Rust: Implement path resolution in QL #18579

hvitved · 2025-01-23T13:42:16Z

This PR adds an initial implementation of path resolution in QL, that is, resolving paths to items.

The implementation is based on the idea of an item graph, which is a labeled directed graph, where an edge item1 --name--> item2 means that item2 is available inside the scope of item1 under the name name. For example, if we have

mod m1 {
    mod m2 { }
}

then there is an edge m1 --m2--> m1::m2.

The implementation has a lot of known limitations for now, for example

Cross-crate imports are not modeled, since we do not have access (yet) to .toml files in QL.
Modeling of the special Self and crate is partial.
Generic paths Foo<Bar> are not modeled.
Qualified paths like <Foo as Bar> are not modeled.
Shadowing glob imports are not handled correctly, and difficult to do without running into non-monotonic recursion. Probably not a big deal in practice.
Path attributes like #[path = "foo.rs"] mod bar; are not handled.
Preludes are not taken into account.

Copilot

Copilot reviewed 14 out of 14 changed files in this pull request and generated no comments.

Tip: Copilot code review supports C#, Go, Java, JavaScript, Markdown, Python, Ruby and TypeScript, with more languages coming soon. Learn more

rust/ql/lib/codeql/rust/AstConsistency.qll

paldepind · 2025-01-29T13:09:36Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

+  pragma[inline_late]
+  predicate isPublic() { exists(this.getVisibility()) }
+
+  /** Gets an element that has this item as immediately enlcosing item. */


Suggested change

/** Gets an element that has this item as immediately enlcosing item. */

/** Gets an element that has this item as immediately enclosing item. */

paldepind · 2025-01-29T13:52:04Z

rust/ql/test/library-tests/modules/modules.ql

+      commmentAt(value, filepath, line) and
+      inMacro = false
+      or
+      (
+        not commmentAt(_, filepath, line)
+        or
+        inMacro = true
+      ) and
+      value = i.getName()


Makes it easier to see that the first disjunct is (roughly) being negated in the second disjunct.

Suggested change

commmentAt(value, filepath, line) and

inMacro = false

or

(

not commmentAt(_, filepath, line)

or

inMacro = true

) and

value = i.getName()

commmentAt(value, filepath, line) and inMacro = false

or

not (commmentAt(_, filepath, line) and inMacro = false) and

paldepind · 2025-01-30T08:27:56Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

+  )
+}
+
+private class RelevantPath extends Path {


Suggested change

private class RelevantPath extends Path {

/** A path that does not access a local variable. */

private class RelevantPath extends Path {

paldepind · 2025-01-30T08:37:31Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

+  override Visibility getVisibility() { none() }
+}
+
+private predicate sourceFileEdge(SourceFile f, string name, ItemNode item) {


Suggested change

private predicate sourceFileEdge(SourceFile f, string name, ItemNode item) {

/** Holds if `item` has the name `name` and is a top-level item inside `f`. */

private predicate sourceFileEdge(SourceFile f, string name, ItemNode item) {

paldepind · 2025-01-30T08:58:00Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

+  /** Holds if this item is declared as `pub`. */
+  bindingset[this]
+  pragma[inline_late]
+  predicate isPublic() { exists(this.getVisibility()) }


Should this predicate handle pub(...) things eventually? If not hasVisibility might be a better name

Yeah, I think eventually pub(...) should be handled.

paldepind · 2025-01-30T09:17:41Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

+  RelevantPath() { not this = any(VariableAccess va).(PathExpr).getPath() }
+
+  pragma[nomagic]
+  predicate isRoot(string name) {


This name gives me the intuition that the path is related to the crate root, but I don't think that's the case. Would it make sense to call it something along the lines of isUnqualifiedPath?

paldepind · 2025-01-30T09:20:38Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

+ * may be looked up inside enclosing item `encl`.
+ */
+pragma[nomagic]
+private predicate rootPathLookup(RelevantPath root, string name, ItemNode encl) {


A lot of the other predicates here have ItemNode as a result, could this one be the same?

I think I prefer it as-is, because encl is not functionally determined.

paldepind · 2025-01-30T09:21:51Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

@@ -0,0 +1,464 @@
+/** Provides functionality for resolving paths. */


Suggested change

/** Provides functionality for resolving paths. */

/** Provides the `resolvePath` predicate for resolving paths. */

I prefer to keep it as-is, since we are also exposing other related concepts, such as ItemNode.

Makes sense. What about adding another sentence that mentions resolvePath then? It seems to be the primary export, so I think it's helpful to mention explicitly.

paldepind · 2025-01-30T09:23:23Z

rust/ql/lib/codeql/rust/elements/internal/PathResolution.qll

+ * any item (named `name`) inside the imported source file `f`. Using this
+ * edge, `m2::foo` can resolve to `f::foo`, which results in the edge
+ * `m1::use m2 --foo--> f::foo`. Lastly, all edges out of `use` nodes are
+ * lifted to predecessors in the graph, so we get an edge `m1 --foo--> f::foo`.


This all helps a lot with understanding 👍

paldepind · 2025-01-30T09:32:20Z

rust/ql/test/library-tests/modules/main.rs

@@ -0,0 +1,201 @@
+mod my; // I1


Perhaps the directory name modules -> path-resolution would make it more clear what is being tested?

paldepind

Looks great :)

github-actions bot added the Rust Pull requests that update Rust code label Jan 23, 2025

hvitved force-pushed the rust/path-resolution branch 10 times, most recently from 707d17c to 9bfd713 Compare January 28, 2025 09:22

hvitved marked this pull request as ready for review January 28, 2025 12:17

Copilot AI review requested due to automatic review settings January 28, 2025 12:17

Copilot AI reviewed Jan 28, 2025

View reviewed changes

paldepind self-assigned this Jan 29, 2025

paldepind requested changes Jan 30, 2025

View reviewed changes

paldepind reviewed Jan 30, 2025

View reviewed changes

hvitved force-pushed the rust/path-resolution branch from 4a57786 to 0d5e408 Compare January 30, 2025 12:58

hvitved added 5 commits January 31, 2025 10:07

Rust: Implement path resolution in QL

0aee2e6

Rust: Rename modules test to path-resolution

1f6d39f

Address review comments

8eb5792

Rust: Elaborate QL doc on PathResolution.qll

9d06f80

Rust: Remove useTreeIsGlobImport workaround

1cb524f

hvitved force-pushed the rust/path-resolution branch from 4361307 to 1cb524f Compare January 31, 2025 09:11

hvitved requested a review from paldepind January 31, 2025 09:11

paldepind approved these changes Jan 31, 2025

View reviewed changes

hvitved merged commit 180782d into github:main Jan 31, 2025
17 checks passed

hvitved deleted the rust/path-resolution branch January 31, 2025 12:22

hvitved mentioned this pull request Jan 31, 2025

Rust: Use PathResolution module in data flow #18609

Merged

	/** Gets an element that has this item as immediately enlcosing item. */
	/** Gets an element that has this item as immediately enclosing item. */

	private class RelevantPath extends Path {
	/** A path that does not access a local variable. */
	private class RelevantPath extends Path {

	private predicate sourceFileEdge(SourceFile f, string name, ItemNode item) {
	/** Holds if `item` has the name `name` and is a top-level item inside `f`. */
	private predicate sourceFileEdge(SourceFile f, string name, ItemNode item) {

		@@ -0,0 +1,464 @@
		/** Provides functionality for resolving paths. */

	/** Provides functionality for resolving paths. */
	/** Provides the `resolvePath` predicate for resolving paths. */

Rust: Implement path resolution in QL #18579

Rust: Implement path resolution in QL #18579

Uh oh!

Conversation

hvitved commented Jan 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

paldepind left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

hvitved commented Jan 23, 2025 •

edited

Loading