debian-mirror-gitlab/doc/development/policies.md

274 lines
11 KiB
Markdown
Raw Normal View History

2021-01-29 00:20:46 +05:30
---
stage: Manage
group: Access
2021-02-22 17:27:13 +05:30
info: To determine the technical writer assigned to the Stage/Group associated with this page, see https://about.gitlab.com/handbook/engineering/ux/technical-writing/#assignments
2021-01-29 00:20:46 +05:30
---
2017-09-10 17:25:29 +05:30
# `DeclarativePolicy` framework
The DeclarativePolicy framework is designed to assist in performance of policy checks, and to enable ease of extension for EE. The DSL code in `app/policies` is what `Ability.allowed?` uses to check whether a particular action is allowed on a subject.
2021-02-22 17:27:13 +05:30
The policy used is based on the subject's class name - so `Ability.allowed?(user, :some_ability, project)` creates a `ProjectPolicy` and check permissions on that.
2017-09-10 17:25:29 +05:30
## Managing Permission Rules
Permissions are broken into two parts: `conditions` and `rules`. Conditions are boolean expressions that can access the database and the environment, while rules are statically configured combinations of expressions and other rules that enable or prevent certain abilities. For an ability to be allowed, it must be enabled by at least one rule, and not prevented by any.
### Conditions
2021-02-22 17:27:13 +05:30
Conditions are defined by the `condition` method, and are given a name and a block. The block is executed in the context of the policy object - so it can access `@user` and `@subject`, as well as call any methods defined on the policy. Note that `@user` may be nil (in the anonymous case), but `@subject` is guaranteed to be a real instance of the subject class.
2017-09-10 17:25:29 +05:30
2019-03-02 22:35:43 +05:30
```ruby
2017-09-10 17:25:29 +05:30
class FooPolicy < BasePolicy
condition(:is_public) do
# @subject guaranteed to be an instance of Foo
@subject.public?
end
# instance methods can be called from the condition as well
condition(:thing) { check_thing }
def check_thing
# ...
end
end
```
2021-02-22 17:27:13 +05:30
When you define a condition, a predicate method is defined on the policy to check whether that condition passes - so in the above example, an instance of `FooPolicy` also responds to `#is_public?` and `#thing?`.
2017-09-10 17:25:29 +05:30
2021-02-22 17:27:13 +05:30
Conditions are cached according to their scope. Scope and ordering is covered later.
2017-09-10 17:25:29 +05:30
### Rules
A `rule` is a logical combination of conditions and other rules, that are configured to enable or prevent certain abilities. It is important to note that the rule configuration is static - a rule's logic cannot touch the database or know about `@user` or `@subject`. This allows us to cache only at the condition level. Rules are specified through the `rule` method, which takes a block of DSL configuration, and returns an object that responds to `#enable` or `#prevent`:
2019-03-02 22:35:43 +05:30
```ruby
2017-09-10 17:25:29 +05:30
class FooPolicy < BasePolicy
# ...
rule { is_public }.enable :read
rule { thing }.prevent :read
# equivalently,
rule { is_public }.policy do
enable :read
end
rule { ~thing }.policy do
prevent :read
end
end
```
Within the rule DSL, you can use:
2019-03-02 22:35:43 +05:30
- A regular word mentions a condition by name - a rule that is in effect when that condition is truthy.
2021-02-22 17:27:13 +05:30
- `~` indicates negation, also available as `negate`.
2019-03-02 22:35:43 +05:30
- `&` and `|` are logical combinations, also available as `all?(...)` and `any?(...)`.
- `can?(:other_ability)` delegates to the rules that apply to `:other_ability`. Note that this is distinct from the instance method `can?`, which can check dynamically - this only configures a delegation to another ability.
2017-09-10 17:25:29 +05:30
2021-02-22 17:27:13 +05:30
`~`, `&` and `|` operators are overridden methods in
2021-06-08 01:23:25 +05:30
[`DeclarativePolicy::Rule::Base`](https://gitlab.com/gitlab-org/declarative-policy/-/blob/main/lib/declarative_policy/rule.rb).
2021-02-22 17:27:13 +05:30
Do not use boolean operators such as `&&` and `||` within the rule DSL,
as conditions within rule blocks are objects, not booleans. The same
applies for ternary operators (`condition ? ... : ...`), and `if`
blocks. These operators cannot be overridden, and are hence banned via a
[custom
cop](https://gitlab.com/gitlab-org/gitlab/-/merge_requests/49771).
2017-09-10 17:25:29 +05:30
## Scores, Order, Performance
2021-02-22 17:27:13 +05:30
To see how the rules get evaluated into a judgment, it is useful in a console to use `policy.debug(:some_ability)`. This prints the rules in the order they are evaluated.
2017-09-10 17:25:29 +05:30
2019-07-07 11:18:12 +05:30
For example, let's say you wanted to debug `IssuePolicy`. You might run
the debugger in this way:
```ruby
user = User.find_by(username: 'john')
issue = Issue.first
policy = IssuePolicy.new(user, issue)
policy.debug(:read_issue)
```
An example debug output would look as follows:
```ruby
- [0] prevent when all?(confidential, ~can_read_confidential) ((@john : Issue/1))
- [0] prevent when archived ((@john : Project/4))
- [0] prevent when issues_disabled ((@john : Project/4))
- [0] prevent when all?(anonymous, ~public_project) ((@john : Project/4))
+ [32] enable when can?(:reporter_access) ((@john : Project/4))
```
Each line represents a rule that was evaluated. There are a few things to note:
1. The `-` or `+` symbol indicates whether the rule block was evaluated to be
`false` or `true`, respectively.
2019-10-12 21:52:04 +05:30
1. The number inside the brackets indicates the score.
2021-09-30 23:02:18 +05:30
1. The last part of the line (for example, `@john : Issue/1`) shows the username
2019-07-07 11:18:12 +05:30
and subject for that rule.
Here you can see that the first four rules were evaluated `false` for
which user and subject. For example, you can see in the last line that
2020-05-24 23:13:21 +05:30
the rule was activated because the user `john` had Reporter access to
2020-03-13 15:44:24 +05:30
`Project/4`.
2019-07-07 11:18:12 +05:30
When a policy is asked whether a particular ability is allowed
(`policy.allowed?(:some_ability)`), it does not necessarily have to
compute all the conditions on the policy. First, only the rules relevant
to that particular ability are selected. Then, the execution model takes
advantage of short-circuiting, and attempts to sort rules based on a
2021-02-22 17:27:13 +05:30
heuristic of how expensive they are to calculate. The sorting is
dynamic and cache-aware, so that previously calculated conditions are
considered first, before computing other conditions.
2019-07-07 11:18:12 +05:30
Note that the score is chosen by a developer via the `score:` parameter
in a `condition` to denote how expensive evaluating this rule would be
relative to other rules.
2017-09-10 17:25:29 +05:30
## Scope
2021-02-22 17:27:13 +05:30
Sometimes, a condition only uses data from `@user` or only from `@subject`. In this case, we want to change the scope of the caching, so that we don't recalculate conditions unnecessarily. For example, given:
2017-09-10 17:25:29 +05:30
2019-03-02 22:35:43 +05:30
```ruby
2017-09-10 17:25:29 +05:30
class FooPolicy < BasePolicy
condition(:expensive_condition) { @subject.expensive_query? }
rule { expensive_condition }.enable :some_ability
end
```
2020-05-24 23:13:21 +05:30
Naively, if we call `Ability.allowed?(user1, :some_ability, foo)` and `Ability.allowed?(user2, :some_ability, foo)`, we would have to calculate the condition twice - since they are for different users. But if we use the `scope: :subject` option:
2017-09-10 17:25:29 +05:30
2019-03-02 22:35:43 +05:30
```ruby
2017-09-10 17:25:29 +05:30
condition(:expensive_condition, scope: :subject) { @subject.expensive_query? }
```
2021-02-22 17:27:13 +05:30
then the result of the condition is cached globally only based on the subject - so it is not calculated repeatedly for different users. Similarly, `scope: :user` caches only based on the user.
2017-09-10 17:25:29 +05:30
**DANGER**: If you use a `:scope` option when the condition actually uses data from
2021-02-22 17:27:13 +05:30
both user and subject (including a simple anonymous check!) your result is cached at too global of a scope and results in cache bugs.
2017-09-10 17:25:29 +05:30
2021-09-30 23:02:18 +05:30
Sometimes we are checking permissions for a lot of users for one subject, or a lot of subjects for one user. In this case, we want to set a *preferred scope* - that is, tell the system that we prefer rules that can be cached on the repeated parameter. For example, in `Ability.users_that_can_read_project`:
2017-09-10 17:25:29 +05:30
2019-03-02 22:35:43 +05:30
```ruby
2017-09-10 17:25:29 +05:30
def users_that_can_read_project(users, project)
DeclarativePolicy.subject_scope do
users.select { |u| allowed?(u, :read_project, project) }
end
end
```
2021-02-22 17:27:13 +05:30
This, for example, prefers checking `project.public?` to checking `user.admin?`.
2017-09-10 17:25:29 +05:30
## Delegation
2019-03-02 22:35:43 +05:30
Delegation is the inclusion of rules from another policy, on a different subject. For example:
2017-09-10 17:25:29 +05:30
2019-03-02 22:35:43 +05:30
```ruby
2017-09-10 17:25:29 +05:30
class FooPolicy < BasePolicy
delegate { @subject.project }
end
```
2021-02-22 17:27:13 +05:30
includes all rules from `ProjectPolicy`. The delegated conditions are evaluated with the correct delegated subject, and are sorted along with the regular rules in the policy. Note that only the relevant rules for a particular ability are actually considered.
2019-12-26 22:10:19 +05:30
2020-07-28 23:09:34 +05:30
### Overrides
We allow policies to opt-out of delegated abilities.
Delegated policies may define some abilities in a way that is incorrect for the
delegating policy. Take for example a child/parent relationship, where some
abilities can be inferred, and some cannot:
```ruby
class ParentPolicy < BasePolicy
condition(:speaks_spanish) { @subject.spoken_languages.include?(:es) }
condition(:has_license) { @subject.driving_license.present? }
condition(:enjoys_broccoli) { @subject.enjoyment_of(:broccoli) > 0 }
2021-01-03 14:25:43 +05:30
2020-07-28 23:09:34 +05:30
rule { speaks_spanish }.enable :read_spanish
rule { has_license }.enable :drive_car
rule { enjoys_broccoli }.enable :eat_broccoli
rule { ~enjoys_broccoli }.prevent :eat_broccoli
end
```
Here, if we delegated the child policy to the parent policy, some values would be
incorrect - we might correctly infer that the child can speak their parent's
language, but it would be incorrect to infer that the child can drive or would
eat broccoli just because the parent can and does.
Some of these things we can deal with - we can forbid driving universally in the
child policy, for example:
```ruby
class ChildPolicy < BasePolicy
delegate { @subject.parent }
2021-01-03 14:25:43 +05:30
2020-07-28 23:09:34 +05:30
rule { default }.prevent :drive_car
end
```
But the food preferences one is harder - because of the `prevent` call in the
parent policy, if the parent dislikes it, even calling `enable` in the child
2021-02-22 17:27:13 +05:30
does not enable `:eat_broccoli`.
2020-07-28 23:09:34 +05:30
We could remove the `prevent` call in the parent policy, but that still doesn't
help us, since the rules are different: parents get to eat what they like, and
children eat what they are given, provided they are well behaved. Allowing
delegation would end up with only children whose parents enjoy green vegetables
eating it. But a parent may well give their child broccoli, even if they dislike
it themselves, because it is good for their child.
The solution it to override the `:eat_broccoli` ability in the child policy:
```ruby
class ChildPolicy < BasePolicy
delegate { @subject.parent }
2021-01-03 14:25:43 +05:30
2020-07-28 23:09:34 +05:30
overrides :eat_broccoli
2021-01-03 14:25:43 +05:30
2020-07-28 23:09:34 +05:30
condition(:good_kid) { @subject.behavior_level >= Child::GOOD }
2021-01-03 14:25:43 +05:30
2020-07-28 23:09:34 +05:30
rule { good_kid }.enable :eat_broccoli
end
```
2021-02-22 17:27:13 +05:30
With this definition, the `ChildPolicy` _never_ looks in the `ParentPolicy` to
2020-07-28 23:09:34 +05:30
satisfy `:eat_broccoli`, but it _will_ use it for any other abilities. The child
policy can then define `:eat_broccoli` in a way that makes sense for `Child` and not
`Parent`.
### Alternatives to using `overrides`
Overriding policy delegation is complex, for the same reason delegation is
complex - it involves reasoning about logical inference, and being clear about
semantics. Misuse of `override` has the potential to duplicate code, and
potentially introduce security bugs, allowing things that should be prevented.
For this reason, it should be used only when other approaches are not feasible.
Other approaches can include for example using different ability names. Choosing
to eat a food and eating foods you are given are semantically distinct, and they
could be named differently (perhaps `chooses_to_eat_broccoli` and
`eats_what_is_given` in this case). It can depend on how polymorphic the call
2021-02-22 17:27:13 +05:30
site is. If you know that we always check the policy with a `Parent` or a
2020-07-28 23:09:34 +05:30
`Child`, then we can choose the appropriate ability name. If the call site is
polymorphic, then we cannot do that.
2019-12-26 22:10:19 +05:30
## Specifying Policy Class
You can also override the Policy used for a given subject:
```ruby
class Foo
def self.declarative_policy_class
'SomeOtherPolicy'
end
end
```
2021-02-22 17:27:13 +05:30
This uses and checks permissions on the `SomeOtherPolicy` class rather than the usual calculated `FooPolicy` class.