctx.Done():
return syscall.Handle(0), ctx.Err()
default:
h, err := createFile(*path, syscall.GENERIC_READ|syscall.GENERIC_WRITE, 0, nil, syscall.OPEN_EXISTING, syscall.FILE_FLAG_OVERLAPPED|cSECURITY_SQOS_PRESENT|cSECURITY_ANONYMOUS, 0)
if err == nil {
return h, nil
}
if err != cERROR_PIPE_BUSY {
return h, &os.PathError{Err: err, Op: "open", Path: *path}
}
// Wait 10 msec and try again. This is a rather simplistic
// view, as we always try each 10 milliseconds.
time.Sleep(time.Millisecond * 10)
}
}
}
// DialPipe connects to a named pipe by path, timing out if the connection
// takes longer than the specified duration. If timeout is nil, then we use
// a default timeout of 2 seconds. (We do not use WaitNamedPipe.)
func DialPipe(path string, timeout *time.Duration) (net.Conn, error) {
var absTimeout time.Time
if timeout != nil {
absTimeout = time.Now().Add(*timeout)
} else {
absTimeout = time.Now().Add(time.Second * 2)
}
ctx, _ := context.WithDeadline(context.Background(), absTimeout)
conn, err := DialPipeContext(ctx, path)
if err == context.DeadlineExceeded {
return nil, ErrTimeout
}
return conn, err
}
// DialPipeContext attempts to connect to a named pipe by `path` until `ctx`
// cancellation or timeout.
func DialPipeContext(ctx context.Context, path string) (net.Conn, error) {
var err error
var h syscall.Handle
h, err = tryDialPipe(ctx, &path)
if err != nil {
return nil, err
}
var flags uint32
err = getNamedPipeInfo(h, &flags, nil, nil, nil)
if err != nil {
return nil, err
}
f, err := makeWin32File(h)
if err != nil {
syscall.Close(h)
return nil, err
}
// If the pipe is in message mode, return a message byte pipe, which
// supports CloseWrite().
if flags&cPIPE_TYPE_MESSAGE != 0 {
return &win32MessageBytePipe{
win32Pipe: win32Pipe{win32File: f, path: path},
}, nil
}
return &win32Pipe{win32File: f, path: path}, nil
}
type acceptResponse struct {
f *win32File
err error
}
type win32PipeListener struct {
firstHandle syscall.Handle
path string
config PipeConfig
acceptCh chan (chan acceptResponse)
closeCh chan int
doneCh chan int
}
func makeServerPipeHandle(path string, sd []byte, c *PipeConfig, first bool) (syscall.Handle, error) {
path16, err := syscall.UTF16FromString(path)
if err != nil {
return 0, &os.PathError{Op: "open", Path: path, Err: err}
}
var oa objectAttributes
oa.Length = unsafe.Sizeof(oa)
var ntPath unicodeString
if err := rtlDosPathNameToNtPathName(&path16[0], &ntPath, 0, 0).Err(); err != nil {
return 0, &os.PathError{Op: "open", Path: path, Err: err}
}
defer localFree(ntPath.Buffer)
oa.ObjectName = &ntPath
// The security descriptor is only needed for the first pipe.
if first {
if sd != nil {
len := uint32(len(sd))
sdb := localAlloc(0, len)
defer localFree(sdb)
copy((*[0xffff]byte)(unsafe.Pointer(sdb))[:], sd)
oa.SecurityDescriptor = (*securityDescriptor)(unsafe.Pointer(sdb))
} else {
// Construct the default named pipe security descriptor.
var dacl uintptr
if err := rtlDefaultNpAcl(&dacl).Err(); err != nil {
return 0, fmt.Errorf("getting default named pipe ACL: %s", err)
}
defer localFree(dacl)
sdb := &securityDescriptor{
Revision: 1,
Control: cSE_DACL_PRESENT,
Dacl: dacl,
}
oa.SecurityDescriptor = sdb
}
}
typ := uint32(cFILE_PIPE_REJECT_REMOTE_CLIENTS)
if c.MessageMode {
typ |= cFILE_PIPE_MESSAGE_TYPE
}
disposition := uint32(cFILE_OPEN)
access := uint32(syscall.GENERIC_READ | syscall.GENERIC_WRITE | syscall.SYNCHRONIZE)
if first {
disposition = cFILE_CREATE
// By not asking for read or write access, the named pipe file system
// will put this pipe into an initially disconnected state, blocking
// client connections until the next call with first == false.
access = syscall.SYNCHRONIZE
}
timeout := int64(-50 * 10000) // 50ms
var (
h syscall.Handle
iosb ioStatusBlock
)
err = ntCreateNamedPipeFile(&h, access, &oa, &iosb, syscall.FILE_SHARE_READ|syscall.FILE_SHARE_WRITE, disposition, 0, typ, 0, 0, 0xffffffff, uint32(c.InputBufferSize), uint32(c.OutputBufferSize), &timeout).Err()
if err != nil {
return 0, &os.PathError{Op: "open", Path: path, Err: err}
}
runtime.KeepAlive(ntPath)
return h, nil
}
func (l *win32PipeListener) makeServerPipe() (*win32File, error) {
h, err := makeServerPipeHandle(l.path, nil, &l.config, false)
if err != nil {
return nil, err
}
f, err := makeWin32File(h)
if err != nil {
syscall.Close(h)
return nil, err
}
return f, nil
}
func (l *win32PipeListener) makeConnectedServerPipe() (*win32File, error) {
p, err := l.makeServerPipe()
if err != nil {
return nil, err
}
// Wait for the client to connect.
ch := make(chan error)
go func(p *win32File) {
ch <- connectPipe(p)
}(p)
select {
case err = <-ch:
if err != nil {
p.Close()
p = nil
}
case <-l.closeCh:
// Abort the connect request by closing the handle.
p.Close()
p = nil
err = <-ch
if err == nil || err == ErrFileClosed {
err = ErrPipeListenerClosed
}
}
return p, err
}
func (l *win32PipeListener) listenerRoutine() {
closed := false
for !closed {
select {
case <-l.closeCh:
closed = true
case responseCh := <-l.acceptCh:
var (
p *win32File
err error
)
for {
p, err = l.makeConnectedServerPipe()
// If the connection was immediately closed by the client, try
// again.
if err != cERROR_NO_DATA {
break
}
}
responseCh <- acceptResponse{p, err}
closed = err == ErrPipeListenerClosed
}
}
syscall.Close(l.firstHandle)
l.firstHandle = 0
// Notify Close() and Accept() callers that the handle has been closed.
close(l.doneCh)
}
// PipeConfig contain configuration for the pipe listener.
type PipeConfig struct {
// SecurityDescriptor contains a Windows security descriptor in SDDL format.
SecurityDescriptor string
// MessageMode determines whether the pipe is in byte or message mode. In either
// case the pipe is read in byte mode by default. The only practical difference in
// this implementation is that CloseWrite() is only supported for message mode pipes;
// CloseWrite() is implemented as a zero-byte write, but zero-byte writes are only
// transferred to the reader (and returned as io.EOF in this implementation)
// when the pipe is in message mode.
MessageMode bool
// InputBufferSize specifies the size the input buffer, in bytes.
InputBufferSize int32
// OutputBufferSize specifies the size the input buffer, in bytes.
OutputBufferSize int32
}
// ListenPipe creates a listener on a Windows named pipe path, e.g. \\.\pipe\mypipe.
// The pipe must not already exist.
func ListenPipe(path string, c *PipeConfig) (net.Listener, error) {
var (
sd []byte
err error
)
if c == nil {
c = &PipeConfig{}
}
if c.SecurityDescriptor != "" {
sd, err = SddlToSecurityDescriptor(c.SecurityDescriptor)
if err != nil {
return nil, err
}
}
h, err := makeServerPipeHandle(path, sd, c, true)
if err != nil {
return nil, err
}
l := &win32PipeListener{
firstHandle: h,
path: path,
config: *c,
acceptCh: make(chan (chan acceptResponse)),
closeCh: make(chan int),
doneCh: make(chan int),
}
go l.listenerRoutine()
return l, nil
}
func connectPipe(p *win32File) error {
c, err := p.prepareIo()
if err != nil {
return err
}
defer p.wg.Done()
err = connectNamedPipe(p.handle, &c.o)
_, err = p.asyncIo(c, nil, 0, err)
if err != nil && err != cERROR_PIPE_CONNECTED {
return err
}
return nil
}
func (l *win32PipeListener) Accept() (net.Conn, error) {
ch := make(chan acceptResponse)
select {
case l.acceptCh <- ch:
response := <-ch
err := response.err
if err != nil {
return nil, err
}
if l.config.MessageMode {
return &win32MessageBytePipe{
win32Pipe: win32Pipe{win32File: response.f, path: l.path},
}, nil
}
return &win32Pipe{win32File: response.f, path: l.path}, nil
case <-l.doneCh:
return nil, ErrPipeListenerClosed
}
}
func (l *win32PipeListener) Close() error {
select {
case l.closeCh <- 1:
<-l.doneCh
case <-l.doneCh:
}
return nil
}
func (l *win32PipeListener) Addr() net.Addr {
return pipeAddress(l.path)
}

View file

@ -1,235 +0,0 @@
// Package guid provides a GUID type. The backing structure for a GUID is
// identical to that used by the golang.org/x/sys/windows GUID type.
// There are two main binary encodings used for a GUID, the big-endian encoding,
// and the Windows (mixed-endian) encoding. See here for details:
// https://en.wikipedia.org/wiki/Universally_unique_identifier#Encoding
package guid
import (
"crypto/rand"
"crypto/sha1"
"encoding"
"encoding/binary"
"fmt"
"strconv"
"golang.org/x/sys/windows"
)
// Variant specifies which GUID variant (or "type") of the GUID. It determines
// how the entirety of the rest of the GUID is interpreted.
type Variant uint8
// The variants specified by RFC 4122.
const (
// VariantUnknown specifies a GUID variant which does not conform to one of
// the variant encodings specified in RFC 4122.
VariantUnknown Variant = iota
VariantNCS
VariantRFC4122
VariantMicrosoft
VariantFuture
)
// Version specifies how the bits in the GUID were generated. For instance, a
// version 4 GUID is randomly generated, and a version 5 is generated from the
// hash of an input string.
type Version uint8
var _ = (encoding.TextMarshaler)(GUID{})
var _ = (encoding.TextUnmarshaler)(&GUID{})
// GUID represents a GUID/UUID. It has the same structure as
// golang.org/x/sys/windows.GUID so that it can be used with functions expecting
// that type. It is defined as its own type so that stringification and
// marshaling can be supported. The representation matches that used by native
// Windows code.
type GUID windows.GUID
// NewV4 returns a new version 4 (pseudorandom) GUID, as defined by RFC 4122.
func NewV4() (GUID, error) {
var b [16]byte
if _, err := rand.Read(b[:]); err != nil {
return GUID{}, err
}
g := FromArray(b)
g.setVersion(4) // Version 4 means randomly generated.
g.setVariant(VariantRFC4122)
return g, nil
}
// NewV5 returns a new version 5 (generated from a string via SHA-1 hashing)
// GUID, as defined by RFC 4122. The RFC is unclear on the encoding of the name,
// and the sample code treats it as a series of bytes, so we do the same here.
//
// Some implementations, such as those found on Windows, treat the name as a
// big-endian UTF16 stream of bytes. If that is desired, the string can be
// encoded as such before being passed to this function.
func NewV5(namespace GUID, name []byte) (GUID, error) {
b := sha1.New()
namespaceBytes := namespace.ToArray()
b.Write(namespaceBytes[:])
b.Write(name)
a := [16]byte{}
copy(a[:], b.Sum(nil))
g := FromArray(a)
g.setVersion(5) // Version 5 means generated from a string.
g.setVariant(VariantRFC4122)
return g, nil
}
func fromArray(b [16]byte, order binary.ByteOrder) GUID {
var g GUID
g.Data1 = order.Uint32(b[0:4])
g.Data2 = order.Uint16(b[4:6])
g.Data3 = order.Uint16(b[6:8])
copy(g.Data4[:], b[8:16])
return g
}
func (g GUID) toArray(order binary.ByteOrder) [16]byte {
b := [16]byte{}
order.PutUint32(b[0:4], g.Data1)
order.PutUint16(b[4:6], g.Data2)
order.PutUint16(b[6:8], g.Data3)
copy(b[8:16], g.Data4[:])
return b
}
// FromArray constructs a GUID from a big-endian encoding array of 16 bytes.
func FromArray(b [16]byte) GUID {
return fromArray(b, binary.BigEndian)
}
// ToArray returns an array of 16 bytes representing the GUID in big-endian
// encoding.
func (g GUID) ToArray() [16]byte {
return g.toArray(binary.BigEndian)
}
// FromWindowsArray constructs a GUID from a Windows encoding array of bytes.
func FromWindowsArray(b [16]byte) GUID {
return fromArray(b, binary.LittleEndian)
}
// ToWindowsArray returns an array of 16 bytes representing the GUID in Windows
// encoding.
func (g GUID) ToWindowsArray() [16]byte {
return g.toArray(binary.LittleEndian)
}
func (g GUID) String() string {
return fmt.Sprintf(
"%08x-%04x-%04x-%04x-%012x",
g.Data1,
g.Data2,
g.Data3,
g.Data4[:2],
g.Data4[2:])
}
// FromString parses a string containing a GUID and returns the GUID. The only
// format currently supported is the `xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx`
// format.
func FromString(s string) (GUID, error) {
if len(s) != 36 {
return GUID{}, fmt.Errorf("invalid GUID %q", s)
}
if s[8] != '-' || s[13] != '-' || s[18] != '-' || s[23] != '-' {
return GUID{}, fmt.Errorf("invalid GUID %q", s)
}
var g GUID
data1, err := strconv.ParseUint(s[0:8], 16, 32)
if err != nil {
return GUID{}, fmt.Errorf("invalid GUID %q", s)
}
g.Data1 = uint32(data1)
data2, err := strconv.ParseUint(s[9:13], 16, 16)
if err != nil {
return GUID{}, fmt.Errorf("invalid GUID %q", s)
}
g.Data2 = uint16(data2)
data3, err := strconv.ParseUint(s[14:18], 16, 16)
if err != nil {
return GUID{}, fmt.Errorf("invalid GUID %q", s)
}
g.Data3 = uint16(data3)
for i, x := range []int{19, 21, 24, 26, 28, 30, 32, 34} {
v, err := strconv.ParseUint(s[x:x+2], 16, 8)
if err != nil {
return GUID{}, fmt.Errorf("invalid GUID %q", s)
}
g.Data4[i] = uint8(v)
}
return g, nil
}
func (g *GUID) setVariant(v Variant) {
d := g.Data4[0]
switch v {
case VariantNCS:
d = (d & 0x7f)
case VariantRFC4122:
d = (d & 0x3f) | 0x80
case VariantMicrosoft:
d = (d & 0x1f) | 0xc0
case VariantFuture:
d = (d & 0x0f) | 0xe0
case VariantUnknown:
fallthrough
default:
panic(fmt.Sprintf("invalid variant: %d", v))
}
g.Data4[0] = d
}
// Variant returns the GUID variant, as defined in RFC 4122.
func (g GUID) Variant() Variant {
b := g.Data4[0]
if b&0x80 == 0 {
return VariantNCS
} else if b&0xc0 == 0x80 {
return VariantRFC4122
} else if b&0xe0 == 0xc0 {
return VariantMicrosoft
} else if b&0xe0 == 0xe0 {
return VariantFuture
}
return VariantUnknown
}
func (g *GUID) setVersion(v Version) {
g.Data3 = (g.Data3 & 0x0fff) | (uint16(v) << 12)
}
// Version returns the GUID version, as defined in RFC 4122.
func (g GUID) Version() Version {
return Version((g.Data3 & 0xF000) >> 12)
}
// MarshalText returns the textual representation of the GUID.
func (g GUID) MarshalText() ([]byte, error) {
return []byte(g.String()), nil
}
// UnmarshalText takes the textual representation of a GUID, and unmarhals it
// into this GUID.
func (g *GUID) UnmarshalText(text []byte) error {
g2, err := FromString(string(text))
if err != nil {
return err
}
*g = g2
return nil
}

View file

@ -1,202 +0,0 @@
// +build windows
package winio
import (
"bytes"
"encoding/binary"
"fmt"
"runtime"
"sync"
"syscall"
"unicode/utf16"
"golang.org/x/sys/windows"
)
//sys adjustTokenPrivileges(token windows.Token, releaseAll bool, input *byte, outputSize uint32, output *byte, requiredSize *uint32) (success bool, err error) [true] = advapi32.AdjustTokenPrivileges
//sys impersonateSelf(level uint32) (err error) = advapi32.ImpersonateSelf
//sys revertToSelf() (err error) = advapi32.RevertToSelf
//sys openThreadToken(thread syscall.Handle, accessMask uint32, openAsSelf bool, token *windows.Token) (err error) = advapi32.OpenThreadToken
//sys getCurrentThread() (h syscall.Handle) = GetCurrentThread
//sys lookupPrivilegeValue(systemName string, name string, luid *uint64) (err error) = advapi32.LookupPrivilegeValueW
//sys lookupPrivilegeName(systemName string, luid *uint64, buffer *uint16, size *uint32) (err error) = advapi32.LookupPrivilegeNameW
//sys lookupPrivilegeDisplayName(systemName string, name *uint16, buffer *uint16, size *uint32, languageId *uint32) (err error) = advapi32.LookupPrivilegeDisplayNameW
const (
SE_PRIVILEGE_ENABLED = 2
ERROR_NOT_ALL_ASSIGNED syscall.Errno = 1300
SeBackupPrivilege = "SeBackupPrivilege"
SeRestorePrivilege = "SeRestorePrivilege"
)
const (
securityAnonymous = iota
securityIdentification
securityImpersonation
securityDelegation
)
var (
privNames = make(map[string]uint64)
privNameMutex sync.Mutex
)
// PrivilegeError represents an error enabling privileges.
type PrivilegeError struct {
privileges []uint64
}
func (e *PrivilegeError) Error() string {
s := ""
if len(e.privileges) > 1 {
s = "Could not enable privileges "
} else {
s = "Could not enable privilege "
}
for i, p := range e.privileges {
if i != 0 {
s += ", "
}
s += `"`
s += getPrivilegeName(p)
s += `"`
}
return s
}
// RunWithPrivilege enables a single privilege for a function call.
func RunWithPrivilege(name string, fn func() error) error {
return RunWithPrivileges([]string{name}, fn)
}
// RunWithPrivileges enables privileges for a function call.
func RunWithPrivileges(names []string, fn func() error) error {
privileges, err := mapPrivileges(names)
if err != nil {
return err
}
runtime.LockOSThread()
defer runtime.UnlockOSThread()
token, err := newThreadToken()
if err != nil {
return err
}
defer releaseThreadToken(token)
err = adjustPrivileges(token, privileges, SE_PRIVILEGE_ENABLED)
if err != nil {
return err
}
return fn()
}
func mapPrivileges(names []string) ([]uint64, error) {
var privileges []uint64
privNameMutex.Lock()
defer privNameMutex.Unlock()
for _, name := range names {
p, ok := privNames[name]
if !ok {
err := lookupPrivilegeValue("", name, &p)
if err != nil {
return nil, err
}
privNames[name] = p
}
privileges = append(privileges, p)
}
return privileges, nil
}
// EnableProcessPrivileges enables privileges globally for the process.
func EnableProcessPrivileges(names []string) error {
return enableDisableProcessPrivilege(names, SE_PRIVILEGE_ENABLED)
}
// DisableProcessPrivileges disables privileges globally for the process.
func DisableProcessPrivileges(names []string) error {
return enableDisableProcessPrivilege(names, 0)
}
func enableDisableProcessPrivilege(names []string, action uint32) error {
privileges, err := mapPrivileges(names)
if err != nil {
return err
}
p, _ := windows.GetCurrentProcess()
var token windows.Token
err = windows.OpenProcessToken(p, windows.TOKEN_ADJUST_PRIVILEGES|windows.TOKEN_QUERY, &token)
if err != nil {
return err
}
defer token.Close()
return adjustPrivileges(token, privileges, action)
}
func adjustPrivileges(token windows.Token, privileges []uint64, action uint32) error {
var b bytes.Buffer
binary.Write(&b, binary.LittleEndian, uint32(len(privileges)))
for _, p := range privileges {
binary.Write(&b, binary.LittleEndian, p)
binary.Write(&b, binary.LittleEndian, action)
}
prevState := make([]byte, b.Len())
reqSize := uint32(0)
success, err := adjustTokenPrivileges(token, false, &b.Bytes()[0], uint32(len(prevState)), &prevState[0], &reqSize)
if !success {
return err
}
if err == ERROR_NOT_ALL_ASSIGNED {
return &PrivilegeError{privileges}
}
return nil
}
func getPrivilegeName(luid uint64) string {
var nameBuffer [256]uint16
bufSize := uint32(len(nameBuffer))
err := lookupPrivilegeName("", &luid, &nameBuffer[0], &bufSize)
if err != nil {
return fmt.Sprintf("<unknown privilege %d>", luid)
}
var displayNameBuffer [256]uint16
displayBufSize := uint32(len(displayNameBuffer))
var langID uint32
err = lookupPrivilegeDisplayName("", &nameBuffer[0], &displayNameBuffer[0], &displayBufSize, &langID)
if err != nil {
return fmt.Sprintf("<unknown privilege %s>", string(utf16.Decode(nameBuffer[:bufSize])))
}
return string(utf16.Decode(displayNameBuffer[:displayBufSize]))
}
func newThreadToken() (windows.Token, error) {
err := impersonateSelf(securityImpersonation)
if err != nil {
return 0, err
}
var token windows.Token
err = openThreadToken(getCurrentThread(), syscall.TOKEN_ADJUST_PRIVILEGES|syscall.TOKEN_QUERY, false, &token)
if err != nil {
rerr := revertToSelf()
if rerr != nil {
panic(rerr)
}
return 0, err
}
return token, nil
}
func releaseThreadToken(h windows.Token) {
err := revertToSelf()
if err != nil {
panic(err)
}
h.Close()
}

View file

@ -1,128 +0,0 @@
package winio
import (
"bytes"
"encoding/binary"
"fmt"
"strings"
"unicode/utf16"
"unsafe"
)
const (
reparseTagMountPoint = 0xA0000003
reparseTagSymlink = 0xA000000C
)
type reparseDataBuffer struct {
ReparseTag uint32
ReparseDataLength uint16
Reserved uint16
SubstituteNameOffset uint16
SubstituteNameLength uint16
PrintNameOffset uint16
PrintNameLength uint16
}
// ReparsePoint describes a Win32 symlink or mount point.
type ReparsePoint struct {
Target string
IsMountPoint bool
}
// UnsupportedReparsePointError is returned when trying to decode a non-symlink or
// mount point reparse point.
type UnsupportedReparsePointError struct {
Tag uint32
}
func (e *UnsupportedReparsePointError) Error() string {
return fmt.Sprintf("unsupported reparse point %x", e.Tag)
}
// DecodeReparsePoint decodes a Win32 REPARSE_DATA_BUFFER structure containing either a symlink
// or a mount point.
func DecodeReparsePoint(b []byte) (*ReparsePoint, error) {
tag := binary.LittleEndian.Uint32(b[0:4])
return DecodeReparsePointData(tag, b[8:])
}
func DecodeReparsePointData(tag uint32, b []byte) (*ReparsePoint, error) {
isMountPoint := false
switch tag {
case reparseTagMountPoint:
isMountPoint = true
case reparseTagSymlink:
default:
return nil, &UnsupportedReparsePointError{tag}
}
nameOffset := 8 + binary.LittleEndian.Uint16(b[4:6])
if !isMountPoint {
nameOffset += 4
}
nameLength := binary.LittleEndian.Uint16(b[6:8])
name := make([]uint16, nameLength/2)
err := binary.Read(bytes.NewReader(b[nameOffset:nameOffset+nameLength]), binary.LittleEndian, &name)
if err != nil {
return nil, err
}
return &ReparsePoint{string(utf16.Decode(name)), isMountPoint}, nil
}
func isDriveLetter(c byte) bool {
return (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z')
}
// EncodeReparsePoint encodes a Win32 REPARSE_DATA_BUFFER structure describing a symlink or
// mount point.
func EncodeReparsePoint(rp *ReparsePoint) []byte {
// Generate an NT path and determine if this is a relative path.
var ntTarget string
relative := false
if strings.HasPrefix(rp.Target, `\\?\`) {
ntTarget = `\??\` + rp.Target[4:]
} else if strings.HasPrefix(rp.Target, `\\`) {
ntTarget = `\??\UNC\` + rp.Target[2:]
} else if len(rp.Target) >= 2 && isDriveLetter(rp.Target[0]) && rp.Target[1] == ':' {
ntTarget = `\??\` + rp.Target
} else {
ntTarget = rp.Target
relative = true
}
// The paths must be NUL-terminated even though they are counted strings.
target16 := utf16.Encode([]rune(rp.Target + "\x00"))
ntTarget16 := utf16.Encode([]rune(ntTarget + "\x00"))
size := int(unsafe.Sizeof(reparseDataBuffer{})) - 8
size += len(ntTarget16)*2 + len(target16)*2
tag := uint32(reparseTagMountPoint)
if !rp.IsMountPoint {
tag = reparseTagSymlink
size += 4 // Add room for symlink flags
}
data := reparseDataBuffer{
ReparseTag: tag,
ReparseDataLength: uint16(size),
SubstituteNameOffset: 0,
SubstituteNameLength: uint16((len(ntTarget16) - 1) * 2),
PrintNameOffset: uint16(len(ntTarget16) * 2),
PrintNameLength: uint16((len(target16) - 1) * 2),
}
var b bytes.Buffer
binary.Write(&b, binary.LittleEndian, &data)
if !rp.IsMountPoint {
flags := uint32(0)
if relative {
flags |= 1
}
binary.Write(&b, binary.LittleEndian, flags)
}
binary.Write(&b, binary.LittleEndian, ntTarget16)
binary.Write(&b, binary.LittleEndian, target16)
return b.Bytes()
}

View file

@ -1,98 +0,0 @@
// +build windows
package winio
import (
"syscall"
"unsafe"
)
//sys lookupAccountName(systemName *uint16, accountName string, sid *byte, sidSize *uint32, refDomain *uint16, refDomainSize *uint32, sidNameUse *uint32) (err error) = advapi32.LookupAccountNameW
//sys convertSidToStringSid(sid *byte, str **uint16) (err error) = advapi32.ConvertSidToStringSidW
//sys convertStringSecurityDescriptorToSecurityDescriptor(str string, revision uint32, sd *uintptr, size *uint32) (err error) = advapi32.ConvertStringSecurityDescriptorToSecurityDescriptorW
//sys convertSecurityDescriptorToStringSecurityDescriptor(sd *byte, revision uint32, secInfo uint32, sddl **uint16, sddlSize *uint32) (err error) = advapi32.ConvertSecurityDescriptorToStringSecurityDescriptorW
//sys localFree(mem uintptr) = LocalFree
//sys getSecurityDescriptorLength(sd uintptr) (len uint32) = advapi32.GetSecurityDescriptorLength
const (
cERROR_NONE_MAPPED = syscall.Errno(1332)
)
type AccountLookupError struct {
Name string
Err error
}
func (e *AccountLookupError) Error() string {
if e.Name == "" {
return "lookup account: empty account name specified"
}
var s string
switch e.Err {
case cERROR_NONE_MAPPED:
s = "not found"
default:
s = e.Err.Error()
}
return "lookup account " + e.Name + ": " + s
}
type SddlConversionError struct {
Sddl string
Err error
}
func (e *SddlConversionError) Error() string {
return "convert " + e.Sddl + ": " + e.Err.Error()
}
// LookupSidByName looks up the SID of an account by name
func LookupSidByName(name string) (sid string, err error) {
if name == "" {
return "", &AccountLookupError{name, cERROR_NONE_MAPPED}
}
var sidSize, sidNameUse, refDomainSize uint32
err = lookupAccountName(nil, name, nil, &sidSize, nil, &refDomainSize, &sidNameUse)
if err != nil && err != syscall.ERROR_INSUFFICIENT_BUFFER {
return "", &AccountLookupError{name, err}
}
sidBuffer := make([]byte, sidSize)
refDomainBuffer := make([]uint16, refDomainSize)
err = lookupAccountName(nil, name, &sidBuffer[0], &sidSize, &refDomainBuffer[0], &refDomainSize, &sidNameUse)
if err != nil {
return "", &AccountLookupError{name, err}
}
var strBuffer *uint16
err = convertSidToStringSid(&sidBuffer[0], &strBuffer)
if err != nil {
return "", &AccountLookupError{name, err}
}
sid = syscall.UTF16ToString((*[0xffff]uint16)(unsafe.Pointer(strBuffer))[:])
localFree(uintptr(unsafe.Pointer(strBuffer)))
return sid, nil
}
func SddlToSecurityDescriptor(sddl string) ([]byte, error) {
var sdBuffer uintptr
err := convertStringSecurityDescriptorToSecurityDescriptor(sddl, 1, &sdBuffer, nil)
if err != nil {
return nil, &SddlConversionError{sddl, err}
}
defer localFree(sdBuffer)
sd := make([]byte, getSecurityDescriptorLength(sdBuffer))
copy(sd, (*[0xffff]byte)(unsafe.Pointer(sdBuffer))[:len(sd)])
return sd, nil
}
func SecurityDescriptorToSddl(sd []byte) (string, error) {
var sddl *uint16
// The returned string length seems to including an aribtrary number of terminating NULs.
// Don't use it.
err := convertSecurityDescriptorToStringSecurityDescriptor(&sd[0], 1, 0xff, &sddl, nil)
if err != nil {
return "", err
}
defer localFree(uintptr(unsafe.Pointer(sddl)))
return syscall.UTF16ToString((*[0xffff]uint16)(unsafe.Pointer(sddl))[:]), nil
}

View file

@ -1,3 +0,0 @@
package winio
//go:generate go run $GOROOT/src/syscall/mksyscall_windows.go -output zsyscall_windows.go file.go pipe.go sd.go fileinfo.go privilege.go backup.go hvsock.go

View file

@ -1,562 +0,0 @@
// Code generated by 'go generate'; DO NOT EDIT.
package winio
import (
"syscall"
"unsafe"
"golang.org/x/sys/windows"
)
var _ unsafe.Pointer
// Do the interface allocations only once for common
// Errno values.
const (
errnoERROR_IO_PENDING = 997
)
var (
errERROR_IO_PENDING error = syscall.Errno(errnoERROR_IO_PENDING)
)
// errnoErr returns common boxed Errno values, to prevent
// allocations at runtime.
func errnoErr(e syscall.Errno) error {
switch e {
case 0:
return nil
case errnoERROR_IO_PENDING:
return errERROR_IO_PENDING
}
// TODO: add more here, after collecting data on the common
// error values see on Windows. (perhaps when running
// all.bat?)
return e
}
var (
modkernel32 = windows.NewLazySystemDLL("kernel32.dll")
modws2_32 = windows.NewLazySystemDLL("ws2_32.dll")
modntdll = windows.NewLazySystemDLL("ntdll.dll")
modadvapi32 = windows.NewLazySystemDLL("advapi32.dll")
procCancelIoEx = modkernel32.NewProc("CancelIoEx")
procCreateIoCompletionPort = modkernel32.NewProc("CreateIoCompletionPort")
procGetQueuedCompletionStatus = modkernel32.NewProc("GetQueuedCompletionStatus")
procSetFileCompletionNotificationModes = modkernel32.NewProc("SetFileCompletionNotificationModes")
procWSAGetOverlappedResult = modws2_32.NewProc("WSAGetOverlappedResult")
procConnectNamedPipe = modkernel32.NewProc("ConnectNamedPipe")
procCreateNamedPipeW = modkernel32.NewProc("CreateNamedPipeW")
procCreateFileW = modkernel32.NewProc("CreateFileW")
procGetNamedPipeInfo = modkernel32.NewProc("GetNamedPipeInfo")
procGetNamedPipeHandleStateW = modkernel32.NewProc("GetNamedPipeHandleStateW")
procLocalAlloc = modkernel32.NewProc("LocalAlloc")
procNtCreateNamedPipeFile = modntdll.NewProc("NtCreateNamedPipeFile")
procRtlNtStatusToDosErrorNoTeb = modntdll.NewProc("RtlNtStatusToDosErrorNoTeb")
procRtlDosPathNameToNtPathName_U = modntdll.NewProc("RtlDosPathNameToNtPathName_U")
procRtlDefaultNpAcl = modntdll.NewProc("RtlDefaultNpAcl")
procLookupAccountNameW = modadvapi32.NewProc("LookupAccountNameW")
procConvertSidToStringSidW = modadvapi32.NewProc("ConvertSidToStringSidW")
procConvertStringSecurityDescriptorToSecurityDescriptorW = modadvapi32.NewProc("ConvertStringSecurityDescriptorToSecurityDescriptorW")
procConvertSecurityDescriptorToStringSecurityDescriptorW = modadvapi32.NewProc("ConvertSecurityDescriptorToStringSecurityDescriptorW")
procLocalFree = modkernel32.NewProc("LocalFree")
procGetSecurityDescriptorLength = modadvapi32.NewProc("GetSecurityDescriptorLength")
procGetFileInformationByHandleEx = modkernel32.NewProc("GetFileInformationByHandleEx")
procSetFileInformationByHandle = modkernel32.NewProc("SetFileInformationByHandle")
procAdjustTokenPrivileges = modadvapi32.NewProc("AdjustTokenPrivileges")
procImpersonateSelf = modadvapi32.NewProc("ImpersonateSelf")
procRevertToSelf = modadvapi32.NewProc("RevertToSelf")
procOpenThreadToken = modadvapi32.NewProc("OpenThreadToken")
procGetCurrentThread = modkernel32.NewProc("GetCurrentThread")
procLookupPrivilegeValueW = modadvapi32.NewProc("LookupPrivilegeValueW")
procLookupPrivilegeNameW = modadvapi32.NewProc("LookupPrivilegeNameW")
procLookupPrivilegeDisplayNameW = modadvapi32.NewProc("LookupPrivilegeDisplayNameW")
procBackupRead = modkernel32.NewProc("BackupRead")
procBackupWrite = modkernel32.NewProc("BackupWrite")
procbind = modws2_32.NewProc("bind")
)
func cancelIoEx(file syscall.Handle, o *syscall.Overlapped) (err error) {
r1, _, e1 := syscall.Syscall(procCancelIoEx.Addr(), 2, uintptr(file), uintptr(unsafe.Pointer(o)), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func createIoCompletionPort(file syscall.Handle, port syscall.Handle, key uintptr, threadCount uint32) (newport syscall.Handle, err error) {
r0, _, e1 := syscall.Syscall6(procCreateIoCompletionPort.Addr(), 4, uintptr(file), uintptr(port), uintptr(key), uintptr(threadCount), 0, 0)
newport = syscall.Handle(r0)
if newport == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func getQueuedCompletionStatus(port syscall.Handle, bytes *uint32, key *uintptr, o **ioOperation, timeout uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procGetQueuedCompletionStatus.Addr(), 5, uintptr(port), uintptr(unsafe.Pointer(bytes)), uintptr(unsafe.Pointer(key)), uintptr(unsafe.Pointer(o)), uintptr(timeout), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func setFileCompletionNotificationModes(h syscall.Handle, flags uint8) (err error) {
r1, _, e1 := syscall.Syscall(procSetFileCompletionNotificationModes.Addr(), 2, uintptr(h), uintptr(flags), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func wsaGetOverlappedResult(h syscall.Handle, o *syscall.Overlapped, bytes *uint32, wait bool, flags *uint32) (err error) {
var _p0 uint32
if wait {
_p0 = 1
} else {
_p0 = 0
}
r1, _, e1 := syscall.Syscall6(procWSAGetOverlappedResult.Addr(), 5, uintptr(h), uintptr(unsafe.Pointer(o)), uintptr(unsafe.Pointer(bytes)), uintptr(_p0), uintptr(unsafe.Pointer(flags)), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func connectNamedPipe(pipe syscall.Handle, o *syscall.Overlapped) (err error) {
r1, _, e1 := syscall.Syscall(procConnectNamedPipe.Addr(), 2, uintptr(pipe), uintptr(unsafe.Pointer(o)), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func createNamedPipe(name string, flags uint32, pipeMode uint32, maxInstances uint32, outSize uint32, inSize uint32, defaultTimeout uint32, sa *syscall.SecurityAttributes) (handle syscall.Handle, err error) {
var _p0 *uint16
_p0, err = syscall.UTF16PtrFromString(name)
if err != nil {
return
}
return _createNamedPipe(_p0, flags, pipeMode, maxInstances, outSize, inSize, defaultTimeout, sa)
}
func _createNamedPipe(name *uint16, flags uint32, pipeMode uint32, maxInstances uint32, outSize uint32, inSize uint32, defaultTimeout uint32, sa *syscall.SecurityAttributes) (handle syscall.Handle, err error) {
r0, _, e1 := syscall.Syscall9(procCreateNamedPipeW.Addr(), 8, uintptr(unsafe.Pointer(name)), uintptr(flags), uintptr(pipeMode), uintptr(maxInstances), uintptr(outSize), uintptr(inSize), uintptr(defaultTimeout), uintptr(unsafe.Pointer(sa)), 0)
handle = syscall.Handle(r0)
if handle == syscall.InvalidHandle {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func createFile(name string, access uint32, mode uint32, sa *syscall.SecurityAttributes, createmode uint32, attrs uint32, templatefile syscall.Handle) (handle syscall.Handle, err error) {
var _p0 *uint16
_p0, err = syscall.UTF16PtrFromString(name)
if err != nil {
return
}
return _createFile(_p0, access, mode, sa, createmode, attrs, templatefile)
}
func _createFile(name *uint16, access uint32, mode uint32, sa *syscall.SecurityAttributes, createmode uint32, attrs uint32, templatefile syscall.Handle) (handle syscall.Handle, err error) {
r0, _, e1 := syscall.Syscall9(procCreateFileW.Addr(), 7, uintptr(unsafe.Pointer(name)), uintptr(access), uintptr(mode), uintptr(unsafe.Pointer(sa)), uintptr(createmode), uintptr(attrs), uintptr(templatefile), 0, 0)
handle = syscall.Handle(r0)
if handle == syscall.InvalidHandle {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func getNamedPipeInfo(pipe syscall.Handle, flags *uint32, outSize *uint32, inSize *uint32, maxInstances *uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procGetNamedPipeInfo.Addr(), 5, uintptr(pipe), uintptr(unsafe.Pointer(flags)), uintptr(unsafe.Pointer(outSize)), uintptr(unsafe.Pointer(inSize)), uintptr(unsafe.Pointer(maxInstances)), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func getNamedPipeHandleState(pipe syscall.Handle, state *uint32, curInstances *uint32, maxCollectionCount *uint32, collectDataTimeout *uint32, userName *uint16, maxUserNameSize uint32) (err error) {
r1, _, e1 := syscall.Syscall9(procGetNamedPipeHandleStateW.Addr(), 7, uintptr(pipe), uintptr(unsafe.Pointer(state)), uintptr(unsafe.Pointer(curInstances)), uintptr(unsafe.Pointer(maxCollectionCount)), uintptr(unsafe.Pointer(collectDataTimeout)), uintptr(unsafe.Pointer(userName)), uintptr(maxUserNameSize), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func localAlloc(uFlags uint32, length uint32) (ptr uintptr) {
r0, _, _ := syscall.Syscall(procLocalAlloc.Addr(), 2, uintptr(uFlags), uintptr(length), 0)
ptr = uintptr(r0)
return
}
func ntCreateNamedPipeFile(pipe *syscall.Handle, access uint32, oa *objectAttributes, iosb *ioStatusBlock, share uint32, disposition uint32, options uint32, typ uint32, readMode uint32, completionMode uint32, maxInstances uint32, inboundQuota uint32, outputQuota uint32, timeout *int64) (status ntstatus) {
r0, _, _ := syscall.Syscall15(procNtCreateNamedPipeFile.Addr(), 14, uintptr(unsafe.Pointer(pipe)), uintptr(access), uintptr(unsafe.Pointer(oa)), uintptr(unsafe.Pointer(iosb)), uintptr(share), uintptr(disposition), uintptr(options), uintptr(typ), uintptr(readMode), uintptr(completionMode), uintptr(maxInstances), uintptr(inboundQuota), uintptr(outputQuota), uintptr(unsafe.Pointer(timeout)), 0)
status = ntstatus(r0)
return
}
func rtlNtStatusToDosError(status ntstatus) (winerr error) {
r0, _, _ := syscall.Syscall(procRtlNtStatusToDosErrorNoTeb.Addr(), 1, uintptr(status), 0, 0)
if r0 != 0 {
winerr = syscall.Errno(r0)
}
return
}
func rtlDosPathNameToNtPathName(name *uint16, ntName *unicodeString, filePart uintptr, reserved uintptr) (status ntstatus) {
r0, _, _ := syscall.Syscall6(procRtlDosPathNameToNtPathName_U.Addr(), 4, uintptr(unsafe.Pointer(name)), uintptr(unsafe.Pointer(ntName)), uintptr(filePart), uintptr(reserved), 0, 0)
status = ntstatus(r0)
return
}
func rtlDefaultNpAcl(dacl *uintptr) (status ntstatus) {
r0, _, _ := syscall.Syscall(procRtlDefaultNpAcl.Addr(), 1, uintptr(unsafe.Pointer(dacl)), 0, 0)
status = ntstatus(r0)
return
}
func lookupAccountName(systemName *uint16, accountName string, sid *byte, sidSize *uint32, refDomain *uint16, refDomainSize *uint32, sidNameUse *uint32) (err error) {
var _p0 *uint16
_p0, err = syscall.UTF16PtrFromString(accountName)
if err != nil {
return
}
return _lookupAccountName(systemName, _p0, sid, sidSize, refDomain, refDomainSize, sidNameUse)
}
func _lookupAccountName(systemName *uint16, accountName *uint16, sid *byte, sidSize *uint32, refDomain *uint16, refDomainSize *uint32, sidNameUse *uint32) (err error) {
r1, _, e1 := syscall.Syscall9(procLookupAccountNameW.Addr(), 7, uintptr(unsafe.Pointer(systemName)), uintptr(unsafe.Pointer(accountName)), uintptr(unsafe.Pointer(sid)), uintptr(unsafe.Pointer(sidSize)), uintptr(unsafe.Pointer(refDomain)), uintptr(unsafe.Pointer(refDomainSize)), uintptr(unsafe.Pointer(sidNameUse)), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func convertSidToStringSid(sid *byte, str **uint16) (err error) {
r1, _, e1 := syscall.Syscall(procConvertSidToStringSidW.Addr(), 2, uintptr(unsafe.Pointer(sid)), uintptr(unsafe.Pointer(str)), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func convertStringSecurityDescriptorToSecurityDescriptor(str string, revision uint32, sd *uintptr, size *uint32) (err error) {
var _p0 *uint16
_p0, err = syscall.UTF16PtrFromString(str)
if err != nil {
return
}
return _convertStringSecurityDescriptorToSecurityDescriptor(_p0, revision, sd, size)
}
func _convertStringSecurityDescriptorToSecurityDescriptor(str *uint16, revision uint32, sd *uintptr, size *uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procConvertStringSecurityDescriptorToSecurityDescriptorW.Addr(), 4, uintptr(unsafe.Pointer(str)), uintptr(revision), uintptr(unsafe.Pointer(sd)), uintptr(unsafe.Pointer(size)), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func convertSecurityDescriptorToStringSecurityDescriptor(sd *byte, revision uint32, secInfo uint32, sddl **uint16, sddlSize *uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procConvertSecurityDescriptorToStringSecurityDescriptorW.Addr(), 5, uintptr(unsafe.Pointer(sd)), uintptr(revision), uintptr(secInfo), uintptr(unsafe.Pointer(sddl)), uintptr(unsafe.Pointer(sddlSize)), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func localFree(mem uintptr) {
syscall.Syscall(procLocalFree.Addr(), 1, uintptr(mem), 0, 0)
return
}
func getSecurityDescriptorLength(sd uintptr) (len uint32) {
r0, _, _ := syscall.Syscall(procGetSecurityDescriptorLength.Addr(), 1, uintptr(sd), 0, 0)
len = uint32(r0)
return
}
func getFileInformationByHandleEx(h syscall.Handle, class uint32, buffer *byte, size uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procGetFileInformationByHandleEx.Addr(), 4, uintptr(h), uintptr(class), uintptr(unsafe.Pointer(buffer)), uintptr(size), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func setFileInformationByHandle(h syscall.Handle, class uint32, buffer *byte, size uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procSetFileInformationByHandle.Addr(), 4, uintptr(h), uintptr(class), uintptr(unsafe.Pointer(buffer)), uintptr(size), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func adjustTokenPrivileges(token windows.Token, releaseAll bool, input *byte, outputSize uint32, output *byte, requiredSize *uint32) (success bool, err error) {
var _p0 uint32
if releaseAll {
_p0 = 1
} else {
_p0 = 0
}
r0, _, e1 := syscall.Syscall6(procAdjustTokenPrivileges.Addr(), 6, uintptr(token), uintptr(_p0), uintptr(unsafe.Pointer(input)), uintptr(outputSize), uintptr(unsafe.Pointer(output)), uintptr(unsafe.Pointer(requiredSize)))
success = r0 != 0
if true {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func impersonateSelf(level uint32) (err error) {
r1, _, e1 := syscall.Syscall(procImpersonateSelf.Addr(), 1, uintptr(level), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func revertToSelf() (err error) {
r1, _, e1 := syscall.Syscall(procRevertToSelf.Addr(), 0, 0, 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func openThreadToken(thread syscall.Handle, accessMask uint32, openAsSelf bool, token *windows.Token) (err error) {
var _p0 uint32
if openAsSelf {
_p0 = 1
} else {
_p0 = 0
}
r1, _, e1 := syscall.Syscall6(procOpenThreadToken.Addr(), 4, uintptr(thread), uintptr(accessMask), uintptr(_p0), uintptr(unsafe.Pointer(token)), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func getCurrentThread() (h syscall.Handle) {
r0, _, _ := syscall.Syscall(procGetCurrentThread.Addr(), 0, 0, 0, 0)
h = syscall.Handle(r0)
return
}
func lookupPrivilegeValue(systemName string, name string, luid *uint64) (err error) {
var _p0 *uint16
_p0, err = syscall.UTF16PtrFromString(systemName)
if err != nil {
return
}
var _p1 *uint16
_p1, err = syscall.UTF16PtrFromString(name)
if err != nil {
return
}
return _lookupPrivilegeValue(_p0, _p1, luid)
}
func _lookupPrivilegeValue(systemName *uint16, name *uint16, luid *uint64) (err error) {
r1, _, e1 := syscall.Syscall(procLookupPrivilegeValueW.Addr(), 3, uintptr(unsafe.Pointer(systemName)), uintptr(unsafe.Pointer(name)), uintptr(unsafe.Pointer(luid)))
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func lookupPrivilegeName(systemName string, luid *uint64, buffer *uint16, size *uint32) (err error) {
var _p0 *uint16
_p0, err = syscall.UTF16PtrFromString(systemName)
if err != nil {
return
}
return _lookupPrivilegeName(_p0, luid, buffer, size)
}
func _lookupPrivilegeName(systemName *uint16, luid *uint64, buffer *uint16, size *uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procLookupPrivilegeNameW.Addr(), 4, uintptr(unsafe.Pointer(systemName)), uintptr(unsafe.Pointer(luid)), uintptr(unsafe.Pointer(buffer)), uintptr(unsafe.Pointer(size)), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func lookupPrivilegeDisplayName(systemName string, name *uint16, buffer *uint16, size *uint32, languageId *uint32) (err error) {
var _p0 *uint16
_p0, err = syscall.UTF16PtrFromString(systemName)
if err != nil {
return
}
return _lookupPrivilegeDisplayName(_p0, name, buffer, size, languageId)
}
func _lookupPrivilegeDisplayName(systemName *uint16, name *uint16, buffer *uint16, size *uint32, languageId *uint32) (err error) {
r1, _, e1 := syscall.Syscall6(procLookupPrivilegeDisplayNameW.Addr(), 5, uintptr(unsafe.Pointer(systemName)), uintptr(unsafe.Pointer(name)), uintptr(unsafe.Pointer(buffer)), uintptr(unsafe.Pointer(size)), uintptr(unsafe.Pointer(languageId)), 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func backupRead(h syscall.Handle, b []byte, bytesRead *uint32, abort bool, processSecurity bool, context *uintptr) (err error) {
var _p0 *byte
if len(b) > 0 {
_p0 = &b[0]
}
var _p1 uint32
if abort {
_p1 = 1
} else {
_p1 = 0
}
var _p2 uint32
if processSecurity {
_p2 = 1
} else {
_p2 = 0
}
r1, _, e1 := syscall.Syscall9(procBackupRead.Addr(), 7, uintptr(h), uintptr(unsafe.Pointer(_p0)), uintptr(len(b)), uintptr(unsafe.Pointer(bytesRead)), uintptr(_p1), uintptr(_p2), uintptr(unsafe.Pointer(context)), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func backupWrite(h syscall.Handle, b []byte, bytesWritten *uint32, abort bool, processSecurity bool, context *uintptr) (err error) {
var _p0 *byte
if len(b) > 0 {
_p0 = &b[0]
}
var _p1 uint32
if abort {
_p1 = 1
} else {
_p1 = 0
}
var _p2 uint32
if processSecurity {
_p2 = 1
} else {
_p2 = 0
}
r1, _, e1 := syscall.Syscall9(procBackupWrite.Addr(), 7, uintptr(h), uintptr(unsafe.Pointer(_p0)), uintptr(len(b)), uintptr(unsafe.Pointer(bytesWritten)), uintptr(_p1), uintptr(_p2), uintptr(unsafe.Pointer(context)), 0, 0)
if r1 == 0 {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}
func bind(s syscall.Handle, name unsafe.Pointer, namelen int32) (err error) {
r1, _, e1 := syscall.Syscall(procbind.Addr(), 3, uintptr(s), uintptr(name), uintptr(namelen))
if r1 == socketError {
if e1 != 0 {
err = errnoErr(e1)
} else {
err = syscall.EINVAL
}
}
return
}

View file

@ -1,21 +0,0 @@
The MIT License (MIT)
Copyright (c) 2015 Microsoft
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

View file

@ -1,57 +0,0 @@
package osversion
import (
"fmt"
"golang.org/x/sys/windows"
)
// OSVersion is a wrapper for Windows version information
// https://msdn.microsoft.com/en-us/library/windows/desktop/ms724439(v=vs.85).aspx
type OSVersion struct {
Version uint32
MajorVersion uint8
MinorVersion uint8
Build uint16
}
// https://msdn.microsoft.com/en-us/library/windows/desktop/ms724833(v=vs.85).aspx
type osVersionInfoEx struct {
OSVersionInfoSize uint32
MajorVersion uint32
MinorVersion uint32
BuildNumber uint32
PlatformID uint32
CSDVersion [128]uint16
ServicePackMajor uint16
ServicePackMinor uint16
SuiteMask uint16
ProductType byte
Reserve byte
}
// Get gets the operating system version on Windows.
// The calling application must be manifested to get the correct version information.
func Get() OSVersion {
var err error
osv := OSVersion{}
osv.Version, err = windows.GetVersion()
if err != nil {
// GetVersion never fails.
panic(err)
}
osv.MajorVersion = uint8(osv.Version & 0xFF)
osv.MinorVersion = uint8(osv.Version >> 8 & 0xFF)
osv.Build = uint16(osv.Version >> 16)
return osv
}
// Build gets the build-number on Windows
// The calling application must be manifested to get the correct version information.
func Build() uint16 {
return Get().Build
}
func (osv OSVersion) ToString() string {
return fmt.Sprintf("%d.%d.%d", osv.MajorVersion, osv.MinorVersion, osv.Build)
}

View file

@ -1,27 +0,0 @@
package osversion
const (
// RS1 (version 1607, codename "Redstone 1") corresponds to Windows Server
// 2016 (ltsc2016) and Windows 10 (Anniversary Update).
RS1 = 14393
// RS2 (version 1703, codename "Redstone 2") was a client-only update, and
// corresponds to Windows 10 (Creators Update).
RS2 = 15063
// RS3 (version 1709, codename "Redstone 3") corresponds to Windows Server
// 1709 (Semi-Annual Channel (SAC)), and Windows 10 (Fall Creators Update).
RS3 = 16299
// RS4 (version 1803, codename "Redstone 4") corresponds to Windows Server
// 1803 (Semi-Annual Channel (SAC)), and Windows 10 (April 2018 Update).
RS4 = 17134
// RS5 (version 1809, codename "Redstone 5") corresponds to Windows Server
// 2019 (ltsc2019), and Windows 10 (October 2018 Update).
RS5 = 17763
// V19H1 (version 1903) corresponds to Windows Server 1903 (semi-annual
// channel).
V19H1 = 18362
)

View file

@ -1,14 +0,0 @@
language: go
sudo: false
go:
- 1.11.x
- tip
matrix:
allow_failures:
- go: tip
script:
- go vet ./...
- go test -v ./...

View file

@ -1,10 +0,0 @@
Brett Vickers (beevik)
Felix Geisendörfer (felixge)
Kamil Kisiel (kisielk)
Graham King (grahamking)
Matt Smith (ma314smith)
Michal Jemala (michaljemala)
Nicolas Piganeau (npiganeau)
Chris Brown (ccbrown)
Earncef Sequeira (earncef)
Gabriel de Labachelerie (wuzuf)

View file

@ -1,24 +0,0 @@
Copyright 2015-2019 Brett Vickers. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
1. Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY COPYRIGHT HOLDER ``AS IS'' AND ANY
EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL COPYRIGHT HOLDER OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

View file

@ -1,205 +0,0 @@
[![Build Status](https://travis-ci.org/beevik/etree.svg?branch=master)](https://travis-ci.org/beevik/etree)
[![GoDoc](https://godoc.org/github.com/beevik/etree?status.svg)](https://godoc.org/github.com/beevik/etree)
etree
=====
The etree package is a lightweight, pure go package that expresses XML in
the form of an element tree. Its design was inspired by the Python
[ElementTree](http://docs.python.org/2/library/xml.etree.elementtree.html)
module.
Some of the package's capabilities and features:
* Represents XML documents as trees of elements for easy traversal.
* Imports, serializes, modifies or creates XML documents from scratch.
* Writes and reads XML to/from files, byte slices, strings and io interfaces.
* Performs simple or complex searches with lightweight XPath-like query APIs.
* Auto-indents XML using spaces or tabs for better readability.
* Implemented in pure go; depends only on standard go libraries.
* Built on top of the go [encoding/xml](http://golang.org/pkg/encoding/xml)
package.
### Creating an XML document
The following example creates an XML document from scratch using the etree
package and outputs its indented contents to stdout.
```go
doc := etree.NewDocument()
doc.CreateProcInst("xml", `version="1.0" encoding="UTF-8"`)
doc.CreateProcInst("xml-stylesheet", `type="text/xsl" href="style.xsl"`)
people := doc.CreateElement("People")
people.CreateComment("These are all known people")
jon := people.CreateElement("Person")
jon.CreateAttr("name", "Jon")
sally := people.CreateElement("Person")
sally.CreateAttr("name", "Sally")
doc.Indent(2)
doc.WriteTo(os.Stdout)
```
Output:
```xml
<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="style.xsl"?>
<People>
<!--These are all known people-->
<Person name="Jon"/>
<Person name="Sally"/>
</People>
```
### Reading an XML file
Suppose you have a file on disk called `bookstore.xml` containing the
following data:
```xml
<bookstore xmlns:p="urn:schemas-books-com:prices">
<book category="COOKING">
<title lang="en">Everyday Italian</title>
<author>Giada De Laurentiis</author>
<year>2005</year>
<p:price>30.00</p:price>
</book>
<book category="CHILDREN">
<title lang="en">Harry Potter</title>
<author>J K. Rowling</author>
<year>2005</year>
<p:price>29.99</p:price>
</book>
<book category="WEB">
<title lang="en">XQuery Kick Start</title>
<author>James McGovern</author>
<author>Per Bothner</author>
<author>Kurt Cagle</author>
<author>James Linn</author>
<author>Vaidyanathan Nagarajan</author>
<year>2003</year>
<p:price>49.99</p:price>
</book>
<book category="WEB">
<title lang="en">Learning XML</title>
<author>Erik T. Ray</author>
<year>2003</year>
<p:price>39.95</p:price>
</book>
</bookstore>
```
This code reads the file's contents into an etree document.
```go
doc := etree.NewDocument()
if err := doc.ReadFromFile("bookstore.xml"); err != nil {
panic(err)
}
```
You can also read XML from a string, a byte slice, or an `io.Reader`.
### Processing elements and attributes
This example illustrates several ways to access elements and attributes using
etree selection queries.
```go
root := doc.SelectElement("bookstore")
fmt.Println("ROOT element:", root.Tag)
for _, book := range root.SelectElements("book") {
fmt.Println("CHILD element:", book.Tag)
if title := book.SelectElement("title"); title != nil {
lang := title.SelectAttrValue("lang", "unknown")
fmt.Printf(" TITLE: %s (%s)\n", title.Text(), lang)
}
for _, attr := range book.Attr {
fmt.Printf(" ATTR: %s=%s\n", attr.Key, attr.Value)
}
}
```
Output:
```
ROOT element: bookstore
CHILD element: book
TITLE: Everyday Italian (en)
ATTR: category=COOKING
CHILD element: book
TITLE: Harry Potter (en)
ATTR: category=CHILDREN
CHILD element: book
TITLE: XQuery Kick Start (en)
ATTR: category=WEB
CHILD element: book
TITLE: Learning XML (en)
ATTR: category=WEB
```
### Path queries
This example uses etree's path functions to select all book titles that fall
into the category of 'WEB'. The double-slash prefix in the path causes the
search for book elements to occur recursively; book elements may appear at any
level of the XML hierarchy.
```go
for _, t := range doc.FindElements("//book[@category='WEB']/title") {
fmt.Println("Title:", t.Text())
}
```
Output:
```
Title: XQuery Kick Start
Title: Learning XML
```
This example finds the first book element under the root bookstore element and
outputs the tag and text of each of its child elements.
```go
for _, e := range doc.FindElements("./bookstore/book[1]/*") {
fmt.Printf("%s: %s\n", e.Tag, e.Text())
}
```
Output:
```
title: Everyday Italian
author: Giada De Laurentiis
year: 2005
price: 30.00
```
This example finds all books with a price of 49.99 and outputs their titles.
```go
path := etree.MustCompilePath("./bookstore/book[p:price='49.99']/title")
for _, e := range doc.FindElementsPath(path) {
fmt.Println(e.Text())
}
```
Output:
```
XQuery Kick Start
```
Note that this example uses the FindElementsPath function, which takes as an
argument a pre-compiled path object. Use precompiled paths when you plan to
search with the same path more than once.
### Other features
These are just a few examples of the things the etree package can do. See the
[documentation](http://godoc.org/github.com/beevik/etree) for a complete
description of its capabilities.
### Contributing
This project accepts contributions. Just fork the repo and submit a pull
request!

View file

@ -1,109 +0,0 @@
Release v1.1.0
==============
**New Features**
* New attribute helpers.
* Added the `Element.SortAttrs` method, which lexicographically sorts an
element's attributes by key.
* New `ReadSettings` properties.
* Added `Entity` for the support of custom entity maps.
* New `WriteSettings` properties.
* Added `UseCRLF` to allow the output of CR-LF newlines instead of the
default LF newlines. This is useful on Windows systems.
* Additional support for text and CDATA sections.
* The `Element.Text` method now returns the concatenation of all consecutive
character data tokens immediately following an element's opening tag.
* Added `Element.SetCData` to replace the character data immediately
following an element's opening tag with a CDATA section.
* Added `Element.CreateCData` to create and add a CDATA section child
`CharData` token to an element.
* Added `Element.CreateText` to create and add a child text `CharData` token
to an element.
* Added `NewCData` to create a parentless CDATA section `CharData` token.
* Added `NewText` to create a parentless text `CharData`
token.
* Added `CharData.IsCData` to detect if the token contains a CDATA section.
* Added `CharData.IsWhitespace` to detect if the token contains whitespace
inserted by one of the document Indent functions.
* Modified `Element.SetText` so that it replaces a run of consecutive
character data tokens following the element's opening tag (instead of just
the first one).
* New "tail text" support.
* Added the `Element.Tail` method, which returns the text immediately
following an element's closing tag.
* Added the `Element.SetTail` method, which modifies the text immediately
following an element's closing tag.
* New element child insertion and removal methods.
* Added the `Element.InsertChildAt` method, which inserts a new child token
before the specified child token index.
* Added the `Element.RemoveChildAt` method, which removes the child token at
the specified child token index.
* New element and attribute queries.
* Added the `Element.Index` method, which returns the element's index within
its parent element's child token list.
* Added the `Element.NamespaceURI` method to return the namespace URI
associated with an element.
* Added the `Attr.NamespaceURI` method to return the namespace URI
associated with an element.
* Added the `Attr.Element` method to return the element that an attribute
belongs to.
* New Path filter functions.
* Added `[local-name()='val']` to keep elements whose unprefixed tag matches
the desired value.
* Added `[name()='val']` to keep elements whose full tag matches the desired
value.
* Added `[namespace-prefix()='val']` to keep elements whose namespace prefix
matches the desired value.
* Added `[namespace-uri()='val']` to keep elements whose namespace URI
matches the desired value.
**Bug Fixes**
* A default XML `CharSetReader` is now used to prevent failed parsing of XML
documents using certain encodings.
([Issue](https://github.com/beevik/etree/issues/53)).
* All characters are now properly escaped according to XML parsing rules.
([Issue](https://github.com/beevik/etree/issues/55)).
* The `Document.Indent` and `Document.IndentTabs` functions no longer insert
empty string `CharData` tokens.
**Deprecated**
* `Element`
* The `InsertChild` method is deprecated. Use `InsertChildAt` instead.
* The `CreateCharData` method is deprecated. Use `CreateText` instead.
* `CharData`
* The `NewCharData` method is deprecated. Use `NewText` instead.
Release v1.0.1
==============
**Changes**
* Added support for absolute etree Path queries. An absolute path begins with
`/` or `//` and begins its search from the element's document root.
* Added [`GetPath`](https://godoc.org/github.com/beevik/etree#Element.GetPath)
and [`GetRelativePath`](https://godoc.org/github.com/beevik/etree#Element.GetRelativePath)
functions to the [`Element`](https://godoc.org/github.com/beevik/etree#Element)
type.
**Breaking changes**
* A path starting with `//` is now interpreted as an absolute path.
Previously, it was interpreted as a relative path starting from the element
whose
[`FindElement`](https://godoc.org/github.com/beevik/etree#Element.FindElement)
method was called. To remain compatible with this release, all paths
prefixed with `//` should be prefixed with `.//` when called from any
element other than the document's root.
* [**edit 2/1/2019**]: Minor releases should not contain breaking changes.
Even though this breaking change was very minor, it was a mistake to include
it in this minor release. In the future, all breaking changes will be
limited to major releases (e.g., version 2.0.0).
Release v1.0.0
==============
Initial release.

1453
vendor/github.com/beevik/etree/etree.go generated vendored

File diff suppressed because it is too large Load diff

View file

@ -1,276 +0,0 @@
// Copyright 2015-2019 Brett Vickers.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package etree
import (
"bufio"
"io"
"strings"
"unicode/utf8"
)
// A simple stack
type stack struct {
data []interface{}
}
func (s *stack) empty() bool {
return len(s.data) == 0
}
func (s *stack) push(value interface{}) {
s.data = append(s.data, value)
}
func (s *stack) pop() interface{} {
value := s.data[len(s.data)-1]
s.data[len(s.data)-1] = nil
s.data = s.data[:len(s.data)-1]
return value
}
func (s *stack) peek() interface{} {
return s.data[len(s.data)-1]
}
// A fifo is a simple first-in-first-out queue.
type fifo struct {
data []interface{}
head, tail int
}
func (f *fifo) add(value interface{}) {
if f.len()+1 >= len(f.data) {
f.grow()
}
f.data[f.tail] = value
if f.tail++; f.tail == len(f.data) {
f.tail = 0
}
}
func (f *fifo) remove() interface{} {
value := f.data[f.head]
f.data[f.head] = nil
if f.head++; f.head == len(f.data) {
f.head = 0
}
return value
}
func (f *fifo) len() int {
if f.tail >= f.head {
return f.tail - f.head
}
return len(f.data) - f.head + f.tail
}
func (f *fifo) grow() {
c := len(f.data) * 2
if c == 0 {
c = 4
}
buf, count := make([]interface{}, c), f.len()
if f.tail >= f.head {
copy(buf[0:count], f.data[f.head:f.tail])
} else {
hindex := len(f.data) - f.head
copy(buf[0:hindex], f.data[f.head:])
copy(buf[hindex:count], f.data[:f.tail])
}
f.data, f.head, f.tail = buf, 0, count
}
// countReader implements a proxy reader that counts the number of
// bytes read from its encapsulated reader.
type countReader struct {
r io.Reader
bytes int64
}
func newCountReader(r io.Reader) *countReader {
return &countReader{r: r}
}
func (cr *countReader) Read(p []byte) (n int, err error) {
b, err := cr.r.Read(p)
cr.bytes += int64(b)
return b, err
}
// countWriter implements a proxy writer that counts the number of
// bytes written by its encapsulated writer.
type countWriter struct {
w io.Writer
bytes int64
}
func newCountWriter(w io.Writer) *countWriter {
return &countWriter{w: w}
}
func (cw *countWriter) Write(p []byte) (n int, err error) {
b, err := cw.w.Write(p)
cw.bytes += int64(b)
return b, err
}
// isWhitespace returns true if the byte slice contains only
// whitespace characters.
func isWhitespace(s string) bool {
for i := 0; i < len(s); i++ {
if c := s[i]; c != ' ' && c != '\t' && c != '\n' && c != '\r' {
return false
}
}
return true
}
// spaceMatch returns true if namespace a is the empty string
// or if namespace a equals namespace b.
func spaceMatch(a, b string) bool {
switch {
case a == "":
return true
default:
return a == b
}
}
// spaceDecompose breaks a namespace:tag identifier at the ':'
// and returns the two parts.
func spaceDecompose(str string) (space, key string) {
colon := strings.IndexByte(str, ':')
if colon == -1 {
return "", str
}
return str[:colon], str[colon+1:]
}
// Strings used by indentCRLF and indentLF
const (
indentSpaces = "\r\n "
indentTabs = "\r\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t"
)
// indentCRLF returns a CRLF newline followed by n copies of the first
// non-CRLF character in the source string.
func indentCRLF(n int, source string) string {
switch {
case n < 0:
return source[:2]
case n < len(source)-1:
return source[:n+2]
default:
return source + strings.Repeat(source[2:3], n-len(source)+2)
}
}
// indentLF returns a LF newline followed by n copies of the first non-LF
// character in the source string.
func indentLF(n int, source string) string {
switch {
case n < 0:
return source[1:2]
case n < len(source)-1:
return source[1 : n+2]
default:
return source[1:] + strings.Repeat(source[2:3], n-len(source)+2)
}
}
// nextIndex returns the index of the next occurrence of sep in s,
// starting from offset. It returns -1 if the sep string is not found.
func nextIndex(s, sep string, offset int) int {
switch i := strings.Index(s[offset:], sep); i {
case -1:
return -1
default:
return offset + i
}
}
// isInteger returns true if the string s contains an integer.
func isInteger(s string) bool {
for i := 0; i < len(s); i++ {
if (s[i] < '0' || s[i] > '9') && !(i == 0 && s[i] == '-') {
return false
}
}
return true
}
type escapeMode byte
const (
escapeNormal escapeMode = iota
escapeCanonicalText
escapeCanonicalAttr
)
// escapeString writes an escaped version of a string to the writer.
func escapeString(w *bufio.Writer, s string, m escapeMode) {
var esc []byte
last := 0
for i := 0; i < len(s); {
r, width := utf8.DecodeRuneInString(s[i:])
i += width
switch r {
case '&':
esc = []byte("&amp;")
case '<':
esc = []byte("&lt;")
case '>':
if m == escapeCanonicalAttr {
continue
}
esc = []byte("&gt;")
case '\'':
if m != escapeNormal {
continue
}
esc = []byte("&apos;")
case '"':
if m == escapeCanonicalText {
continue
}
esc = []byte("&quot;")
case '\t':
if m != escapeCanonicalAttr {
continue
}
esc = []byte("&#x9;")
case '\n':
if m != escapeCanonicalAttr {
continue
}
esc = []byte("&#xA;")
case '\r':
if m == escapeNormal {
continue
}
esc = []byte("&#xD;")
default:
if !isInCharacterRange(r) || (r == 0xFFFD && width == 1) {
esc = []byte("\uFFFD")
break
}
continue
}
w.WriteString(s[last : i-width])
w.Write(esc)
last = i
}
w.WriteString(s[last:])
}
func isInCharacterRange(r rune) bool {
return r == 0x09 ||
r == 0x0A ||
r == 0x0D ||
r >= 0x20 && r <= 0xD7FF ||
r >= 0xE000 && r <= 0xFFFD ||
r >= 0x10000 && r <= 0x10FFFF
}

View file

@ -1,582 +0,0 @@
// Copyright 2015-2019 Brett Vickers.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package etree
import (
"strconv"
"strings"
)
/*
A Path is a string that represents a search path through an etree starting
from the document root or an arbitrary element. Paths are used with the
Element object's Find* methods to locate and return desired elements.
A Path consists of a series of slash-separated "selectors", each of which may
be modified by one or more bracket-enclosed "filters". Selectors are used to
traverse the etree from element to element, while filters are used to narrow
the list of candidate elements at each node.
Although etree Path strings are similar to XPath strings
(https://www.w3.org/TR/1999/REC-xpath-19991116/), they have a more limited set
of selectors and filtering options.
The following selectors are supported by etree Path strings:
. Select the current element.
.. Select the parent of the current element.
* Select all child elements of the current element.
/ Select the root element when used at the start of a path.
// Select all descendants of the current element.
tag Select all child elements with a name matching the tag.
The following basic filters are supported by etree Path strings:
[@attrib] Keep elements with an attribute named attrib.
[@attrib='val'] Keep elements with an attribute named attrib and value matching val.
[tag] Keep elements with a child element named tag.
[tag='val'] Keep elements with a child element named tag and text matching val.
[n] Keep the n-th element, where n is a numeric index starting from 1.
The following function filters are also supported:
[text()] Keep elements with non-empty text.
[text()='val'] Keep elements whose text matches val.
[local-name()='val'] Keep elements whose un-prefixed tag matches val.
[name()='val'] Keep elements whose full tag exactly matches val.
[namespace-prefix()='val'] Keep elements whose namespace prefix matches val.
[namespace-uri()='val'] Keep elements whose namespace URI matches val.
Here are some examples of Path strings:
- Select the bookstore child element of the root element:
/bookstore
- Beginning from the root element, select the title elements of all
descendant book elements having a 'category' attribute of 'WEB':
//book[@category='WEB']/title
- Beginning from the current element, select the first descendant
book element with a title child element containing the text 'Great
Expectations':
.//book[title='Great Expectations'][1]
- Beginning from the current element, select all child elements of
book elements with an attribute 'language' set to 'english':
./book/*[@language='english']
- Beginning from the current element, select all child elements of
book elements containing the text 'special':
./book/*[text()='special']
- Beginning from the current element, select all descendant book
elements whose title child element has a 'language' attribute of 'french':
.//book/title[@language='french']/..
- Beginning from the current element, select all book elements
belonging to the http://www.w3.org/TR/html4/ namespace:
.//book[namespace-uri()='http://www.w3.org/TR/html4/']
*/
type Path struct {
segments []segment
}
// ErrPath is returned by path functions when an invalid etree path is provided.
type ErrPath string
// Error returns the string describing a path error.
func (err ErrPath) Error() string {
return "etree: " + string(err)
}
// CompilePath creates an optimized version of an XPath-like string that
// can be used to query elements in an element tree.
func CompilePath(path string) (Path, error) {
var comp compiler
segments := comp.parsePath(path)
if comp.err != ErrPath("") {
return Path{nil}, comp.err
}
return Path{segments}, nil
}
// MustCompilePath creates an optimized version of an XPath-like string that
// can be used to query elements in an element tree. Panics if an error
// occurs. Use this function to create Paths when you know the path is
// valid (i.e., if it's hard-coded).
func MustCompilePath(path string) Path {
p, err := CompilePath(path)
if err != nil {
panic(err)
}
return p
}
// A segment is a portion of a path between "/" characters.
// It contains one selector and zero or more [filters].
type segment struct {
sel selector
filters []filter
}
func (seg *segment) apply(e *Element, p *pather) {
seg.sel.apply(e, p)
for _, f := range seg.filters {
f.apply(p)
}
}
// A selector selects XML elements for consideration by the
// path traversal.
type selector interface {
apply(e *Element, p *pather)
}
// A filter pares down a list of candidate XML elements based
// on a path filter in [brackets].
type filter interface {
apply(p *pather)
}
// A pather is helper object that traverses an element tree using
// a Path object. It collects and deduplicates all elements matching
// the path query.
type pather struct {
queue fifo
results []*Element
inResults map[*Element]bool
candidates []*Element
scratch []*Element // used by filters
}
// A node represents an element and the remaining path segments that
// should be applied against it by the pather.
type node struct {
e *Element
segments []segment
}
func newPather() *pather {
return &pather{
results: make([]*Element, 0),
inResults: make(map[*Element]bool),
candidates: make([]*Element, 0),
scratch: make([]*Element, 0),
}
}
// traverse follows the path from the element e, collecting
// and then returning all elements that match the path's selectors
// and filters.
func (p *pather) traverse(e *Element, path Path) []*Element {
for p.queue.add(node{e, path.segments}); p.queue.len() > 0; {
p.eval(p.queue.remove().(node))
}
return p.results
}
// eval evalutes the current path node by applying the remaining
// path's selector rules against the node's element.
func (p *pather) eval(n node) {
p.candidates = p.candidates[0:0]
seg, remain := n.segments[0], n.segments[1:]
seg.apply(n.e, p)
if len(remain) == 0 {
for _, c := range p.candidates {
if in := p.inResults[c]; !in {
p.inResults[c] = true
p.results = append(p.results, c)
}
}
} else {
for _, c := range p.candidates {
p.queue.add(node{c, remain})
}
}
}
// A compiler generates a compiled path from a path string.
type compiler struct {
err ErrPath
}
// parsePath parses an XPath-like string describing a path
// through an element tree and returns a slice of segment
// descriptors.
func (c *compiler) parsePath(path string) []segment {
// If path ends with //, fix it
if strings.HasSuffix(path, "//") {
path = path + "*"
}
var segments []segment
// Check for an absolute path
if strings.HasPrefix(path, "/") {
segments = append(segments, segment{new(selectRoot), []filter{}})
path = path[1:]
}
// Split path into segments
for _, s := range splitPath(path) {
segments = append(segments, c.parseSegment(s))
if c.err != ErrPath("") {
break
}
}
return segments
}
func splitPath(path string) []string {
pieces := make([]string, 0)
start := 0
inquote := false
for i := 0; i+1 <= len(path); i++ {
if path[i] == '\'' {
inquote = !inquote
} else if path[i] == '/' && !inquote {
pieces = append(pieces, path[start:i])
start = i + 1
}
}
return append(pieces, path[start:])
}
// parseSegment parses a path segment between / characters.
func (c *compiler) parseSegment(path string) segment {
pieces := strings.Split(path, "[")
seg := segment{
sel: c.parseSelector(pieces[0]),
filters: []filter{},
}
for i := 1; i < len(pieces); i++ {
fpath := pieces[i]
if fpath[len(fpath)-1] != ']' {
c.err = ErrPath("path has invalid filter [brackets].")
break
}
seg.filters = append(seg.filters, c.parseFilter(fpath[:len(fpath)-1]))
}
return seg
}
// parseSelector parses a selector at the start of a path segment.
func (c *compiler) parseSelector(path string) selector {
switch path {
case ".":
return new(selectSelf)
case "..":
return new(selectParent)
case "*":
return new(selectChildren)
case "":
return new(selectDescendants)
default:
return newSelectChildrenByTag(path)
}
}
var fnTable = map[string]struct {
hasFn func(e *Element) bool
getValFn func(e *Element) string
}{
"local-name": {nil, (*Element).name},
"name": {nil, (*Element).FullTag},
"namespace-prefix": {nil, (*Element).namespacePrefix},
"namespace-uri": {nil, (*Element).NamespaceURI},
"text": {(*Element).hasText, (*Element).Text},
}
// parseFilter parses a path filter contained within [brackets].
func (c *compiler) parseFilter(path string) filter {
if len(path) == 0 {
c.err = ErrPath("path contains an empty filter expression.")
return nil
}
// Filter contains [@attr='val'], [fn()='val'], or [tag='val']?
eqindex := strings.Index(path, "='")
if eqindex >= 0 {
rindex := nextIndex(path, "'", eqindex+2)
if rindex != len(path)-1 {
c.err = ErrPath("path has mismatched filter quotes.")
return nil
}
key := path[:eqindex]
value := path[eqindex+2 : rindex]
switch {
case key[0] == '@':
return newFilterAttrVal(key[1:], value)
case strings.HasSuffix(key, "()"):
fn := key[:len(key)-2]
if t, ok := fnTable[fn]; ok && t.getValFn != nil {
return newFilterFuncVal(t.getValFn, value)
}
c.err = ErrPath("path has unknown function " + fn)
return nil
default:
return newFilterChildText(key, value)
}
}
// Filter contains [@attr], [N], [tag] or [fn()]
switch {
case path[0] == '@':
return newFilterAttr(path[1:])
case strings.HasSuffix(path, "()"):
fn := path[:len(path)-2]
if t, ok := fnTable[fn]; ok && t.hasFn != nil {
return newFilterFunc(t.hasFn)
}
c.err = ErrPath("path has unknown function " + fn)
return nil
case isInteger(path):
pos, _ := strconv.Atoi(path)
switch {
case pos > 0:
return newFilterPos(pos - 1)
default:
return newFilterPos(pos)
}
default:
return newFilterChild(path)
}
}
// selectSelf selects the current element into the candidate list.
type selectSelf struct{}
func (s *selectSelf) apply(e *Element, p *pather) {
p.candidates = append(p.candidates, e)
}
// selectRoot selects the element's root node.
type selectRoot struct{}
func (s *selectRoot) apply(e *Element, p *pather) {
root := e
for root.parent != nil {
root = root.parent
}
p.candidates = append(p.candidates, root)
}
// selectParent selects the element's parent into the candidate list.
type selectParent struct{}
func (s *selectParent) apply(e *Element, p *pather) {
if e.parent != nil {
p.candidates = append(p.candidates, e.parent)
}
}
// selectChildren selects the element's child elements into the
// candidate list.
type selectChildren struct{}
func (s *selectChildren) apply(e *Element, p *pather) {
for _, c := range e.Child {
if c, ok := c.(*Element); ok {
p.candidates = append(p.candidates, c)
}
}
}
// selectDescendants selects all descendant child elements
// of the element into the candidate list.
type selectDescendants struct{}
func (s *selectDescendants) apply(e *Element, p *pather) {
var queue fifo
for queue.add(e); queue.len() > 0; {
e := queue.remove().(*Element)
p.candidates = append(p.candidates, e)
for _, c := range e.Child {
if c, ok := c.(*Element); ok {
queue.add(c)
}
}
}
}
// selectChildrenByTag selects into the candidate list all child
// elements of the element having the specified tag.
type selectChildrenByTag struct {
space, tag string
}
func newSelectChildrenByTag(path string) *selectChildrenByTag {
s, l := spaceDecompose(path)
return &selectChildrenByTag{s, l}
}
func (s *selectChildrenByTag) apply(e *Element, p *pather) {
for _, c := range e.Child {
if c, ok := c.(*Element); ok && spaceMatch(s.space, c.Space) && s.tag == c.Tag {
p.candidates = append(p.candidates, c)
}
}
}
// filterPos filters the candidate list, keeping only the
// candidate at the specified index.
type filterPos struct {
index int
}
func newFilterPos(pos int) *filterPos {
return &filterPos{pos}
}
func (f *filterPos) apply(p *pather) {
if f.index >= 0 {
if f.index < len(p.candidates) {
p.scratch = append(p.scratch, p.candidates[f.index])
}
} else {
if -f.index <= len(p.candidates) {
p.scratch = append(p.scratch, p.candidates[len(p.candidates)+f.index])
}
}
p.candidates, p.scratch = p.scratch, p.candidates[0:0]
}
// filterAttr filters the candidate list for elements having
// the specified attribute.
type filterAttr struct {
space, key string
}
func newFilterAttr(str string) *filterAttr {
s, l := spaceDecompose(str)
return &filterAttr{s, l}
}
func (f *filterAttr) apply(p *pather) {
for _, c := range p.candidates {
for _, a := range c.Attr {
if spaceMatch(f.space, a.Space) && f.key == a.Key {
p.scratch = append(p.scratch, c)
break
}
}
}
p.candidates, p.scratch = p.scratch, p.candidates[0:0]
}
// filterAttrVal filters the candidate list for elements having
// the specified attribute with the specified value.
type filterAttrVal struct {
space, key, val string
}
func newFilterAttrVal(str, value string) *filterAttrVal {
s, l := spaceDecompose(str)
return &filterAttrVal{s, l, value}
}
func (f *filterAttrVal) apply(p *pather) {
for _, c := range p.candidates {
for _, a := range c.Attr {
if spaceMatch(f.space, a.Space) && f.key == a.Key && f.val == a.Value {
p.scratch = append(p.scratch, c)
break
}
}
}
p.candidates, p.scratch = p.scratch, p.candidates[0:0]
}
// filterFunc filters the candidate list for elements satisfying a custom
// boolean function.
type filterFunc struct {
fn func(e *Element) bool
}
func newFilterFunc(fn func(e *Element) bool) *filterFunc {
return &filterFunc{fn}
}
func (f *filterFunc) apply(p *pather) {
for _, c := range p.candidates {
if f.fn(c) {
p.scratch = append(p.scratch, c)
}
}
p.candidates, p.scratch = p.scratch, p.candidates[0:0]
}
// filterFuncVal filters the candidate list for elements containing a value
// matching the result of a custom function.
type filterFuncVal struct {
fn func(e *Element) string
val string
}
func newFilterFuncVal(fn func(e *Element) string, value string) *filterFuncVal {
return &filterFuncVal{fn, value}
}
func (f *filterFuncVal) apply(p *pather) {
for _, c := range p.candidates {
if f.fn(c) == f.val {
p.scratch = append(p.scratch, c)
}
}
p.candidates, p.scratch = p.scratch, p.candidates[0:0]
}
// filterChild filters the candidate list for elements having
// a child element with the specified tag.
type filterChild struct {
space, tag string
}
func newFilterChild(str string) *filterChild {
s, l := spaceDecompose(str)
return &filterChild{s, l}
}
func (f *filterChild) apply(p *pather) {
for _, c := range p.candidates {
for _, cc := range c.Child {
if cc, ok := cc.(*Element); ok &&
spaceMatch(f.space, cc.Space) &&
f.tag == cc.Tag {
p.scratch = append(p.scratch, c)
}
}
}
p.candidates, p.scratch = p.scratch, p.candidates[0:0]
}
// filterChildText filters the candidate list for elements having
// a child element with the specified tag and text.
type filterChildText struct {
space, tag, text string
}
func newFilterChildText(str, text string) *filterChildText {
s, l := spaceDecompose(str)
return &filterChildText{s, l, text}
}
func (f *filterChildText) apply(p *pather) {
for _, c := range p.candidates {
for _, cc := range c.Child {
if cc, ok := cc.(*Element); ok &&
spaceMatch(f.space, cc.Space) &&
f.tag == cc.Tag &&
f.text == cc.Text() {
p.scratch = append(p.scratch, c)
}
}
}
p.candidates, p.scratch = p.scratch, p.candidates[0:0]
}

View file

@ -1,20 +0,0 @@
Copyright (C) 2013 Blake Mizerany
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

File diff suppressed because it is too large Load diff

View file

@ -1,316 +0,0 @@
// Package quantile computes approximate quantiles over an unbounded data
// stream within low memory and CPU bounds.
//
// A small amount of accuracy is traded to achieve the above properties.
//
// Multiple streams can be merged before calling Query to generate a single set
// of results. This is meaningful when the streams represent the same type of
// data. See Merge and Samples.
//
// For more detailed information about the algorithm used, see:
//
// Effective Computation of Biased Quantiles over Data Streams
//
// http://www.cs.rutgers.edu/~muthu/bquant.pdf
package quantile
import (
"math"
"sort"
)
// Sample holds an observed value and meta information for compression. JSON
// tags have been added for convenience.
type Sample struct {
Value float64 `json:",string"`
Width float64 `json:",string"`
Delta float64 `json:",string"`
}
// Samples represents a slice of samples. It implements sort.Interface.
type Samples []Sample
func (a Samples) Len() int { return len(a) }
func (a Samples) Less(i, j int) bool { return a[i].Value < a[j].Value }
func (a Samples) Swap(i, j int) { a[i], a[j] = a[j], a[i] }
type invariant func(s *stream, r float64) float64
// NewLowBiased returns an initialized Stream for low-biased quantiles
// (e.g. 0.01, 0.1, 0.5) where the needed quantiles are not known a priori, but
// error guarantees can still be given even for the lower ranks of the data
// distribution.
//
// The provided epsilon is a relative error, i.e. the true quantile of a value
// returned by a query is guaranteed to be within (1±Epsilon)*Quantile.
//
// See http://www.cs.rutgers.edu/~muthu/bquant.pdf for time, space, and error
// properties.
func NewLowBiased(epsilon float64) *Stream {
ƒ := func(s *stream, r float64) float64 {
return 2 * epsilon * r
}
return newStream(ƒ)
}
// NewHighBiased returns an initialized Stream for high-biased quantiles
// (e.g. 0.01, 0.1, 0.5) where the needed quantiles are not known a priori, but
// error guarantees can still be given even for the higher ranks of the data
// distribution.
//
// The provided epsilon is a relative error, i.e. the true quantile of a value
// returned by a query is guaranteed to be within 1-(1±Epsilon)*(1-Quantile).
//
// See http://www.cs.rutgers.edu/~muthu/bquant.pdf for time, space, and error
// properties.
func NewHighBiased(epsilon float64) *Stream {
ƒ := func(s *stream, r float64) float64 {
return 2 * epsilon * (s.n - r)
}
return newStream(ƒ)
}
// NewTargeted returns an initialized Stream concerned with a particular set of
// quantile values that are supplied a priori. Knowing these a priori reduces
// space and computation time. The targets map maps the desired quantiles to
// their absolute errors, i.e. the true quantile of a value returned by a query
// is guaranteed to be within (Quantile±Epsilon).
//
// See http://www.cs.rutgers.edu/~muthu/bquant.pdf for time, space, and error properties.
func NewTargeted(targetMap map[float64]float64) *Stream {
// Convert map to slice to avoid slow iterations on a map.
// ƒ is called on the hot path, so converting the map to a slice
// beforehand results in significant CPU savings.
targets := targetMapToSlice(targetMap)
ƒ := func(s *stream, r float64) float64 {
var m = math.MaxFloat64
var f float64
for _, t := range targets {
if t.quantile*s.n <= r {
f = (2 * t.epsilon * r) / t.quantile
} else {
f = (2 * t.epsilon * (s.n - r)) / (1 - t.quantile)
}
if f < m {
m = f
}
}
return m
}
return newStream(ƒ)
}
type target struct {
quantile float64
epsilon float64
}
func targetMapToSlice(targetMap map[float64]float64) []target {
targets := make([]target, 0, len(targetMap))
for quantile, epsilon := range targetMap {
t := target{
quantile: quantile,
epsilon: epsilon,
}
targets = append(targets, t)
}
return targets
}
// Stream computes quantiles for a stream of float64s. It is not thread-safe by
// design. Take care when using across multiple goroutines.
type Stream struct {
*stream
b Samples
sorted bool
}
func newStream(ƒ invariant) *Stream {
x := &stream{ƒ: ƒ}
return &Stream{x, make(Samples, 0, 500), true}
}
// Insert inserts v into the stream.
func (s *Stream) Insert(v float64) {
s.insert(Sample{Value: v, Width: 1})
}
func (s *Stream) insert(sample Sample) {
s.b = append(s.b, sample)
s.sorted = false
if len(s.b) == cap(s.b) {
s.flush()
}
}
// Query returns the computed qth percentiles value. If s was created with
// NewTargeted, and q is not in the set of quantiles provided a priori, Query
// will return an unspecified result.
func (s *Stream) Query(q float64) float64 {
if !s.flushed() {
// Fast path when there hasn't been enough data for a flush;
// this also yields better accuracy for small sets of data.
l := len(s.b)
if l == 0 {
return 0
}
i := int(math.Ceil(float64(l) * q))
if i > 0 {
i -= 1
}
s.maybeSort()
return s.b[i].Value
}
s.flush()
return s.stream.query(q)
}
// Merge merges samples into the underlying streams samples. This is handy when
// merging multiple streams from separate threads, database shards, etc.
//
// ATTENTION: This method is broken and does not yield correct results. The
// underlying algorithm is not capable of merging streams correctly.
func (s *Stream) Merge(samples Samples) {
sort.Sort(samples)
s.stream.merge(samples)
}
// Reset reinitializes and clears the list reusing the samples buffer memory.
func (s *Stream) Reset() {
s.stream.reset()
s.b = s.b[:0]
}
// Samples returns stream samples held by s.
func (s *Stream) Samples() Samples {
if !s.flushed() {
return s.b
}
s.flush()
return s.stream.samples()
}
// Count returns the total number of samples observed in the stream
// since initialization.
func (s *Stream) Count() int {
return len(s.b) + s.stream.count()
}
func (s *Stream) flush() {
s.maybeSort()
s.stream.merge(s.b)
s.b = s.b[:0]
}
func (s *Stream) maybeSort() {
if !s.sorted {
s.sorted = true
sort.Sort(s.b)
}
}
func (s *Stream) flushed() bool {
return len(s.stream.l) > 0
}
type stream struct {
n float64
l []Sample
ƒ invariant
}
func (s *stream) reset() {
s.l = s.l[:0]
s.n = 0
}
func (s *stream) insert(v float64) {
s.merge(Samples{{v, 1, 0}})
}
func (s *stream) merge(samples Samples) {
// TODO(beorn7): This tries to merge not only individual samples, but
// whole summaries. The paper doesn't mention merging summaries at
// all. Unittests show that the merging is inaccurate. Find out how to
// do merges properly.
var r float64
i := 0
for _, sample := range samples {
for ; i < len(s.l); i++ {
c := s.l[i]
if c.Value > sample.Value {
// Insert at position i.
s.l = append(s.l, Sample{})
copy(s.l[i+1:], s.l[i:])
s.l[i] = Sample{
sample.Value,
sample.Width,
math.Max(sample.Delta, math.Floor(s.ƒ(s, r))-1),
// TODO(beorn7): How to calculate delta correctly?
}
i++
goto inserted
}
r += c.Width
}
s.l = append(s.l, Sample{sample.Value, sample.Width, 0})
i++
inserted:
s.n += sample.Width
r += sample.Width
}
s.compress()
}
func (s *stream) count() int {
return int(s.n)
}
func (s *stream) query(q float64) float64 {
t := math.Ceil(q * s.n)
t += math.Ceil(s.ƒ(s, t) / 2)
p := s.l[0]
var r float64
for _, c := range s.l[1:] {
r += p.Width
if r+c.Width+c.Delta > t {
return p.Value
}
p = c
}
return p.Value
}
func (s *stream) compress() {
if len(s.l) < 2 {
return
}
x := s.l[len(s.l)-1]
xi := len(s.l) - 1
r := s.n - 1 - x.Width
for i := len(s.l) - 2; i >= 0; i-- {
c := s.l[i]
if c.Width+x.Width+x.Delta <= s.ƒ(s, r) {
x.Width += c.Width
s.l[xi] = x
// Remove element at i.
copy(s.l[i:], s.l[i+1:])
s.l = s.l[:len(s.l)-1]
xi -= 1
} else {
x = c
xi = i
}
r -= c.Width
}
}
func (s *stream) samples() Samples {
samples := make(Samples, len(s.l))
copy(samples, s.l)
return samples
}

View file

@ -1,22 +0,0 @@
# Compiled Object files, Static and Dynamic libs (Shared Objects)
*.o
*.a
*.so
# Folders
_obj
_test
# Architecture specific extensions/prefixes
*.[568vq]
[568vq].out
*.cgo1.go
*.cgo2.c
_cgo_defun.c
_cgo_gotypes.go
_cgo_export.*
_testmain.go
*.exe

View file

@ -1,10 +0,0 @@
language: go
go:
- 1.7
- 1.x
- tip
before_install:
- go get github.com/mattn/goveralls
- go get golang.org/x/tools/cmd/cover
script:
- $HOME/gopath/bin/goveralls -service=travis-ci

View file

@ -1,20 +0,0 @@
The MIT License (MIT)
Copyright (c) 2014 Cenk Altı
Permission is hereby granted, free of charge, to any person obtaining a copy of
this software and associated documentation files (the "Software"), to deal in
the Software without restriction, including without limitation the rights to
use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
the Software, and to permit persons to whom the Software is furnished to do so,
subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

View file

@ -1,30 +0,0 @@
# Exponential Backoff [![GoDoc][godoc image]][godoc] [![Build Status][travis image]][travis] [![Coverage Status][coveralls image]][coveralls]
This is a Go port of the exponential backoff algorithm from [Google's HTTP Client Library for Java][google-http-java-client].
[Exponential backoff][exponential backoff wiki]
is an algorithm that uses feedback to multiplicatively decrease the rate of some process,
in order to gradually find an acceptable rate.
The retries exponentially increase and stop increasing when a certain threshold is met.
## Usage
See https://godoc.org/github.com/cenkalti/backoff#pkg-examples
## Contributing
* I would like to keep this library as small as possible.
* Please don't send a PR without opening an issue and discussing it first.
* If proposed change is not a common use case, I will probably not accept it.
[godoc]: https://godoc.org/github.com/cenkalti/backoff
[godoc image]: https://godoc.org/github.com/cenkalti/backoff?status.png
[travis]: https://travis-ci.org/cenkalti/backoff
[travis image]: https://travis-ci.org/cenkalti/backoff.png?branch=master
[coveralls]: https://coveralls.io/github/cenkalti/backoff?branch=master
[coveralls image]: https://coveralls.io/repos/github/cenkalti/backoff/badge.svg?branch=master
[google-http-java-client]: https://github.com/google/google-http-java-client/blob/da1aa993e90285ec18579f1553339b00e19b3ab5/google-http-client/src/main/java/com/google/api/client/util/ExponentialBackOff.java
[exponential backoff wiki]: http://en.wikipedia.org/wiki/Exponential_backoff
[advanced example]: https://godoc.org/github.com/cenkalti/backoff#example_

View file

@ -1,66 +0,0 @@
// Package backoff implements backoff algorithms for retrying operations.
//
// Use Retry function for retrying operations that may fail.
// If Retry does not meet your needs,
// copy/paste the function into your project and modify as you wish.
//
// There is also Ticker type similar to time.Ticker.
// You can use it if you need to work with channels.
//
// See Examples section below for usage examples.
package backoff
import "time"
// BackOff is a backoff policy for retrying an operation.
type BackOff interface {
// NextBackOff returns the duration to wait before retrying the operation,
// or backoff. Stop to indicate that no more retries should be made.
//
// Example usage:
//
// duration := backoff.NextBackOff();
// if (duration == backoff.Stop) {
// // Do not retry operation.
// } else {
// // Sleep for duration and retry operation.
// }
//
NextBackOff() time.Duration
// Reset to initial state.
Reset()
}
// Stop indicates that no more retries should be made for use in NextBackOff().
const Stop time.Duration = -1
// ZeroBackOff is a fixed backoff policy whose backoff time is always zero,
// meaning that the operation is retried immediately without waiting, indefinitely.
type ZeroBackOff struct{}
func (b *ZeroBackOff) Reset() {}
func (b *ZeroBackOff) NextBackOff() time.Duration { return 0 }
// StopBackOff is a fixed backoff policy that always returns backoff.Stop for
// NextBackOff(), meaning that the operation should never be retried.
type StopBackOff struct{}
func (b *StopBackOff) Reset() {}
func (b *StopBackOff) NextBackOff() time.Duration { return Stop }
// ConstantBackOff is a backoff policy that always returns the same backoff delay.
// This is in contrast to an exponential backoff policy,
// which returns a delay that grows longer as you call NextBackOff() over and over again.
type ConstantBackOff struct {
Interval time.Duration
}
func (b *ConstantBackOff) Reset() {}
func (b *ConstantBackOff) NextBackOff() time.Duration { return b.Interval }
func NewConstantBackOff(d time.Duration) *ConstantBackOff {
return &ConstantBackOff{Interval: d}
}

View file

@ -1,63 +0,0 @@
package backoff
import (
"context"
"time"
)
// BackOffContext is a backoff policy that stops retrying after the context
// is canceled.
type BackOffContext interface {
BackOff
Context() context.Context
}
type backOffContext struct {
BackOff
ctx context.Context
}
// WithContext returns a BackOffContext with context ctx
//
// ctx must not be nil
func WithContext(b BackOff, ctx context.Context) BackOffContext {
if ctx == nil {
panic("nil context")
}
if b, ok := b.(*backOffContext); ok {
return &backOffContext{
BackOff: b.BackOff,
ctx: ctx,
}
}
return &backOffContext{
BackOff: b,
ctx: ctx,
}
}
func ensureContext(b BackOff) BackOffContext {
if cb, ok := b.(BackOffContext); ok {
return cb
}
return WithContext(b, context.Background())
}
func (b *backOffContext) Context() context.Context {
return b.ctx
}
func (b *backOffContext) NextBackOff() time.Duration {
select {
case <-b.ctx.Done():
return Stop
default:
}
next := b.BackOff.NextBackOff()
if deadline, ok := b.ctx.Deadline(); ok && deadline.Sub(time.Now()) < next {
return Stop
}
return next
}

View file

@ -1,153 +0,0 @@
package backoff
import (
"math/rand"
"time"
)
/*
ExponentialBackOff is a backoff implementation that increases the backoff
period for each retry attempt using a randomization function that grows exponentially.
NextBackOff() is calculated using the following formula:
randomized interval =
RetryInterval * (random value in range [1 - RandomizationFactor, 1 + RandomizationFactor])
In other words NextBackOff() will range between the randomization factor
percentage below and above the retry interval.
For example, given the following parameters:
RetryInterval = 2
RandomizationFactor = 0.5
Multiplier = 2
the actual backoff period used in the next retry attempt will range between 1 and 3 seconds,
multiplied by the exponential, that is, between 2 and 6 seconds.
Note: MaxInterval caps the RetryInterval and not the randomized interval.
If the time elapsed since an ExponentialBackOff instance is created goes past the
MaxElapsedTime, then the method NextBackOff() starts returning backoff.Stop.
The elapsed time can be reset by calling Reset().
Example: Given the following default arguments, for 10 tries the sequence will be,
and assuming we go over the MaxElapsedTime on the 10th try:
Request # RetryInterval (seconds) Randomized Interval (seconds)
1 0.5 [0.25, 0.75]
2 0.75 [0.375, 1.125]
3 1.125 [0.562, 1.687]
4 1.687 [0.8435, 2.53]
5 2.53 [1.265, 3.795]
6 3.795 [1.897, 5.692]
7 5.692 [2.846, 8.538]
8 8.538 [4.269, 12.807]
9 12.807 [6.403, 19.210]
10 19.210 backoff.Stop
Note: Implementation is not thread-safe.
*/
type ExponentialBackOff struct {
InitialInterval time.Duration
RandomizationFactor float64
Multiplier float64
MaxInterval time.Duration
// After MaxElapsedTime the ExponentialBackOff stops.
// It never stops if MaxElapsedTime == 0.
MaxElapsedTime time.Duration
Clock Clock
currentInterval time.Duration
startTime time.Time
}
// Clock is an interface that returns current time for BackOff.
type Clock interface {
Now() time.Time
}
// Default values for ExponentialBackOff.
const (
DefaultInitialInterval = 500 * time.Millisecond
DefaultRandomizationFactor = 0.5
DefaultMultiplier = 1.5
DefaultMaxInterval = 60 * time.Second
DefaultMaxElapsedTime = 15 * time.Minute
)
// NewExponentialBackOff creates an instance of ExponentialBackOff using default values.
func NewExponentialBackOff() *ExponentialBackOff {
b := &ExponentialBackOff{
InitialInterval: DefaultInitialInterval,
RandomizationFactor: DefaultRandomizationFactor,
Multiplier: DefaultMultiplier,
MaxInterval: DefaultMaxInterval,
MaxElapsedTime: DefaultMaxElapsedTime,
Clock: SystemClock,
}
b.Reset()
return b
}
type systemClock struct{}
func (t systemClock) Now() time.Time {
return time.Now()
}
// SystemClock implements Clock interface that uses time.Now().
var SystemClock = systemClock{}
// Reset the interval back to the initial retry interval and restarts the timer.
func (b *ExponentialBackOff) Reset() {
b.currentInterval = b.InitialInterval
b.startTime = b.Clock.Now()
}
// NextBackOff calculates the next backoff interval using the formula:
// Randomized interval = RetryInterval +/- (RandomizationFactor * RetryInterval)
func (b *ExponentialBackOff) NextBackOff() time.Duration {
// Make sure we have not gone over the maximum elapsed time.
if b.MaxElapsedTime != 0 && b.GetElapsedTime() > b.MaxElapsedTime {
return Stop
}
defer b.incrementCurrentInterval()
return getRandomValueFromInterval(b.RandomizationFactor, rand.Float64(), b.currentInterval)
}
// GetElapsedTime returns the elapsed time since an ExponentialBackOff instance
// is created and is reset when Reset() is called.
//
// The elapsed time is computed using time.Now().UnixNano(). It is
// safe to call even while the backoff policy is used by a running
// ticker.
func (b *ExponentialBackOff) GetElapsedTime() time.Duration {
return b.Clock.Now().Sub(b.startTime)
}
// Increments the current interval by multiplying it with the multiplier.
func (b *ExponentialBackOff) incrementCurrentInterval() {
// Check for overflow, if overflow is detected set the current interval to the max interval.
if float64(b.currentInterval) >= float64(b.MaxInterval)/b.Multiplier {
b.currentInterval = b.MaxInterval
} else {
b.currentInterval = time.Duration(float64(b.currentInterval) * b.Multiplier)
}
}
// Returns a random value from the following interval:
// [randomizationFactor * currentInterval, randomizationFactor * currentInterval].
func getRandomValueFromInterval(randomizationFactor, random float64, currentInterval time.Duration) time.Duration {
var delta = randomizationFactor * float64(currentInterval)
var minInterval = float64(currentInterval) - delta
var maxInterval = float64(currentInterval) + delta
// Get a random value from the range [minInterval, maxInterval].
// The formula used below has a +1 because if the minInterval is 1 and the maxInterval is 3 then
// we want a 33% chance for selecting either 1, 2 or 3.
return time.Duration(minInterval + (random * (maxInterval - minInterval + 1)))
}

View file

@ -1,82 +0,0 @@
package backoff
import "time"
// An Operation is executing by Retry() or RetryNotify().
// The operation will be retried using a backoff policy if it returns an error.
type Operation func() error
// Notify is a notify-on-error function. It receives an operation error and
// backoff delay if the operation failed (with an error).
//
// NOTE that if the backoff policy stated to stop retrying,
// the notify function isn't called.
type Notify func(error, time.Duration)
// Retry the operation o until it does not return error or BackOff stops.
// o is guaranteed to be run at least once.
//
// If o returns a *PermanentError, the operation is not retried, and the
// wrapped error is returned.
//
// Retry sleeps the goroutine for the duration returned by BackOff after a
// failed operation returns.
func Retry(o Operation, b BackOff) error { return RetryNotify(o, b, nil) }
// RetryNotify calls notify function with the error and wait duration
// for each failed attempt before sleep.
func RetryNotify(operation Operation, b BackOff, notify Notify) error {
var err error
var next time.Duration
var t *time.Timer
cb := ensureContext(b)
b.Reset()
for {
if err = operation(); err == nil {
return nil
}
if permanent, ok := err.(*PermanentError); ok {
return permanent.Err
}
if next = cb.NextBackOff(); next == Stop {
return err
}
if notify != nil {
notify(err, next)
}
if t == nil {
t = time.NewTimer(next)
defer t.Stop()
} else {
t.Reset(next)
}
select {
case <-cb.Context().Done():
return err
case <-t.C:
}
}
}
// PermanentError signals that the operation should not be retried.
type PermanentError struct {
Err error
}
func (e *PermanentError) Error() string {
return e.Err.Error()
}
// Permanent wraps the given err in a *PermanentError.
func Permanent(err error) *PermanentError {
return &PermanentError{
Err: err,
}
}

View file

@ -1,82 +0,0 @@
package backoff
import (
"sync"
"time"
)
// Ticker holds a channel that delivers `ticks' of a clock at times reported by a BackOff.
//
// Ticks will continue to arrive when the previous operation is still running,
// so operations that take a while to fail could run in quick succession.
type Ticker struct {
C <-chan time.Time
c chan time.Time
b BackOffContext
stop chan struct{}
stopOnce sync.Once
}
// NewTicker returns a new Ticker containing a channel that will send
// the time at times specified by the BackOff argument. Ticker is
// guaranteed to tick at least once. The channel is closed when Stop
// method is called or BackOff stops. It is not safe to manipulate the
// provided backoff policy (notably calling NextBackOff or Reset)
// while the ticker is running.
func NewTicker(b BackOff) *Ticker {
c := make(chan time.Time)
t := &Ticker{
C: c,
c: c,
b: ensureContext(b),
stop: make(chan struct{}),
}
t.b.Reset()
go t.run()
return t
}
// Stop turns off a ticker. After Stop, no more ticks will be sent.
func (t *Ticker) Stop() {
t.stopOnce.Do(func() { close(t.stop) })
}
func (t *Ticker) run() {
c := t.c
defer close(c)
// Ticker is guaranteed to tick at least once.
afterC := t.send(time.Now())
for {
if afterC == nil {
return
}
select {
case tick := <-afterC:
afterC = t.send(tick)
case <-t.stop:
t.c = nil // Prevent future ticks from being sent to the channel.
return
case <-t.b.Context().Done():
return
}
}
}
func (t *Ticker) send(tick time.Time) <-chan time.Time {
select {
case t.c <- tick:
case <-t.stop:
return nil
}
next := t.b.NextBackOff()
if next == Stop {
t.Stop()
return nil
}
return time.After(next)
}

View file

@ -1,35 +0,0 @@
package backoff
import "time"
/*
WithMaxRetries creates a wrapper around another BackOff, which will
return Stop if NextBackOff() has been called too many times since
the last time Reset() was called
Note: Implementation is not thread-safe.
*/
func WithMaxRetries(b BackOff, max uint64) BackOff {
return &backOffTries{delegate: b, maxTries: max}
}
type backOffTries struct {
delegate BackOff
maxTries uint64
numTries uint64
}
func (b *backOffTries) NextBackOff() time.Duration {
if b.maxTries > 0 {
if b.maxTries <= b.numTries {
return Stop
}
b.numTries++
}
return b.delegate.NextBackOff()
}
func (b *backOffTries) Reset() {
b.numTries = 0
b.delegate.Reset()
}

View file

@ -1,8 +0,0 @@
language: go
go:
- "1.x"
- master
env:
- TAGS=""
- TAGS="-tags purego"
script: go test $TAGS -v ./...

View file

@ -1,22 +0,0 @@
Copyright (c) 2016 Caleb Spare
MIT License
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

View file

@ -1,67 +0,0 @@
# xxhash
[![GoDoc](https://godoc.org/github.com/cespare/xxhash?status.svg)](https://godoc.org/github.com/cespare/xxhash)
[![Build Status](https://travis-ci.org/cespare/xxhash.svg?branch=master)](https://travis-ci.org/cespare/xxhash)
xxhash is a Go implementation of the 64-bit
[xxHash](http://cyan4973.github.io/xxHash/) algorithm, XXH64. This is a
high-quality hashing algorithm that is much faster than anything in the Go
standard library.
This package provides a straightforward API:
```
func Sum64(b []byte) uint64
func Sum64String(s string) uint64
type Digest struct{ ... }
func New() *Digest
```
The `Digest` type implements hash.Hash64. Its key methods are:
```
func (*Digest) Write([]byte) (int, error)
func (*Digest) WriteString(string) (int, error)
func (*Digest) Sum64() uint64
```
This implementation provides a fast pure-Go implementation and an even faster
assembly implementation for amd64.
## Compatibility
This package is in a module and the latest code is in version 2 of the module.
You need a version of Go with at least "minimal module compatibility" to use
github.com/cespare/xxhash/v2:
* 1.9.7+ for Go 1.9
* 1.10.3+ for Go 1.10
* Go 1.11 or later
I recommend using the latest release of Go.
## Benchmarks
Here are some quick benchmarks comparing the pure-Go and assembly
implementations of Sum64.
| input size | purego | asm |
| --- | --- | --- |
| 5 B | 979.66 MB/s | 1291.17 MB/s |
| 100 B | 7475.26 MB/s | 7973.40 MB/s |
| 4 KB | 17573.46 MB/s | 17602.65 MB/s |
| 10 MB | 17131.46 MB/s | 17142.16 MB/s |
These numbers were generated on Ubuntu 18.04 with an Intel i7-8700K CPU using
the following commands under Go 1.11.2:
```
$ go test -tags purego -benchtime 10s -bench '/xxhash,direct,bytes'
$ go test -benchtime 10s -bench '/xxhash,direct,bytes'
```
## Projects using this package
- [InfluxDB](https://github.com/influxdata/influxdb)
- [Prometheus](https://github.com/prometheus/prometheus)
- [FreeCache](https://github.com/coocood/freecache)

View file

@ -1,3 +0,0 @@
module github.com/cespare/xxhash/v2
go 1.11

View file

View file

@ -1,236 +0,0 @@
// Package xxhash implements the 64-bit variant of xxHash (XXH64) as described
// at http://cyan4973.github.io/xxHash/.
package xxhash
import (
"encoding/binary"
"errors"
"math/bits"
)
const (
prime1 uint64 = 11400714785074694791
prime2 uint64 = 14029467366897019727
prime3 uint64 = 1609587929392839161
prime4 uint64 = 9650029242287828579
prime5 uint64 = 2870177450012600261
)
// NOTE(caleb): I'm using both consts and vars of the primes. Using consts where
// possible in the Go code is worth a small (but measurable) performance boost
// by avoiding some MOVQs. Vars are needed for the asm and also are useful for
// convenience in the Go code in a few places where we need to intentionally
// avoid constant arithmetic (e.g., v1 := prime1 + prime2 fails because the
// result overflows a uint64).
var (
prime1v = prime1
prime2v = prime2
prime3v = prime3
prime4v = prime4
prime5v = prime5
)
// Digest implements hash.Hash64.
type Digest struct {
v1 uint64
v2 uint64
v3 uint64
v4 uint64
total uint64
mem [32]byte
n int // how much of mem is used
}
// New creates a new Digest that computes the 64-bit xxHash algorithm.
func New() *Digest {
var d Digest
d.Reset()
return &d
}
// Reset clears the Digest's state so that it can be reused.
func (d *Digest) Reset() {
d.v1 = prime1v + prime2
d.v2 = prime2
d.v3 = 0
d.v4 = -prime1v
d.total = 0
d.n = 0
}
// Size always returns 8 bytes.
func (d *Digest) Size() int { return 8 }
// BlockSize always returns 32 bytes.
func (d *Digest) BlockSize() int { return 32 }
// Write adds more data to d. It always returns len(b), nil.
func (d *Digest) Write(b []byte) (n int, err error) {
n = len(b)
d.total += uint64(n)
if d.n+n < 32 {
// This new data doesn't even fill the current block.
copy(d.mem[d.n:], b)
d.n += n
return
}
if d.n > 0 {
// Finish off the partial block.
copy(d.mem[d.n:], b)
d.v1 = round(d.v1, u64(d.mem[0:8]))
d.v2 = round(d.v2, u64(d.mem[8:16]))
d.v3 = round(d.v3, u64(d.mem[16:24]))
d.v4 = round(d.v4, u64(d.mem[24:32]))
b = b[32-d.n:]
d.n = 0
}
if len(b) >= 32 {
// One or more full blocks left.
nw := writeBlocks(d, b)
b = b[nw:]
}
// Store any remaining partial block.
copy(d.mem[:], b)
d.n = len(b)
return
}
// Sum appends the current hash to b and returns the resulting slice.
func (d *Digest) Sum(b []byte) []byte {
s := d.Sum64()
return append(
b,
byte(s>>56),
byte(s>>48),
byte(s>>40),
byte(s>>32),
byte(s>>24),
byte(s>>16),
byte(s>>8),
byte(s),
)
}
// Sum64 returns the current hash.
func (d *Digest) Sum64() uint64 {
var h uint64
if d.total >= 32 {
v1, v2, v3, v4 := d.v1, d.v2, d.v3, d.v4
h = rol1(v1) + rol7(v2) + rol12(v3) + rol18(v4)
h = mergeRound(h, v1)
h = mergeRound(h, v2)
h = mergeRound(h, v3)
h = mergeRound(h, v4)
} else {
h = d.v3 + prime5
}
h += d.total
i, end := 0, d.n
for ; i+8 <= end; i += 8 {
k1 := round(0, u64(d.mem[i:i+8]))
h ^= k1
h = rol27(h)*prime1 + prime4
}
if i+4 <= end {
h ^= uint64(u32(d.mem[i:i+4])) * prime1
h = rol23(h)*prime2 + prime3
i += 4
}
for i < end {
h ^= uint64(d.mem[i]) * prime5
h = rol11(h) * prime1
i++
}
h ^= h >> 33
h *= prime2
h ^= h >> 29
h *= prime3
h ^= h >> 32
return h
}
const (
magic = "xxh\x06"
marshaledSize = len(magic) + 8*5 + 32
)
// MarshalBinary implements the encoding.BinaryMarshaler interface.
func (d *Digest) MarshalBinary() ([]byte, error) {
b := make([]byte, 0, marshaledSize)
b = append(b, magic...)
b = appendUint64(b, d.v1)
b = appendUint64(b, d.v2)
b = appendUint64(b, d.v3)
b = appendUint64(b, d.v4)
b = appendUint64(b, d.total)
b = append(b, d.mem[:d.n]...)
b = b[:len(b)+len(d.mem)-d.n]
return b, nil
}
// UnmarshalBinary implements the encoding.BinaryUnmarshaler interface.
func (d *Digest) UnmarshalBinary(b []byte) error {
if len(b) < len(magic) || string(b[:len(magic)]) != magic {
return errors.New("xxhash: invalid hash state identifier")
}
if len(b) != marshaledSize {
return errors.New("xxhash: invalid hash state size")
}
b = b[len(magic):]
b, d.v1 = consumeUint64(b)
b, d.v2 = consumeUint64(b)
b, d.v3 = consumeUint64(b)
b, d.v4 = consumeUint64(b)
b, d.total = consumeUint64(b)
copy(d.mem[:], b)
b = b[len(d.mem):]
d.n = int(d.total % uint64(len(d.mem)))
return nil
}
func appendUint64(b []byte, x uint64) []byte {
var a [8]byte
binary.LittleEndian.PutUint64(a[:], x)
return append(b, a[:]...)
}
func consumeUint64(b []byte) ([]byte, uint64) {
x := u64(b)
return b[8:], x
}
func u64(b []byte) uint64 { return binary.LittleEndian.Uint64(b) }
func u32(b []byte) uint32 { return binary.LittleEndian.Uint32(b) }
func round(acc, input uint64) uint64 {
acc += input * prime2
acc = rol31(acc)
acc *= prime1
return acc
}
func mergeRound(acc, val uint64) uint64 {
val = round(0, val)
acc ^= val
acc = acc*prime1 + prime4
return acc
}
func rol1(x uint64) uint64 { return bits.RotateLeft64(x, 1) }
func rol7(x uint64) uint64 { return bits.RotateLeft64(x, 7) }
func rol11(x uint64) uint64 { return bits.RotateLeft64(x, 11) }
func rol12(x uint64) uint64 { return bits.RotateLeft64(x, 12) }
func rol18(x uint64) uint64 { return bits.RotateLeft64(x, 18) }
func rol23(x uint64) uint64 { return bits.RotateLeft64(x, 23) }
func rol27(x uint64) uint64 { return bits.RotateLeft64(x, 27) }
func rol31(x uint64) uint64 { return bits.RotateLeft64(x, 31) }

View file

@ -1,13 +0,0 @@
// +build !appengine
// +build gc
// +build !purego
package xxhash
// Sum64 computes the 64-bit xxHash digest of b.
//
//go:noescape
func Sum64(b []byte) uint64
//go:noescape
func writeBlocks(d *Digest, b []byte) int

View file

@ -1,215 +0,0 @@
// +build !appengine
// +build gc
// +build !purego
#include "textflag.h"
// Register allocation:
// AX h
// CX pointer to advance through b
// DX n
// BX loop end
// R8 v1, k1
// R9 v2
// R10 v3
// R11 v4
// R12 tmp
// R13 prime1v
// R14 prime2v
// R15 prime4v
// round reads from and advances the buffer pointer in CX.
// It assumes that R13 has prime1v and R14 has prime2v.
#define round(r) \
MOVQ (CX), R12 \
ADDQ $8, CX \
IMULQ R14, R12 \
ADDQ R12, r \
ROLQ $31, r \
IMULQ R13, r
// mergeRound applies a merge round on the two registers acc and val.
// It assumes that R13 has prime1v, R14 has prime2v, and R15 has prime4v.
#define mergeRound(acc, val) \
IMULQ R14, val \
ROLQ $31, val \
IMULQ R13, val \
XORQ val, acc \
IMULQ R13, acc \
ADDQ R15, acc
// func Sum64(b []byte) uint64
TEXT ·Sum64(SB), NOSPLIT, $0-32
// Load fixed primes.
MOVQ ·prime1v(SB), R13
MOVQ ·prime2v(SB), R14
MOVQ ·prime4v(SB), R15
// Load slice.
MOVQ b_base+0(FP), CX
MOVQ b_len+8(FP), DX
LEAQ (CX)(DX*1), BX
// The first loop limit will be len(b)-32.
SUBQ $32, BX
// Check whether we have at least one block.
CMPQ DX, $32
JLT noBlocks
// Set up initial state (v1, v2, v3, v4).
MOVQ R13, R8
ADDQ R14, R8
MOVQ R14, R9
XORQ R10, R10
XORQ R11, R11
SUBQ R13, R11
// Loop until CX > BX.
blockLoop:
round(R8)
round(R9)
round(R10)
round(R11)
CMPQ CX, BX
JLE blockLoop
MOVQ R8, AX
ROLQ $1, AX
MOVQ R9, R12
ROLQ $7, R12
ADDQ R12, AX
MOVQ R10, R12
ROLQ $12, R12
ADDQ R12, AX
MOVQ R11, R12
ROLQ $18, R12
ADDQ R12, AX
mergeRound(AX, R8)
mergeRound(AX, R9)
mergeRound(AX, R10)
mergeRound(AX, R11)
JMP afterBlocks
noBlocks:
MOVQ ·prime5v(SB), AX
afterBlocks:
ADDQ DX, AX
// Right now BX has len(b)-32, and we want to loop until CX > len(b)-8.
ADDQ $24, BX
CMPQ CX, BX
JG fourByte
wordLoop:
// Calculate k1.
MOVQ (CX), R8
ADDQ $8, CX
IMULQ R14, R8
ROLQ $31, R8
IMULQ R13, R8
XORQ R8, AX
ROLQ $27, AX
IMULQ R13, AX
ADDQ R15, AX
CMPQ CX, BX
JLE wordLoop
fourByte:
ADDQ $4, BX
CMPQ CX, BX
JG singles
MOVL (CX), R8
ADDQ $4, CX
IMULQ R13, R8
XORQ R8, AX
ROLQ $23, AX
IMULQ R14, AX
ADDQ ·prime3v(SB), AX
singles:
ADDQ $4, BX
CMPQ CX, BX
JGE finalize
singlesLoop:
MOVBQZX (CX), R12
ADDQ $1, CX
IMULQ ·prime5v(SB), R12
XORQ R12, AX
ROLQ $11, AX
IMULQ R13, AX
CMPQ CX, BX
JL singlesLoop
finalize:
MOVQ AX, R12
SHRQ $33, R12
XORQ R12, AX
IMULQ R14, AX
MOVQ AX, R12
SHRQ $29, R12
XORQ R12, AX
IMULQ ·prime3v(SB), AX
MOVQ AX, R12
SHRQ $32, R12
XORQ R12, AX
MOVQ AX, ret+24(FP)
RET
// writeBlocks uses the same registers as above except that it uses AX to store
// the d pointer.
// func writeBlocks(d *Digest, b []byte) int
TEXT ·writeBlocks(SB), NOSPLIT, $0-40
// Load fixed primes needed for round.
MOVQ ·prime1v(SB), R13
MOVQ ·prime2v(SB), R14
// Load slice.
MOVQ b_base+8(FP), CX
MOVQ b_len+16(FP), DX
LEAQ (CX)(DX*1), BX
SUBQ $32, BX
// Load vN from d.
MOVQ d+0(FP), AX
MOVQ 0(AX), R8 // v1
MOVQ 8(AX), R9 // v2
MOVQ 16(AX), R10 // v3
MOVQ 24(AX), R11 // v4
// We don't need to check the loop condition here; this function is
// always called with at least one block of data to process.
blockLoop:
round(R8)
round(R9)
round(R10)
round(R11)
CMPQ CX, BX
JLE blockLoop
// Copy vN back to d.
MOVQ R8, 0(AX)
MOVQ R9, 8(AX)
MOVQ R10, 16(AX)
MOVQ R11, 24(AX)
// The number of bytes written is CX minus the old base pointer.
SUBQ b_base+8(FP), CX
MOVQ CX, ret+32(FP)
RET

View file

@ -1,76 +0,0 @@
// +build !amd64 appengine !gc purego
package xxhash
// Sum64 computes the 64-bit xxHash digest of b.
func Sum64(b []byte) uint64 {
// A simpler version would be
// d := New()
// d.Write(b)
// return d.Sum64()
// but this is faster, particularly for small inputs.
n := len(b)
var h uint64
if n >= 32 {
v1 := prime1v + prime2
v2 := prime2
v3 := uint64(0)
v4 := -prime1v
for len(b) >= 32 {
v1 = round(v1, u64(b[0:8:len(b)]))
v2 = round(v2, u64(b[8:16:len(b)]))
v3 = round(v3, u64(b[16:24:len(b)]))
v4 = round(v4, u64(b[24:32:len(b)]))
b = b[32:len(b):len(b)]
}
h = rol1(v1) + rol7(v2) + rol12(v3) + rol18(v4)
h = mergeRound(h, v1)
h = mergeRound(h, v2)
h = mergeRound(h, v3)
h = mergeRound(h, v4)
} else {
h = prime5
}
h += uint64(n)
i, end := 0, len(b)
for ; i+8 <= end; i += 8 {
k1 := round(0, u64(b[i:i+8:len(b)]))
h ^= k1
h = rol27(h)*prime1 + prime4
}
if i+4 <= end {
h ^= uint64(u32(b[i:i+4:len(b)])) * prime1
h = rol23(h)*prime2 + prime3
i += 4
}
for ; i < end; i++ {
h ^= uint64(b[i]) * prime5
h = rol11(h) * prime1
}
h ^= h >> 33
h *= prime2
h ^= h >> 29
h *= prime3
h ^= h >> 32
return h
}
func writeBlocks(d *Digest, b []byte) int {
v1, v2, v3, v4 := d.v1, d.v2, d.v3, d.v4
n := len(b)
for len(b) >= 32 {
v1 = round(v1, u64(b[0:8:len(b)]))
v2 = round(v2, u64(b[8:16:len(b)]))
v3 = round(v3, u64(b[16:24:len(b)]))
v4 = round(v4, u64(b[24:32:len(b)]))
b = b[32:len(b):len(b)]
}
d.v1, d.v2, d.v3, d.v4 = v1, v2, v3, v4
return n - len(b)
}

View file

@ -1,15 +0,0 @@
// +build appengine
// This file contains the safe implementations of otherwise unsafe-using code.
package xxhash
// Sum64String computes the 64-bit xxHash digest of s.
func Sum64String(s string) uint64 {
return Sum64([]byte(s))
}
// WriteString adds more data to d. It always returns len(s), nil.
func (d *Digest) WriteString(s string) (n int, err error) {
return d.Write([]byte(s))
}

View file

@ -1,46 +0,0 @@
// +build !appengine
// This file encapsulates usage of unsafe.
// xxhash_safe.go contains the safe implementations.
package xxhash
import (
"reflect"
"unsafe"
)
// Notes:
//
// See https://groups.google.com/d/msg/golang-nuts/dcjzJy-bSpw/tcZYBzQqAQAJ
// for some discussion about these unsafe conversions.
//
// In the future it's possible that compiler optimizations will make these
// unsafe operations unnecessary: https://golang.org/issue/2205.
//
// Both of these wrapper functions still incur function call overhead since they
// will not be inlined. We could write Go/asm copies of Sum64 and Digest.Write
// for strings to squeeze out a bit more speed. Mid-stack inlining should
// eventually fix this.
// Sum64String computes the 64-bit xxHash digest of s.
// It may be faster than Sum64([]byte(s)) by avoiding a copy.
func Sum64String(s string) uint64 {
var b []byte
bh := (*reflect.SliceHeader)(unsafe.Pointer(&b))
bh.Data = (*reflect.StringHeader)(unsafe.Pointer(&s)).Data
bh.Len = len(s)
bh.Cap = len(s)
return Sum64(b)
}
// WriteString adds more data to d. It always returns len(s), nil.
// It may be faster than Write([]byte(s)) by avoiding a copy.
func (d *Digest) WriteString(s string) (n int, err error) {
var b []byte
bh := (*reflect.SliceHeader)(unsafe.Pointer(&b))
bh.Data = (*reflect.StringHeader)(unsafe.Pointer(&s)).Data
bh.Len = len(s)
bh.Cap = len(s)
return d.Write(b)
}

View file

@ -1,16 +0,0 @@
Aaron Lehmann <aaron.lehmann@docker.com>
Akash Gupta <akagup@microsoft.com>
Akihiro Suda <suda.akihiro@lab.ntt.co.jp>
Andrew Pennebaker <apennebaker@datapipe.com>
Brandon Philips <brandon.philips@coreos.com>
Christopher Jones <tophj@linux.vnet.ibm.com>
Daniel, Dao Quang Minh <dqminh89@gmail.com>
Derek McGowan <derek@mcgstyle.net>
Edward Pilatowicz <edward.pilatowicz@oracle.com>
Ian Campbell <ijc@docker.com>
Justin Cormack <justin.cormack@docker.com>
Justin Cummins <sul3n3t@gmail.com>
Phil Estes <estesp@gmail.com>
Stephen J Day <stephen.day@docker.com>
Tobias Klauser <tklauser@distanz.ch>
Tonis Tiigi <tonistiigi@gmail.com>

View file

@ -1,191 +0,0 @@
Apache License
Version 2.0, January 2004
https://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
Copyright The containerd Authors
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
https://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

View file

@ -1,172 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"io/ioutil"
"os"
"path/filepath"
"sync"
"github.com/pkg/errors"
)
var bufferPool = &sync.Pool{
New: func() interface{} {
buffer := make([]byte, 32*1024)
return &buffer
},
}
// XAttrErrorHandlers transform a non-nil xattr error.
// Return nil to ignore an error.
// xattrKey can be empty for listxattr operation.
type XAttrErrorHandler func(dst, src, xattrKey string, err error) error
type copyDirOpts struct {
xeh XAttrErrorHandler
}
type CopyDirOpt func(*copyDirOpts) error
// WithXAttrErrorHandler allows specifying XAttrErrorHandler
// If nil XAttrErrorHandler is specified (default), CopyDir stops
// on a non-nil xattr error.
func WithXAttrErrorHandler(xeh XAttrErrorHandler) CopyDirOpt {
return func(o *copyDirOpts) error {
o.xeh = xeh
return nil
}
}
// WithAllowXAttrErrors allows ignoring xattr errors.
func WithAllowXAttrErrors() CopyDirOpt {
xeh := func(dst, src, xattrKey string, err error) error {
return nil
}
return WithXAttrErrorHandler(xeh)
}
// CopyDir copies the directory from src to dst.
// Most efficient copy of files is attempted.
func CopyDir(dst, src string, opts ...CopyDirOpt) error {
var o copyDirOpts
for _, opt := range opts {
if err := opt(&o); err != nil {
return err
}
}
inodes := map[uint64]string{}
return copyDirectory(dst, src, inodes, &o)
}
func copyDirectory(dst, src string, inodes map[uint64]string, o *copyDirOpts) error {
stat, err := os.Stat(src)
if err != nil {
return errors.Wrapf(err, "failed to stat %s", src)
}
if !stat.IsDir() {
return errors.Errorf("source is not directory")
}
if st, err := os.Stat(dst); err != nil {
if err := os.Mkdir(dst, stat.Mode()); err != nil {
return errors.Wrapf(err, "failed to mkdir %s", dst)
}
} else if !st.IsDir() {
return errors.Errorf("cannot copy to non-directory: %s", dst)
} else {
if err := os.Chmod(dst, stat.Mode()); err != nil {
return errors.Wrapf(err, "failed to chmod on %s", dst)
}
}
fis, err := ioutil.ReadDir(src)
if err != nil {
return errors.Wrapf(err, "failed to read %s", src)
}
if err := copyFileInfo(stat, dst); err != nil {
return errors.Wrapf(err, "failed to copy file info for %s", dst)
}
for _, fi := range fis {
source := filepath.Join(src, fi.Name())
target := filepath.Join(dst, fi.Name())
switch {
case fi.IsDir():
if err := copyDirectory(target, source, inodes, o); err != nil {
return err
}
continue
case (fi.Mode() & os.ModeType) == 0:
link, err := getLinkSource(target, fi, inodes)
if err != nil {
return errors.Wrap(err, "failed to get hardlink")
}
if link != "" {
if err := os.Link(link, target); err != nil {
return errors.Wrap(err, "failed to create hard link")
}
} else if err := CopyFile(target, source); err != nil {
return errors.Wrap(err, "failed to copy files")
}
case (fi.Mode() & os.ModeSymlink) == os.ModeSymlink:
link, err := os.Readlink(source)
if err != nil {
return errors.Wrapf(err, "failed to read link: %s", source)
}
if err := os.Symlink(link, target); err != nil {
return errors.Wrapf(err, "failed to create symlink: %s", target)
}
case (fi.Mode() & os.ModeDevice) == os.ModeDevice:
if err := copyDevice(target, fi); err != nil {
return errors.Wrapf(err, "failed to create device")
}
default:
// TODO: Support pipes and sockets
return errors.Wrapf(err, "unsupported mode %s", fi.Mode())
}
if err := copyFileInfo(fi, target); err != nil {
return errors.Wrap(err, "failed to copy file info")
}
if err := copyXAttrs(target, source, o.xeh); err != nil {
return errors.Wrap(err, "failed to copy xattrs")
}
}
return nil
}
// CopyFile copies the source file to the target.
// The most efficient means of copying is used for the platform.
func CopyFile(target, source string) error {
src, err := os.Open(source)
if err != nil {
return errors.Wrapf(err, "failed to open source %s", source)
}
defer src.Close()
tgt, err := os.Create(target)
if err != nil {
return errors.Wrapf(err, "failed to open target %s", target)
}
defer tgt.Close()
return copyFileContent(tgt, src)
}

View file

@ -1,144 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"io"
"os"
"syscall"
"github.com/containerd/continuity/sysx"
"github.com/pkg/errors"
"golang.org/x/sys/unix"
)
func copyFileInfo(fi os.FileInfo, name string) error {
st := fi.Sys().(*syscall.Stat_t)
if err := os.Lchown(name, int(st.Uid), int(st.Gid)); err != nil {
if os.IsPermission(err) {
// Normally if uid/gid are the same this would be a no-op, but some
// filesystems may still return EPERM... for instance NFS does this.
// In such a case, this is not an error.
if dstStat, err2 := os.Lstat(name); err2 == nil {
st2 := dstStat.Sys().(*syscall.Stat_t)
if st.Uid == st2.Uid && st.Gid == st2.Gid {
err = nil
}
}
}
if err != nil {
return errors.Wrapf(err, "failed to chown %s", name)
}
}
if (fi.Mode() & os.ModeSymlink) != os.ModeSymlink {
if err := os.Chmod(name, fi.Mode()); err != nil {
return errors.Wrapf(err, "failed to chmod %s", name)
}
}
timespec := []unix.Timespec{unix.Timespec(StatAtime(st)), unix.Timespec(StatMtime(st))}
if err := unix.UtimesNanoAt(unix.AT_FDCWD, name, timespec, unix.AT_SYMLINK_NOFOLLOW); err != nil {
return errors.Wrapf(err, "failed to utime %s", name)
}
return nil
}
const maxSSizeT = int64(^uint(0) >> 1)
func copyFileContent(dst, src *os.File) error {
st, err := src.Stat()
if err != nil {
return errors.Wrap(err, "unable to stat source")
}
size := st.Size()
first := true
srcFd := int(src.Fd())
dstFd := int(dst.Fd())
for size > 0 {
// Ensure that we are never trying to copy more than SSIZE_MAX at a
// time and at the same time avoids overflows when the file is larger
// than 4GB on 32-bit systems.
var copySize int
if size > maxSSizeT {
copySize = int(maxSSizeT)
} else {
copySize = int(size)
}
n, err := unix.CopyFileRange(srcFd, nil, dstFd, nil, copySize, 0)
if err != nil {
if (err != unix.ENOSYS && err != unix.EXDEV) || !first {
return errors.Wrap(err, "copy file range failed")
}
buf := bufferPool.Get().(*[]byte)
_, err = io.CopyBuffer(dst, src, *buf)
bufferPool.Put(buf)
return errors.Wrap(err, "userspace copy failed")
}
first = false
size -= int64(n)
}
return nil
}
func copyXAttrs(dst, src string, xeh XAttrErrorHandler) error {
xattrKeys, err := sysx.LListxattr(src)
if err != nil {
e := errors.Wrapf(err, "failed to list xattrs on %s", src)
if xeh != nil {
e = xeh(dst, src, "", e)
}
return e
}
for _, xattr := range xattrKeys {
data, err := sysx.LGetxattr(src, xattr)
if err != nil {
e := errors.Wrapf(err, "failed to get xattr %q on %s", xattr, src)
if xeh != nil {
if e = xeh(dst, src, xattr, e); e == nil {
continue
}
}
return e
}
if err := sysx.LSetxattr(dst, xattr, data, 0); err != nil {
e := errors.Wrapf(err, "failed to set xattr %q on %s", xattr, dst)
if xeh != nil {
if e = xeh(dst, src, xattr, e); e == nil {
continue
}
}
return e
}
}
return nil
}
func copyDevice(dst string, fi os.FileInfo) error {
st, ok := fi.Sys().(*syscall.Stat_t)
if !ok {
return errors.New("unsupported stat type")
}
return unix.Mknod(dst, uint32(fi.Mode()), int(st.Rdev))
}

View file

@ -1,112 +0,0 @@
// +build solaris darwin freebsd
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"io"
"os"
"syscall"
"github.com/containerd/continuity/sysx"
"github.com/pkg/errors"
"golang.org/x/sys/unix"
)
func copyFileInfo(fi os.FileInfo, name string) error {
st := fi.Sys().(*syscall.Stat_t)
if err := os.Lchown(name, int(st.Uid), int(st.Gid)); err != nil {
if os.IsPermission(err) {
// Normally if uid/gid are the same this would be a no-op, but some
// filesystems may still return EPERM... for instance NFS does this.
// In such a case, this is not an error.
if dstStat, err2 := os.Lstat(name); err2 == nil {
st2 := dstStat.Sys().(*syscall.Stat_t)
if st.Uid == st2.Uid && st.Gid == st2.Gid {
err = nil
}
}
}
if err != nil {
return errors.Wrapf(err, "failed to chown %s", name)
}
}
if (fi.Mode() & os.ModeSymlink) != os.ModeSymlink {
if err := os.Chmod(name, fi.Mode()); err != nil {
return errors.Wrapf(err, "failed to chmod %s", name)
}
}
timespec := []syscall.Timespec{StatAtime(st), StatMtime(st)}
if err := syscall.UtimesNano(name, timespec); err != nil {
return errors.Wrapf(err, "failed to utime %s", name)
}
return nil
}
func copyFileContent(dst, src *os.File) error {
buf := bufferPool.Get().(*[]byte)
_, err := io.CopyBuffer(dst, src, *buf)
bufferPool.Put(buf)
return err
}
func copyXAttrs(dst, src string, xeh XAttrErrorHandler) error {
xattrKeys, err := sysx.LListxattr(src)
if err != nil {
e := errors.Wrapf(err, "failed to list xattrs on %s", src)
if xeh != nil {
e = xeh(dst, src, "", e)
}
return e
}
for _, xattr := range xattrKeys {
data, err := sysx.LGetxattr(src, xattr)
if err != nil {
e := errors.Wrapf(err, "failed to get xattr %q on %s", xattr, src)
if xeh != nil {
if e = xeh(dst, src, xattr, e); e == nil {
continue
}
}
return e
}
if err := sysx.LSetxattr(dst, xattr, data, 0); err != nil {
e := errors.Wrapf(err, "failed to set xattr %q on %s", xattr, dst)
if xeh != nil {
if e = xeh(dst, src, xattr, e); e == nil {
continue
}
}
return e
}
}
return nil
}
func copyDevice(dst string, fi os.FileInfo) error {
st, ok := fi.Sys().(*syscall.Stat_t)
if !ok {
return errors.New("unsupported stat type")
}
return unix.Mknod(dst, uint32(fi.Mode()), int(st.Rdev))
}

View file

@ -1,49 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"io"
"os"
"github.com/pkg/errors"
)
func copyFileInfo(fi os.FileInfo, name string) error {
if err := os.Chmod(name, fi.Mode()); err != nil {
return errors.Wrapf(err, "failed to chmod %s", name)
}
// TODO: copy windows specific metadata
return nil
}
func copyFileContent(dst, src *os.File) error {
buf := bufferPool.Get().(*[]byte)
_, err := io.CopyBuffer(dst, src, *buf)
bufferPool.Put(buf)
return err
}
func copyXAttrs(dst, src string, xeh XAttrErrorHandler) error {
return nil
}
func copyDevice(dst string, fi os.FileInfo) error {
return errors.New("device copy not supported")
}

View file

@ -1,326 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"context"
"os"
"path/filepath"
"strings"
"golang.org/x/sync/errgroup"
"github.com/sirupsen/logrus"
)
// ChangeKind is the type of modification that
// a change is making.
type ChangeKind int
const (
// ChangeKindUnmodified represents an unmodified
// file
ChangeKindUnmodified = iota
// ChangeKindAdd represents an addition of
// a file
ChangeKindAdd
// ChangeKindModify represents a change to
// an existing file
ChangeKindModify
// ChangeKindDelete represents a delete of
// a file
ChangeKindDelete
)
func (k ChangeKind) String() string {
switch k {
case ChangeKindUnmodified:
return "unmodified"
case ChangeKindAdd:
return "add"
case ChangeKindModify:
return "modify"
case ChangeKindDelete:
return "delete"
default:
return ""
}
}
// Change represents single change between a diff and its parent.
type Change struct {
Kind ChangeKind
Path string
}
// ChangeFunc is the type of function called for each change
// computed during a directory changes calculation.
type ChangeFunc func(ChangeKind, string, os.FileInfo, error) error
// Changes computes changes between two directories calling the
// given change function for each computed change. The first
// directory is intended to the base directory and second
// directory the changed directory.
//
// The change callback is called by the order of path names and
// should be appliable in that order.
// Due to this apply ordering, the following is true
// - Removed directory trees only create a single change for the root
// directory removed. Remaining changes are implied.
// - A directory which is modified to become a file will not have
// delete entries for sub-path items, their removal is implied
// by the removal of the parent directory.
//
// Opaque directories will not be treated specially and each file
// removed from the base directory will show up as a removal.
//
// File content comparisons will be done on files which have timestamps
// which may have been truncated. If either of the files being compared
// has a zero value nanosecond value, each byte will be compared for
// differences. If 2 files have the same seconds value but different
// nanosecond values where one of those values is zero, the files will
// be considered unchanged if the content is the same. This behavior
// is to account for timestamp truncation during archiving.
func Changes(ctx context.Context, a, b string, changeFn ChangeFunc) error {
if a == "" {
logrus.Debugf("Using single walk diff for %s", b)
return addDirChanges(ctx, changeFn, b)
} else if diffOptions := detectDirDiff(b, a); diffOptions != nil {
logrus.Debugf("Using single walk diff for %s from %s", diffOptions.diffDir, a)
return diffDirChanges(ctx, changeFn, a, diffOptions)
}
logrus.Debugf("Using double walk diff for %s from %s", b, a)
return doubleWalkDiff(ctx, changeFn, a, b)
}
func addDirChanges(ctx context.Context, changeFn ChangeFunc, root string) error {
return filepath.Walk(root, func(path string, f os.FileInfo, err error) error {
if err != nil {
return err
}
// Rebase path
path, err = filepath.Rel(root, path)
if err != nil {
return err
}
path = filepath.Join(string(os.PathSeparator), path)
// Skip root
if path == string(os.PathSeparator) {
return nil
}
return changeFn(ChangeKindAdd, path, f, nil)
})
}
// diffDirOptions is used when the diff can be directly calculated from
// a diff directory to its base, without walking both trees.
type diffDirOptions struct {
diffDir string
skipChange func(string) (bool, error)
deleteChange func(string, string, os.FileInfo) (string, error)
}
// diffDirChanges walks the diff directory and compares changes against the base.
func diffDirChanges(ctx context.Context, changeFn ChangeFunc, base string, o *diffDirOptions) error {
changedDirs := make(map[string]struct{})
return filepath.Walk(o.diffDir, func(path string, f os.FileInfo, err error) error {
if err != nil {
return err
}
// Rebase path
path, err = filepath.Rel(o.diffDir, path)
if err != nil {
return err
}
path = filepath.Join(string(os.PathSeparator), path)
// Skip root
if path == string(os.PathSeparator) {
return nil
}
// TODO: handle opaqueness, start new double walker at this
// location to get deletes, and skip tree in single walker
if o.skipChange != nil {
if skip, err := o.skipChange(path); skip {
return err
}
}
var kind ChangeKind
deletedFile, err := o.deleteChange(o.diffDir, path, f)
if err != nil {
return err
}
// Find out what kind of modification happened
if deletedFile != "" {
path = deletedFile
kind = ChangeKindDelete
f = nil
} else {
// Otherwise, the file was added
kind = ChangeKindAdd
// ...Unless it already existed in a base, in which case, it's a modification
stat, err := os.Stat(filepath.Join(base, path))
if err != nil && !os.IsNotExist(err) {
return err
}
if err == nil {
// The file existed in the base, so that's a modification
// However, if it's a directory, maybe it wasn't actually modified.
// If you modify /foo/bar/baz, then /foo will be part of the changed files only because it's the parent of bar
if stat.IsDir() && f.IsDir() {
if f.Size() == stat.Size() && f.Mode() == stat.Mode() && sameFsTime(f.ModTime(), stat.ModTime()) {
// Both directories are the same, don't record the change
return nil
}
}
kind = ChangeKindModify
}
}
// If /foo/bar/file.txt is modified, then /foo/bar must be part of the changed files.
// This block is here to ensure the change is recorded even if the
// modify time, mode and size of the parent directory in the rw and ro layers are all equal.
// Check https://github.com/docker/docker/pull/13590 for details.
if f.IsDir() {
changedDirs[path] = struct{}{}
}
if kind == ChangeKindAdd || kind == ChangeKindDelete {
parent := filepath.Dir(path)
if _, ok := changedDirs[parent]; !ok && parent != "/" {
pi, err := os.Stat(filepath.Join(o.diffDir, parent))
if err := changeFn(ChangeKindModify, parent, pi, err); err != nil {
return err
}
changedDirs[parent] = struct{}{}
}
}
return changeFn(kind, path, f, nil)
})
}
// doubleWalkDiff walks both directories to create a diff
func doubleWalkDiff(ctx context.Context, changeFn ChangeFunc, a, b string) (err error) {
g, ctx := errgroup.WithContext(ctx)
var (
c1 = make(chan *currentPath)
c2 = make(chan *currentPath)
f1, f2 *currentPath
rmdir string
)
g.Go(func() error {
defer close(c1)
return pathWalk(ctx, a, c1)
})
g.Go(func() error {
defer close(c2)
return pathWalk(ctx, b, c2)
})
g.Go(func() error {
for c1 != nil || c2 != nil {
if f1 == nil && c1 != nil {
f1, err = nextPath(ctx, c1)
if err != nil {
return err
}
if f1 == nil {
c1 = nil
}
}
if f2 == nil && c2 != nil {
f2, err = nextPath(ctx, c2)
if err != nil {
return err
}
if f2 == nil {
c2 = nil
}
}
if f1 == nil && f2 == nil {
continue
}
var f os.FileInfo
k, p := pathChange(f1, f2)
switch k {
case ChangeKindAdd:
if rmdir != "" {
rmdir = ""
}
f = f2.f
f2 = nil
case ChangeKindDelete:
// Check if this file is already removed by being
// under of a removed directory
if rmdir != "" && strings.HasPrefix(f1.path, rmdir) {
f1 = nil
continue
} else if f1.f.IsDir() {
rmdir = f1.path + string(os.PathSeparator)
} else if rmdir != "" {
rmdir = ""
}
f1 = nil
case ChangeKindModify:
same, err := sameFile(f1, f2)
if err != nil {
return err
}
if f1.f.IsDir() && !f2.f.IsDir() {
rmdir = f1.path + string(os.PathSeparator)
} else if rmdir != "" {
rmdir = ""
}
f = f2.f
f1 = nil
f2 = nil
if same {
if !isLinked(f) {
continue
}
k = ChangeKindUnmodified
}
}
if err := changeFn(k, p, f, nil); err != nil {
return err
}
}
return nil
})
return g.Wait()
}

View file

@ -1,74 +0,0 @@
// +build !windows
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"bytes"
"os"
"syscall"
"github.com/containerd/continuity/sysx"
"github.com/pkg/errors"
)
// detectDirDiff returns diff dir options if a directory could
// be found in the mount info for upper which is the direct
// diff with the provided lower directory
func detectDirDiff(upper, lower string) *diffDirOptions {
// TODO: get mount options for upper
// TODO: detect AUFS
// TODO: detect overlay
return nil
}
// compareSysStat returns whether the stats are equivalent,
// whether the files are considered the same file, and
// an error
func compareSysStat(s1, s2 interface{}) (bool, error) {
ls1, ok := s1.(*syscall.Stat_t)
if !ok {
return false, nil
}
ls2, ok := s2.(*syscall.Stat_t)
if !ok {
return false, nil
}
return ls1.Mode == ls2.Mode && ls1.Uid == ls2.Uid && ls1.Gid == ls2.Gid && ls1.Rdev == ls2.Rdev, nil
}
func compareCapabilities(p1, p2 string) (bool, error) {
c1, err := sysx.LGetxattr(p1, "security.capability")
if err != nil && err != sysx.ENODATA {
return false, errors.Wrapf(err, "failed to get xattr for %s", p1)
}
c2, err := sysx.LGetxattr(p2, "security.capability")
if err != nil && err != sysx.ENODATA {
return false, errors.Wrapf(err, "failed to get xattr for %s", p2)
}
return bytes.Equal(c1, c2), nil
}
func isLinked(f os.FileInfo) bool {
s, ok := f.Sys().(*syscall.Stat_t)
if !ok {
return false
}
return !f.IsDir() && s.Nlink > 1
}

View file

@ -1,48 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"os"
"golang.org/x/sys/windows"
)
func detectDirDiff(upper, lower string) *diffDirOptions {
return nil
}
func compareSysStat(s1, s2 interface{}) (bool, error) {
f1, ok := s1.(windows.Win32FileAttributeData)
if !ok {
return false, nil
}
f2, ok := s2.(windows.Win32FileAttributeData)
if !ok {
return false, nil
}
return f1.FileAttributes == f2.FileAttributes, nil
}
func compareCapabilities(p1, p2 string) (bool, error) {
// TODO: Use windows equivalent
return true, nil
}
func isLinked(os.FileInfo) bool {
return false
}

View file

@ -1,103 +0,0 @@
// +build linux
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"fmt"
"io/ioutil"
"os"
"syscall"
"unsafe"
)
func locateDummyIfEmpty(path string) (string, error) {
children, err := ioutil.ReadDir(path)
if err != nil {
return "", err
}
if len(children) != 0 {
return "", nil
}
dummyFile, err := ioutil.TempFile(path, "fsutils-dummy")
if err != nil {
return "", err
}
name := dummyFile.Name()
err = dummyFile.Close()
return name, err
}
// SupportsDType returns whether the filesystem mounted on path supports d_type
func SupportsDType(path string) (bool, error) {
// locate dummy so that we have at least one dirent
dummy, err := locateDummyIfEmpty(path)
if err != nil {
return false, err
}
if dummy != "" {
defer os.Remove(dummy)
}
visited := 0
supportsDType := true
fn := func(ent *syscall.Dirent) bool {
visited++
if ent.Type == syscall.DT_UNKNOWN {
supportsDType = false
// stop iteration
return true
}
// continue iteration
return false
}
if err = iterateReadDir(path, fn); err != nil {
return false, err
}
if visited == 0 {
return false, fmt.Errorf("did not hit any dirent during iteration %s", path)
}
return supportsDType, nil
}
func iterateReadDir(path string, fn func(*syscall.Dirent) bool) error {
d, err := os.Open(path)
if err != nil {
return err
}
defer d.Close()
fd := int(d.Fd())
buf := make([]byte, 4096)
for {
nbytes, err := syscall.ReadDirent(fd, buf)
if err != nil {
return err
}
if nbytes == 0 {
break
}
for off := 0; off < nbytes; {
ent := (*syscall.Dirent)(unsafe.Pointer(&buf[off]))
if stop := fn(ent); stop {
return nil
}
off += int(ent.Reclen)
}
}
return nil
}

View file

@ -1,38 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import "context"
// Usage of disk information
type Usage struct {
Inodes int64
Size int64
}
// DiskUsage counts the number of inodes and disk usage for the resources under
// path.
func DiskUsage(ctx context.Context, roots ...string) (Usage, error) {
return diskUsage(ctx, roots...)
}
// DiffUsage counts the numbers of inodes and disk usage in the
// diff between the 2 directories. The first path is intended
// as the base directory and the second as the changed directory.
func DiffUsage(ctx context.Context, a, b string) (Usage, error) {
return diffUsage(ctx, a, b)
}

View file

@ -1,110 +0,0 @@
// +build !windows
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"context"
"os"
"path/filepath"
"syscall"
)
type inode struct {
// TODO(stevvooe): Can probably reduce memory usage by not tracking
// device, but we can leave this right for now.
dev, ino uint64
}
func newInode(stat *syscall.Stat_t) inode {
return inode{
// Dev is uint32 on darwin/bsd, uint64 on linux/solaris
dev: uint64(stat.Dev), // nolint: unconvert
// Ino is uint32 on bsd, uint64 on darwin/linux/solaris
ino: uint64(stat.Ino), // nolint: unconvert
}
}
func diskUsage(ctx context.Context, roots ...string) (Usage, error) {
var (
size int64
inodes = map[inode]struct{}{} // expensive!
)
for _, root := range roots {
if err := filepath.Walk(root, func(path string, fi os.FileInfo, err error) error {
if err != nil {
return err
}
select {
case <-ctx.Done():
return ctx.Err()
default:
}
inoKey := newInode(fi.Sys().(*syscall.Stat_t))
if _, ok := inodes[inoKey]; !ok {
inodes[inoKey] = struct{}{}
size += fi.Size()
}
return nil
}); err != nil {
return Usage{}, err
}
}
return Usage{
Inodes: int64(len(inodes)),
Size: size,
}, nil
}
func diffUsage(ctx context.Context, a, b string) (Usage, error) {
var (
size int64
inodes = map[inode]struct{}{} // expensive!
)
if err := Changes(ctx, a, b, func(kind ChangeKind, _ string, fi os.FileInfo, err error) error {
if err != nil {
return err
}
if kind == ChangeKindAdd || kind == ChangeKindModify {
inoKey := newInode(fi.Sys().(*syscall.Stat_t))
if _, ok := inodes[inoKey]; !ok {
inodes[inoKey] = struct{}{}
size += fi.Size()
}
return nil
}
return nil
}); err != nil {
return Usage{}, err
}
return Usage{
Inodes: int64(len(inodes)),
Size: size,
}, nil
}

View file

@ -1,82 +0,0 @@
// +build windows
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"context"
"os"
"path/filepath"
)
func diskUsage(ctx context.Context, roots ...string) (Usage, error) {
var (
size int64
)
// TODO(stevvooe): Support inodes (or equivalent) for windows.
for _, root := range roots {
if err := filepath.Walk(root, func(path string, fi os.FileInfo, err error) error {
if err != nil {
return err
}
select {
case <-ctx.Done():
return ctx.Err()
default:
}
size += fi.Size()
return nil
}); err != nil {
return Usage{}, err
}
}
return Usage{
Size: size,
}, nil
}
func diffUsage(ctx context.Context, a, b string) (Usage, error) {
var (
size int64
)
if err := Changes(ctx, a, b, func(kind ChangeKind, _ string, fi os.FileInfo, err error) error {
if err != nil {
return err
}
if kind == ChangeKindAdd || kind == ChangeKindModify {
size += fi.Size()
return nil
}
return nil
}); err != nil {
return Usage{}, err
}
return Usage{
Size: size,
}, nil
}

View file

@ -1,43 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import "os"
// GetLinkInfo returns an identifier representing the node a hardlink is pointing
// to. If the file is not hard linked then 0 will be returned.
func GetLinkInfo(fi os.FileInfo) (uint64, bool) {
return getLinkInfo(fi)
}
// getLinkSource returns a path for the given name and
// file info to its link source in the provided inode
// map. If the given file name is not in the map and
// has other links, it is added to the inode map
// to be a source for other link locations.
func getLinkSource(name string, fi os.FileInfo, inodes map[uint64]string) (string, error) {
inode, isHardlink := getLinkInfo(fi)
if !isHardlink {
return "", nil
}
path, ok := inodes[inode]
if !ok {
inodes[inode] = name
}
return path, nil
}

View file

@ -1,34 +0,0 @@
// +build !windows
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"os"
"syscall"
)
func getLinkInfo(fi os.FileInfo) (uint64, bool) {
s, ok := fi.Sys().(*syscall.Stat_t)
if !ok {
return 0, false
}
// Ino is uint32 on bsd, uint64 on darwin/linux/solaris
return uint64(s.Ino), !fi.IsDir() && s.Nlink > 1 // nolint: unconvert
}

View file

@ -1,23 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import "os"
func getLinkInfo(fi os.FileInfo) (uint64, bool) {
return 0, false
}

View file

@ -1,313 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"bytes"
"context"
"io"
"os"
"path/filepath"
"github.com/pkg/errors"
)
var (
errTooManyLinks = errors.New("too many links")
)
type currentPath struct {
path string
f os.FileInfo
fullPath string
}
func pathChange(lower, upper *currentPath) (ChangeKind, string) {
if lower == nil {
if upper == nil {
panic("cannot compare nil paths")
}
return ChangeKindAdd, upper.path
}
if upper == nil {
return ChangeKindDelete, lower.path
}
switch i := directoryCompare(lower.path, upper.path); {
case i < 0:
// File in lower that is not in upper
return ChangeKindDelete, lower.path
case i > 0:
// File in upper that is not in lower
return ChangeKindAdd, upper.path
default:
return ChangeKindModify, upper.path
}
}
func directoryCompare(a, b string) int {
l := len(a)
if len(b) < l {
l = len(b)
}
for i := 0; i < l; i++ {
c1, c2 := a[i], b[i]
if c1 == filepath.Separator {
c1 = byte(0)
}
if c2 == filepath.Separator {
c2 = byte(0)
}
if c1 < c2 {
return -1
}
if c1 > c2 {
return +1
}
}
if len(a) < len(b) {
return -1
}
if len(a) > len(b) {
return +1
}
return 0
}
func sameFile(f1, f2 *currentPath) (bool, error) {
if os.SameFile(f1.f, f2.f) {
return true, nil
}
equalStat, err := compareSysStat(f1.f.Sys(), f2.f.Sys())
if err != nil || !equalStat {
return equalStat, err
}
if eq, err := compareCapabilities(f1.fullPath, f2.fullPath); err != nil || !eq {
return eq, err
}
// If not a directory also check size, modtime, and content
if !f1.f.IsDir() {
if f1.f.Size() != f2.f.Size() {
return false, nil
}
t1 := f1.f.ModTime()
t2 := f2.f.ModTime()
if t1.Unix() != t2.Unix() {
return false, nil
}
// If the timestamp may have been truncated in both of the
// files, check content of file to determine difference
if t1.Nanosecond() == 0 && t2.Nanosecond() == 0 {
var eq bool
if (f1.f.Mode() & os.ModeSymlink) == os.ModeSymlink {
eq, err = compareSymlinkTarget(f1.fullPath, f2.fullPath)
} else if f1.f.Size() > 0 {
eq, err = compareFileContent(f1.fullPath, f2.fullPath)
}
if err != nil || !eq {
return eq, err
}
} else if t1.Nanosecond() != t2.Nanosecond() {
return false, nil
}
}
return true, nil
}
func compareSymlinkTarget(p1, p2 string) (bool, error) {
t1, err := os.Readlink(p1)
if err != nil {
return false, err
}
t2, err := os.Readlink(p2)
if err != nil {
return false, err
}
return t1 == t2, nil
}
const compareChuckSize = 32 * 1024
// compareFileContent compares the content of 2 same sized files
// by comparing each byte.
func compareFileContent(p1, p2 string) (bool, error) {
f1, err := os.Open(p1)
if err != nil {
return false, err
}
defer f1.Close()
f2, err := os.Open(p2)
if err != nil {
return false, err
}
defer f2.Close()
b1 := make([]byte, compareChuckSize)
b2 := make([]byte, compareChuckSize)
for {
n1, err1 := f1.Read(b1)
if err1 != nil && err1 != io.EOF {
return false, err1
}
n2, err2 := f2.Read(b2)
if err2 != nil && err2 != io.EOF {
return false, err2
}
if n1 != n2 || !bytes.Equal(b1[:n1], b2[:n2]) {
return false, nil
}
if err1 == io.EOF && err2 == io.EOF {
return true, nil
}
}
}
func pathWalk(ctx context.Context, root string, pathC chan<- *currentPath) error {
return filepath.Walk(root, func(path string, f os.FileInfo, err error) error {
if err != nil {
return err
}
// Rebase path
path, err = filepath.Rel(root, path)
if err != nil {
return err
}
path = filepath.Join(string(os.PathSeparator), path)
// Skip root
if path == string(os.PathSeparator) {
return nil
}
p := &currentPath{
path: path,
f: f,
fullPath: filepath.Join(root, path),
}
select {
case <-ctx.Done():
return ctx.Err()
case pathC <- p:
return nil
}
})
}
func nextPath(ctx context.Context, pathC <-chan *currentPath) (*currentPath, error) {
select {
case <-ctx.Done():
return nil, ctx.Err()
case p := <-pathC:
return p, nil
}
}
// RootPath joins a path with a root, evaluating and bounding any
// symlink to the root directory.
func RootPath(root, path string) (string, error) {
if path == "" {
return root, nil
}
var linksWalked int // to protect against cycles
for {
i := linksWalked
newpath, err := walkLinks(root, path, &linksWalked)
if err != nil {
return "", err
}
path = newpath
if i == linksWalked {
newpath = filepath.Join("/", newpath)
if path == newpath {
return filepath.Join(root, newpath), nil
}
path = newpath
}
}
}
func walkLink(root, path string, linksWalked *int) (newpath string, islink bool, err error) {
if *linksWalked > 255 {
return "", false, errTooManyLinks
}
path = filepath.Join("/", path)
if path == "/" {
return path, false, nil
}
realPath := filepath.Join(root, path)
fi, err := os.Lstat(realPath)
if err != nil {
// If path does not yet exist, treat as non-symlink
if os.IsNotExist(err) {
return path, false, nil
}
return "", false, err
}
if fi.Mode()&os.ModeSymlink == 0 {
return path, false, nil
}
newpath, err = os.Readlink(realPath)
if err != nil {
return "", false, err
}
*linksWalked++
return newpath, true, nil
}
func walkLinks(root, path string, linksWalked *int) (string, error) {
switch dir, file := filepath.Split(path); {
case dir == "":
newpath, _, err := walkLink(root, file, linksWalked)
return newpath, err
case file == "":
if os.IsPathSeparator(dir[len(dir)-1]) {
if dir == "/" {
return dir, nil
}
return walkLinks(root, dir[:len(dir)-1], linksWalked)
}
newpath, _, err := walkLink(root, dir, linksWalked)
return newpath, err
default:
newdir, err := walkLinks(root, dir, linksWalked)
if err != nil {
return "", err
}
newpath, islink, err := walkLink(root, filepath.Join(newdir, file), linksWalked)
if err != nil {
return "", err
}
if !islink {
return newpath, nil
}
if filepath.IsAbs(newpath) {
return newpath, nil
}
return filepath.Join(newdir, newpath), nil
}
}

View file

@ -1,44 +0,0 @@
// +build darwin freebsd
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"syscall"
"time"
)
// StatAtime returns the access time from a stat struct
func StatAtime(st *syscall.Stat_t) syscall.Timespec {
return st.Atimespec
}
// StatCtime returns the created time from a stat struct
func StatCtime(st *syscall.Stat_t) syscall.Timespec {
return st.Ctimespec
}
// StatMtime returns the modified time from a stat struct
func StatMtime(st *syscall.Stat_t) syscall.Timespec {
return st.Mtimespec
}
// StatATimeAsTime returns the access time as a time.Time
func StatATimeAsTime(st *syscall.Stat_t) time.Time {
return time.Unix(int64(st.Atimespec.Sec), int64(st.Atimespec.Nsec)) // nolint: unconvert
}

View file

@ -1,43 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import (
"syscall"
"time"
)
// StatAtime returns the Atim
func StatAtime(st *syscall.Stat_t) syscall.Timespec {
return st.Atim
}
// StatCtime returns the Ctim
func StatCtime(st *syscall.Stat_t) syscall.Timespec {
return st.Ctim
}
// StatMtime returns the Mtim
func StatMtime(st *syscall.Stat_t) syscall.Timespec {
return st.Mtim
}
// StatATimeAsTime returns st.Atim as a time.Time
func StatATimeAsTime(st *syscall.Stat_t) time.Time {
// The int64 conversions ensure the line compiles for 32-bit systems as well.
return time.Unix(int64(st.Atim.Sec), int64(st.Atim.Nsec)) // nolint: unconvert
}

View file

@ -1,29 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package fs
import "time"
// Gnu tar and the go tar writer don't have sub-second mtime
// precision, which is problematic when we apply changes via tar
// files, we handle this by comparing for exact times, *or* same
// second count and either a or b having exactly 0 nanoseconds
func sameFsTime(a, b time.Time) bool {
return a == b ||
(a.Unix() == b.Unix() &&
(a.Nanosecond() == 0 || b.Nanosecond() == 0))
}

View file

@ -1,101 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package pathdriver
import (
"path/filepath"
)
// PathDriver provides all of the path manipulation functions in a common
// interface. The context should call these and never use the `filepath`
// package or any other package to manipulate paths.
type PathDriver interface {
Join(paths ...string) string
IsAbs(path string) bool
Rel(base, target string) (string, error)
Base(path string) string
Dir(path string) string
Clean(path string) string
Split(path string) (dir, file string)
Separator() byte
Abs(path string) (string, error)
Walk(string, filepath.WalkFunc) error
FromSlash(path string) string
ToSlash(path string) string
Match(pattern, name string) (matched bool, err error)
}
// pathDriver is a simple default implementation calls the filepath package.
type pathDriver struct{}
// LocalPathDriver is the exported pathDriver struct for convenience.
var LocalPathDriver PathDriver = &pathDriver{}
func (*pathDriver) Join(paths ...string) string {
return filepath.Join(paths...)
}
func (*pathDriver) IsAbs(path string) bool {
return filepath.IsAbs(path)
}
func (*pathDriver) Rel(base, target string) (string, error) {
return filepath.Rel(base, target)
}
func (*pathDriver) Base(path string) string {
return filepath.Base(path)
}
func (*pathDriver) Dir(path string) string {
return filepath.Dir(path)
}
func (*pathDriver) Clean(path string) string {
return filepath.Clean(path)
}
func (*pathDriver) Split(path string) (dir, file string) {
return filepath.Split(path)
}
func (*pathDriver) Separator() byte {
return filepath.Separator
}
func (*pathDriver) Abs(path string) (string, error) {
return filepath.Abs(path)
}
// Note that filepath.Walk calls os.Stat, so if the context wants to
// to call Driver.Stat() for Walk, they need to create a new struct that
// overrides this method.
func (*pathDriver) Walk(root string, walkFn filepath.WalkFunc) error {
return filepath.Walk(root, walkFn)
}
func (*pathDriver) FromSlash(path string) string {
return filepath.FromSlash(path)
}
func (*pathDriver) ToSlash(path string) string {
return filepath.ToSlash(path)
}
func (*pathDriver) Match(pattern, name string) (bool, error) {
return filepath.Match(pattern, name)
}

View file

@ -1,26 +0,0 @@
// +build !windows
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package syscallx
import "syscall"
// Readlink returns the destination of the named symbolic link.
func Readlink(path string, buf []byte) (n int, err error) {
return syscall.Readlink(path, buf)
}

View file

@ -1,112 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package syscallx
import (
"syscall"
"unsafe"
)
type reparseDataBuffer struct {
ReparseTag uint32
ReparseDataLength uint16
Reserved uint16
// GenericReparseBuffer
reparseBuffer byte
}
type mountPointReparseBuffer struct {
SubstituteNameOffset uint16
SubstituteNameLength uint16
PrintNameOffset uint16
PrintNameLength uint16
PathBuffer [1]uint16
}
type symbolicLinkReparseBuffer struct {
SubstituteNameOffset uint16
SubstituteNameLength uint16
PrintNameOffset uint16
PrintNameLength uint16
Flags uint32
PathBuffer [1]uint16
}
const (
_IO_REPARSE_TAG_MOUNT_POINT = 0xA0000003
_SYMLINK_FLAG_RELATIVE = 1
)
// Readlink returns the destination of the named symbolic link.
func Readlink(path string, buf []byte) (n int, err error) {
fd, err := syscall.CreateFile(syscall.StringToUTF16Ptr(path), syscall.GENERIC_READ, 0, nil, syscall.OPEN_EXISTING,
syscall.FILE_FLAG_OPEN_REPARSE_POINT|syscall.FILE_FLAG_BACKUP_SEMANTICS, 0)
if err != nil {
return -1, err
}
defer syscall.CloseHandle(fd)
rdbbuf := make([]byte, syscall.MAXIMUM_REPARSE_DATA_BUFFER_SIZE)
var bytesReturned uint32
err = syscall.DeviceIoControl(fd, syscall.FSCTL_GET_REPARSE_POINT, nil, 0, &rdbbuf[0], uint32(len(rdbbuf)), &bytesReturned, nil)
if err != nil {
return -1, err
}
rdb := (*reparseDataBuffer)(unsafe.Pointer(&rdbbuf[0]))
var s string
switch rdb.ReparseTag {
case syscall.IO_REPARSE_TAG_SYMLINK:
data := (*symbolicLinkReparseBuffer)(unsafe.Pointer(&rdb.reparseBuffer))
p := (*[0xffff]uint16)(unsafe.Pointer(&data.PathBuffer[0]))
s = syscall.UTF16ToString(p[data.SubstituteNameOffset/2 : (data.SubstituteNameOffset+data.SubstituteNameLength)/2])
if data.Flags&_SYMLINK_FLAG_RELATIVE == 0 {
if len(s) >= 4 && s[:4] == `\??\` {
s = s[4:]
switch {
case len(s) >= 2 && s[1] == ':': // \??\C:\foo\bar
// do nothing
case len(s) >= 4 && s[:4] == `UNC\`: // \??\UNC\foo\bar
s = `\\` + s[4:]
default:
// unexpected; do nothing
}
} else {
// unexpected; do nothing
}
}
case _IO_REPARSE_TAG_MOUNT_POINT:
data := (*mountPointReparseBuffer)(unsafe.Pointer(&rdb.reparseBuffer))
p := (*[0xffff]uint16)(unsafe.Pointer(&data.PathBuffer[0]))
s = syscall.UTF16ToString(p[data.SubstituteNameOffset/2 : (data.SubstituteNameOffset+data.SubstituteNameLength)/2])
if len(s) >= 4 && s[:4] == `\??\` { // \??\C:\foo\bar
if len(s) < 48 || s[:11] != `\??\Volume{` {
s = s[4:]
}
} else {
// unexpected; do nothing
}
default:
// the path is not a symlink or junction but another type of reparse
// point
return -1, syscall.ENOENT
}
n = copy(buf, []byte(s))
return n, nil
}

View file

@ -1,3 +0,0 @@
This package is for internal use only. It is intended to only have
temporary changes before they are upstreamed to golang.org/x/sys/
(a.k.a. https://github.com/golang/sys).

View file

@ -1,128 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package sysx
import (
"os"
"path/filepath"
"github.com/containerd/continuity/syscallx"
)
// Readlink returns the destination of the named symbolic link.
// If there is an error, it will be of type *PathError.
func Readlink(name string) (string, error) {
for len := 128; ; len *= 2 {
b := make([]byte, len)
n, e := fixCount(syscallx.Readlink(fixLongPath(name), b))
if e != nil {
return "", &os.PathError{Op: "readlink", Path: name, Err: e}
}
if n < len {
return string(b[0:n]), nil
}
}
}
// Many functions in package syscall return a count of -1 instead of 0.
// Using fixCount(call()) instead of call() corrects the count.
func fixCount(n int, err error) (int, error) {
if n < 0 {
n = 0
}
return n, err
}
// fixLongPath returns the extended-length (\\?\-prefixed) form of
// path when needed, in order to avoid the default 260 character file
// path limit imposed by Windows. If path is not easily converted to
// the extended-length form (for example, if path is a relative path
// or contains .. elements), or is short enough, fixLongPath returns
// path unmodified.
//
// See https://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx#maxpath
func fixLongPath(path string) string {
// Do nothing (and don't allocate) if the path is "short".
// Empirically (at least on the Windows Server 2013 builder),
// the kernel is arbitrarily okay with < 248 bytes. That
// matches what the docs above say:
// "When using an API to create a directory, the specified
// path cannot be so long that you cannot append an 8.3 file
// name (that is, the directory name cannot exceed MAX_PATH
// minus 12)." Since MAX_PATH is 260, 260 - 12 = 248.
//
// The MSDN docs appear to say that a normal path that is 248 bytes long
// will work; empirically the path must be less then 248 bytes long.
if len(path) < 248 {
// Don't fix. (This is how Go 1.7 and earlier worked,
// not automatically generating the \\?\ form)
return path
}
// The extended form begins with \\?\, as in
// \\?\c:\windows\foo.txt or \\?\UNC\server\share\foo.txt.
// The extended form disables evaluation of . and .. path
// elements and disables the interpretation of / as equivalent
// to \. The conversion here rewrites / to \ and elides
// . elements as well as trailing or duplicate separators. For
// simplicity it avoids the conversion entirely for relative
// paths or paths containing .. elements. For now,
// \\server\share paths are not converted to
// \\?\UNC\server\share paths because the rules for doing so
// are less well-specified.
if len(path) >= 2 && path[:2] == `\\` {
// Don't canonicalize UNC paths.
return path
}
if !filepath.IsAbs(path) {
// Relative path
return path
}
const prefix = `\\?`
pathbuf := make([]byte, len(prefix)+len(path)+len(`\`))
copy(pathbuf, prefix)
n := len(path)
r, w := 0, len(prefix)
for r < n {
switch {
case os.IsPathSeparator(path[r]):
// empty block
r++
case path[r] == '.' && (r+1 == n || os.IsPathSeparator(path[r+1])):
// /./
r++
case r+1 < n && path[r] == '.' && path[r+1] == '.' && (r+2 == n || os.IsPathSeparator(path[r+2])):
// /../ is currently unhandled
return path
default:
pathbuf[w] = '\\'
w++
for ; r < n && !os.IsPathSeparator(path[r]); r++ {
pathbuf[w] = path[r]
w++
}
}
}
// A drive's root directory needs a trailing \
if w == len(`\\?\c:`) {
pathbuf[w] = '\\'
w++
}
return string(pathbuf[:w])
}

View file

@ -1,52 +0,0 @@
#!/bin/bash
# Copyright The containerd Authors.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# http://www.apache.org/licenses/LICENSE-2.0
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
set -e
mksyscall="$(go env GOROOT)/src/syscall/mksyscall.pl"
fix() {
sed 's,^package syscall$,package sysx,' \
| sed 's,^import "unsafe"$,import (\n\t"syscall"\n\t"unsafe"\n),' \
| gofmt -r='BytePtrFromString -> syscall.BytePtrFromString' \
| gofmt -r='Syscall6 -> syscall.Syscall6' \
| gofmt -r='Syscall -> syscall.Syscall' \
| gofmt -r='SYS_GETXATTR -> syscall.SYS_GETXATTR' \
| gofmt -r='SYS_LISTXATTR -> syscall.SYS_LISTXATTR' \
| gofmt -r='SYS_SETXATTR -> syscall.SYS_SETXATTR' \
| gofmt -r='SYS_REMOVEXATTR -> syscall.SYS_REMOVEXATTR' \
| gofmt -r='SYS_LGETXATTR -> syscall.SYS_LGETXATTR' \
| gofmt -r='SYS_LLISTXATTR -> syscall.SYS_LLISTXATTR' \
| gofmt -r='SYS_LSETXATTR -> syscall.SYS_LSETXATTR' \
| gofmt -r='SYS_LREMOVEXATTR -> syscall.SYS_LREMOVEXATTR'
}
if [ "$GOARCH" == "" ] || [ "$GOOS" == "" ]; then
echo "Must specify \$GOARCH and \$GOOS"
exit 1
fi
mkargs=""
if [ "$GOARCH" == "386" ] || [ "$GOARCH" == "arm" ]; then
mkargs="-l32"
fi
for f in "$@"; do
$mksyscall $mkargs "${f}_${GOOS}.go" | fix > "${f}_${GOOS}_${GOARCH}.go"
done

View file

@ -1,23 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package sysx
import (
"syscall"
)
const ENODATA = syscall.ENODATA

View file

@ -1,24 +0,0 @@
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package sysx
import (
"syscall"
)
// This should actually be a set that contains ENOENT and EPERM
const ENODATA = syscall.ENOENT

View file

@ -1,25 +0,0 @@
// +build darwin freebsd
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package sysx
import (
"syscall"
)
const ENODATA = syscall.ENOATTR

View file

@ -1,125 +0,0 @@
// +build linux darwin
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package sysx
import (
"bytes"
"syscall"
"golang.org/x/sys/unix"
)
// Listxattr calls syscall listxattr and reads all content
// and returns a string array
func Listxattr(path string) ([]string, error) {
return listxattrAll(path, unix.Listxattr)
}
// Removexattr calls syscall removexattr
func Removexattr(path string, attr string) (err error) {
return unix.Removexattr(path, attr)
}
// Setxattr calls syscall setxattr
func Setxattr(path string, attr string, data []byte, flags int) (err error) {
return unix.Setxattr(path, attr, data, flags)
}
// Getxattr calls syscall getxattr
func Getxattr(path, attr string) ([]byte, error) {
return getxattrAll(path, attr, unix.Getxattr)
}
// LListxattr lists xattrs, not following symlinks
func LListxattr(path string) ([]string, error) {
return listxattrAll(path, unix.Llistxattr)
}
// LRemovexattr removes an xattr, not following symlinks
func LRemovexattr(path string, attr string) (err error) {
return unix.Lremovexattr(path, attr)
}
// LSetxattr sets an xattr, not following symlinks
func LSetxattr(path string, attr string, data []byte, flags int) (err error) {
return unix.Lsetxattr(path, attr, data, flags)
}
// LGetxattr gets an xattr, not following symlinks
func LGetxattr(path, attr string) ([]byte, error) {
return getxattrAll(path, attr, unix.Lgetxattr)
}
const defaultXattrBufferSize = 5
type listxattrFunc func(path string, dest []byte) (int, error)
func listxattrAll(path string, listFunc listxattrFunc) ([]string, error) {
var p []byte // nil on first execution
for {
n, err := listFunc(path, p) // first call gets buffer size.
if err != nil {
return nil, err
}
if n > len(p) {
p = make([]byte, n)
continue
}
p = p[:n]
ps := bytes.Split(bytes.TrimSuffix(p, []byte{0}), []byte{0})
var entries []string
for _, p := range ps {
s := string(p)
if s != "" {
entries = append(entries, s)
}
}
return entries, nil
}
}
type getxattrFunc func(string, string, []byte) (int, error)
func getxattrAll(path, attr string, getFunc getxattrFunc) ([]byte, error) {
p := make([]byte, defaultXattrBufferSize)
for {
n, err := getFunc(path, attr, p)
if err != nil {
if errno, ok := err.(syscall.Errno); ok && errno == syscall.ERANGE {
p = make([]byte, len(p)*2) // this can't be ideal.
continue // try again!
}
return nil, err
}
// realloc to correct size and repeat
if n > len(p) {
p = make([]byte, n)
continue
}
return p[:n], nil
}
}

View file

@ -1,67 +0,0 @@
// +build !linux,!darwin
/*
Copyright The containerd Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package sysx
import (
"errors"
"runtime"
)
var unsupported = errors.New("extended attributes unsupported on " + runtime.GOOS)
// Listxattr calls syscall listxattr and reads all content
// and returns a string array
func Listxattr(path string) ([]string, error) {
return []string{}, nil
}
// Removexattr calls syscall removexattr
func Removexattr(path string, attr string) (err error) {
return unsupported
}
// Setxattr calls syscall setxattr
func Setxattr(path string, attr string, data []byte, flags int) (err error) {
return unsupported
}
// Getxattr calls syscall getxattr
func Getxattr(path, attr string) ([]byte, error) {
return []byte{}, unsupported
}
// LListxattr lists xattrs, not following symlinks
func LListxattr(path string) ([]string, error) {
return []string{}, nil
}
// LRemovexattr removes an xattr, not following symlinks
func LRemovexattr(path string, attr string) (err error) {
return unsupported
}
// LSetxattr sets an xattr, not following symlinks
func LSetxattr(path string, attr string, data []byte, flags int) (err error) {
return unsupported
}
// LGetxattr gets an xattr, not following symlinks
func LGetxattr(path, attr string) ([]byte, error) {
return []byte{}, nil
}

View file

@ -1,2 +0,0 @@
/bin
/gopath

View file

@ -1,16 +0,0 @@
language: go
go:
- "1.12"
- "1.13"
install:
- go get -v -t github.com/coreos/go-oidc/...
- go get golang.org/x/tools/cmd/cover
- go get golang.org/x/lint/golint
script:
- ./test
notifications:
email: false

View file

@ -1,71 +0,0 @@
# How to Contribute
CoreOS projects are [Apache 2.0 licensed](LICENSE) and accept contributions via
GitHub pull requests. This document outlines some of the conventions on
development workflow, commit message formatting, contact points and other
resources to make it easier to get your contribution accepted.
# Certificate of Origin
By contributing to this project you agree to the Developer Certificate of
Origin (DCO). This document was created by the Linux Kernel community and is a
simple statement that you, as a contributor, have the legal right to make the
contribution. See the [DCO](DCO) file for details.
# Email and Chat
The project currently uses the general CoreOS email list and IRC channel:
- Email: [coreos-dev](https://groups.google.com/forum/#!forum/coreos-dev)
- IRC: #[coreos](irc://irc.freenode.org:6667/#coreos) IRC channel on freenode.org
Please avoid emailing maintainers found in the MAINTAINERS file directly. They
are very busy and read the mailing lists.
## Getting Started
- Fork the repository on GitHub
- Read the [README](README.md) for build and test instructions
- Play with the project, submit bugs, submit patches!
## Contribution Flow
This is a rough outline of what a contributor's workflow looks like:
- Create a topic branch from where you want to base your work (usually master).
- Make commits of logical units.
- Make sure your commit messages are in the proper format (see below).
- Push your changes to a topic branch in your fork of the repository.
- Make sure the tests pass, and add any new tests as appropriate.
- Submit a pull request to the original repository.
Thanks for your contributions!
### Format of the Commit Message
We follow a rough convention for commit messages that is designed to answer two
questions: what changed and why. The subject line should feature the what and
the body of the commit should describe the why.
```
scripts: add the test-cluster command
this uses tmux to setup a test cluster that you can easily kill and
start for debugging.
Fixes #38
```
The format can be described more formally as follows:
```
<subsystem>: <what changed>
<BLANK LINE>
<why this change was made>
<BLANK LINE>
<footer>
```
The first line is the subject and should be no longer than 70 characters, the
second line is always blank, and other lines should be wrapped at 80 characters.
This allows the message to be easier to read on GitHub as well as in various
git tools.

36
vendor/github.com/coreos/go-oidc/DCO generated vendored
View file

@ -1,36 +0,0 @@
Developer Certificate of Origin
Version 1.1
Copyright (C) 2004, 2006 The Linux Foundation and its contributors.
660 York Street, Suite 102,
San Francisco, CA 94110 USA
Everyone is permitted to copy and distribute verbatim copies of this
license document, but changing it is not allowed.
Developer's Certificate of Origin 1.1
By making a contribution to this project, I certify that:
(a) The contribution was created in whole or in part by me and I
have the right to submit it under the open source license
indicated in the file; or
(b) The contribution is based upon previous work that, to the best
of my knowledge, is covered under an appropriate open source
license and I have the right under that license to submit that
work with modifications, whether created in whole or in part
by me, under the same open source license (unless I am
permitted to submit under a different license), as indicated
in the file; or
(c) The contribution was provided directly to me by some other
person who certified (a), (b) or (c) and I have not modified
it.
(d) I understand and agree that this project and the contribution
are public and that a record of the contribution (including all
personal information I submit with it, including my sign-off) is
maintained indefinitely and may be redistributed consistent with
this project or the open source license(s) involved.

View file

@ -1,202 +0,0 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "{}"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright {yyyy} {name of copyright owner}
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

View file

@ -1,3 +0,0 @@
Eric Chiang <ericchiang@google.com> (@ericchiang)
Mike Danese <mikedanese@google.com> (@mikedanese)
Rithu Leena John <rjohn@redhat.com> (@rithujohn191)

View file

@ -1,5 +0,0 @@
CoreOS Project
Copyright 2014 CoreOS, Inc
This product includes software developed at CoreOS, Inc.
(http://www.coreos.com/).

View file

@ -1,72 +0,0 @@
# go-oidc
[![GoDoc](https://godoc.org/github.com/coreos/go-oidc?status.svg)](https://godoc.org/github.com/coreos/go-oidc)
[![Build Status](https://travis-ci.org/coreos/go-oidc.png?branch=master)](https://travis-ci.org/coreos/go-oidc)
## OpenID Connect support for Go
This package enables OpenID Connect support for the [golang.org/x/oauth2](https://godoc.org/golang.org/x/oauth2) package.
```go
provider, err := oidc.NewProvider(ctx, "https://accounts.google.com")
if err != nil {
// handle error
}
// Configure an OpenID Connect aware OAuth2 client.
oauth2Config := oauth2.Config{
ClientID: clientID,
ClientSecret: clientSecret,
RedirectURL: redirectURL,
// Discovery returns the OAuth2 endpoints.
Endpoint: provider.Endpoint(),
// "openid" is a required scope for OpenID Connect flows.
Scopes: []string{oidc.ScopeOpenID, "profile", "email"},
}
```
OAuth2 redirects are unchanged.
```go
func handleRedirect(w http.ResponseWriter, r *http.Request) {
http.Redirect(w, r, oauth2Config.AuthCodeURL(state), http.StatusFound)
}
```
The on responses, the provider can be used to verify ID Tokens.
```go
var verifier = provider.Verifier(&oidc.Config{ClientID: clientID})
func handleOAuth2Callback(w http.ResponseWriter, r *http.Request) {
// Verify state and errors.
oauth2Token, err := oauth2Config.Exchange(ctx, r.URL.Query().Get("code"))
if err != nil {
// handle error
}
// Extract the ID Token from OAuth2 token.
rawIDToken, ok := oauth2Token.Extra("id_token").(string)
if !ok {
// handle missing token
}
// Parse and verify ID Token payload.
idToken, err := verifier.Verify(ctx, rawIDToken)
if err != nil {
// handle error
}
// Extract custom claims
var claims struct {
Email string `json:"email"`
Verified bool `json:"email_verified"`
}
if err := idToken.Claims(&claims); err != nil {
// handle error
}
}
```

View file

@ -1,61 +0,0 @@
## CoreOS Community Code of Conduct
### Contributor Code of Conduct
As contributors and maintainers of this project, and in the interest of
fostering an open and welcoming community, we pledge to respect all people who
contribute through reporting issues, posting feature requests, updating
documentation, submitting pull requests or patches, and other activities.
We are committed to making participation in this project a harassment-free
experience for everyone, regardless of level of experience, gender, gender
identity and expression, sexual orientation, disability, personal appearance,
body size, race, ethnicity, age, religion, or nationality.
Examples of unacceptable behavior by participants include:
* The use of sexualized language or imagery
* Personal attacks
* Trolling or insulting/derogatory comments
* Public or private harassment
* Publishing others' private information, such as physical or electronic addresses, without explicit permission
* Other unethical or unprofessional conduct.
Project maintainers have the right and responsibility to remove, edit, or
reject comments, commits, code, wiki edits, issues, and other contributions
that are not aligned to this Code of Conduct. By adopting this Code of Conduct,
project maintainers commit themselves to fairly and consistently applying these
principles to every aspect of managing this project. Project maintainers who do
not follow or enforce the Code of Conduct may be permanently removed from the
project team.
This code of conduct applies both within project spaces and in public spaces
when an individual is representing the project or its community.
Instances of abusive, harassing, or otherwise unacceptable behavior may be
reported by contacting a project maintainer, Brandon Philips
<brandon.philips@coreos.com>, and/or Rithu John <rithu.john@coreos.com>.
This Code of Conduct is adapted from the Contributor Covenant
(http://contributor-covenant.org), version 1.2.0, available at
http://contributor-covenant.org/version/1/2/0/
### CoreOS Events Code of Conduct
CoreOS events are working conferences intended for professional networking and
collaboration in the CoreOS community. Attendees are expected to behave
according to professional standards and in accordance with their employers
policies on appropriate workplace behavior.
While at CoreOS events or related social networking opportunities, attendees
should not engage in discriminatory or offensive speech or actions including
but not limited to gender, sexuality, race, age, disability, or religion.
Speakers should be especially aware of these concerns.
CoreOS does not condone any statements by speakers contrary to these standards.
CoreOS reserves the right to deny entrance and/or eject from an event (without
refund) any individual found to be engaging in discriminatory or offensive
speech or actions.
Please bring any concerns to the immediate attention of designated on-site
staff, Brandon Philips <brandon.philips@coreos.com>, and/or Rithu John <rithu.john@coreos.com>.

Some files were not shown because too many files have changed in this diff Show more