forgejo-federation/services/wiki/wiki_path.go
wxiaoguang b39a5bbbd6
Make wiki title supports dashes and improve wiki name related features (#24143)
Close #7570


1. Clearly define the wiki path behaviors, see
`services/wiki/wiki_path.go` and tests
2. Keep compatibility with old contents
3. Allow to use dashes in titles, eg: "2000-01-02 Meeting record"
4. Add a "Pages" link in the dropdown, otherwise users can't go to the
Pages page easily.
5. Add a "View original git file" link in the Pages list, even if some
file names are broken, users still have a chance to edit or remove it,
without cloning the wiki repo to local.
6. Fix 500 error when the name contains prefix spaces.


This PR also introduces the ability to support sub-directories, but it
can't be done at the moment due to there are a lot of legacy wiki data,
which use "%2F" in file names.



![image](https://user-images.githubusercontent.com/2114189/232239004-3359d7b9-7bf3-4ff3-8446-bfb0e79645dd.png)


![image](https://user-images.githubusercontent.com/2114189/232239020-74b92c72-bf73-4377-a319-1c85609f82b1.png)

Co-authored-by: Giteabot <teabot@gitea.io>
2023-04-19 13:50:10 -04:00

153 lines
4.4 KiB
Go

// Copyright 2023 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package wiki
import (
"net/url"
"path"
"strings"
repo_model "code.gitea.io/gitea/models/repo"
"code.gitea.io/gitea/modules/util"
)
// To define the wiki related concepts:
// * Display Segment: the text what user see for a wiki page (aka, the title):
// - "Home Page"
// - "100% Free"
// - "2000-01-02 meeting"
// * Web Path:
// - "/wiki/Home-Page"
// - "/wiki/100%25+Free"
// - "/wiki/2000-01-02+meeting.-"
// - If a segment has a suffix "DashMarker(.-)", it means that there is no dash-space conversion for this segment.
// - If a WebPath is a "*.md" pattern, then use it directly as GitPath, to make users can access the raw file.
// * Git Path (only space doesn't need to be escaped):
// - "/.wiki.git/Home-Page.md"
// - "/.wiki.git/100%25 Free.md"
// - "/.wiki.git/2000-01-02 meeting.-.md"
// TODO: support subdirectory in the future
//
// Although this package now has the ablity to support subdirectory, but the route package doesn't:
// * Double-escaping problem: the URL "/wiki/abc%2Fdef" becomes "/wiki/abc/def" by ctx.Params, which is incorrect
// * The old wiki code's behavior is always using %2F, instead of subdirectory, so there are a lot of legacy "%2F" files in user wikis.
type WebPath string
var reservedWikiNames = []string{"_pages", "_new", "_edit", "raw"}
func validateWebPath(name WebPath) error {
for _, s := range WebPathSegments(name) {
if util.SliceContainsString(reservedWikiNames, s) {
return repo_model.ErrWikiReservedName{Title: s}
}
}
return nil
}
func hasDashMarker(s string) bool {
return strings.HasSuffix(s, ".-")
}
func removeDashMarker(s string) string {
return strings.TrimSuffix(s, ".-")
}
func addDashMarker(s string) string {
return s + ".-"
}
func unescapeSegment(s string) (string, error) {
if hasDashMarker(s) {
s = removeDashMarker(s)
} else {
s = strings.ReplaceAll(s, "-", " ")
}
unescaped, err := url.QueryUnescape(s)
if err != nil {
return s, err // un-escaping failed, but it's still safe to return the original string, because it is only a title for end users
}
return unescaped, nil
}
func escapeSegToWeb(s string, hadDashMarker bool) string {
if hadDashMarker || strings.Contains(s, "-") {
s = addDashMarker(s)
} else {
s = strings.ReplaceAll(s, " ", "-")
}
s = url.QueryEscape(s)
return s
}
func WebPathSegments(s WebPath) []string {
a := strings.Split(string(s), "/")
for i := range a {
a[i], _ = unescapeSegment(a[i])
}
return a
}
func WebPathToGitPath(s WebPath) string {
if strings.HasSuffix(string(s), ".md") {
return string(s)
}
a := strings.Split(string(s), "/")
for i := range a {
shouldAddDashMarker := hasDashMarker(a[i])
a[i], _ = unescapeSegment(a[i])
a[i] = escapeSegToWeb(a[i], shouldAddDashMarker)
a[i] = strings.ReplaceAll(a[i], "%20", " ") // space is safe to be kept in git path
a[i] = strings.ReplaceAll(a[i], "+", " ")
}
return strings.Join(a, "/") + ".md"
}
func GitPathToWebPath(s string) (wp WebPath, err error) {
if !strings.HasSuffix(s, ".md") {
return "", repo_model.ErrWikiInvalidFileName{FileName: s}
}
s = strings.TrimSuffix(s, ".md")
a := strings.Split(s, "/")
for i := range a {
shouldAddDashMarker := hasDashMarker(a[i])
if a[i], err = unescapeSegment(a[i]); err != nil {
return "", err
}
a[i] = escapeSegToWeb(a[i], shouldAddDashMarker)
}
return WebPath(strings.Join(a, "/")), nil
}
func WebPathToUserTitle(s WebPath) (dir, display string) {
dir = path.Dir(string(s))
display = path.Base(string(s))
display = strings.TrimSuffix(display, ".md")
display, _ = unescapeSegment(display)
return dir, display
}
func WebPathToURLPath(s WebPath) string {
return string(s)
}
func WebPathFromRequest(s string) WebPath {
s = util.PathJoinRelX(s)
// The old wiki code's behavior is always using %2F, instead of subdirectory.
s = strings.ReplaceAll(s, "/", "%2F")
return WebPath(s)
}
func UserTitleToWebPath(base, title string) WebPath {
// TODO: ctx.Params does un-escaping, so the URL "/wiki/abc%2Fdef" becomes "wiki path = `abc/def`", which is incorrect.
// And the old wiki code's behavior is always using %2F, instead of subdirectory.
// So we do not add the support for writing slashes in title at the moment.
title = strings.TrimSpace(title)
title = util.PathJoinRelX(base, escapeSegToWeb(title, false))
if title == "" || title == "." {
title = "unnamed"
}
return WebPath(title)
}