Add Code And Bindings For Generating Test Case CheckSums (#691)

## Context 

Test Case IDs act as a checksum for tests - they're V5 UUID generated using aspects of the test like its filepath and name. If nothing about the test changes (including the variant used when running it), the ID remains fixed

## What Does The PR Do

Currently, generating the ID / Checksum for test runs is done [in the ETL when it runs](https://github.com/trunk-io/trunk/blob/5117073037075cc5f51c2bf59fe5422c8360c8fd/trunk/py/metrics_data_pipeline/trunk_etl_common/id_schema.py#L30). This PR moves that functionality over to rust bindings so that both the ETL and other processes (like our root cause AI agent) can use it

## Testing

This PR adds rust tests and tests for the TS and Python bindings. Also verified this against a recently uploaded test in staging, specifically https://app.trunk-staging.io/trunk-staging-org/flaky-tests/test/3f507aef-e834-523b-a8ad-edaba6b137be?repo=trunk-io%2Ftrunk (see actual rust test in `id.rs`

Author: pv72895

bundle/src/types.rs

@@ -1,4 +1,4 @@
-use context::repo::RepoUrlParts;
+use context::{meta::id::gen_info_id, repo::RepoUrlParts};
 #[cfg(feature = "pyo3")]
 use pyo3::prelude::*;
 #[cfg(feature = "pyo3")]
@@ -57,17 +57,16 @@ impl Test {
     }
 
     pub fn set_id<T: AsRef<str>>(&mut self, org_slug: T, repo: &RepoUrlParts) {
-        let info_id_input = [
+        self.id = gen_info_id(
             org_slug.as_ref(),
             repo.repo_full_name().as_str(),
-            self.file.as_deref().unwrap_or(""),
-            self.class_name.as_deref().unwrap_or(""),
-            &self.parent_name,
-            &self.name,
-            "JUNIT_TESTCASE",
-        ]
-        .join("#");
-        self.id = Uuid::new_v5(&Uuid::NAMESPACE_URL, info_id_input.as_bytes()).to_string()
+            self.file.as_deref(),
+            self.class_name.as_deref(),
+            Some(self.parent_name.as_str()),
+            Some(self.name.as_str()),
+            None,
+            "",
+        );
     }
 
     pub fn generate_custom_uuid<T: AsRef<str>>(&mut self, org_slug: T, repo: &RepoUrlParts, id: T) {

context-js/src/lib.rs

@@ -6,7 +6,7 @@ use bundle::{
     parse_meta_from_tarball as parse_tarball_meta, FileSetTestRunnerReport, VersionedBundle,
     VersionedBundleWithBindingsReport,
 };
-use context::{env, junit, repo};
+use context::{env, junit, meta::id::gen_info_id as gen_info_id_impl, repo};
 use futures::{future::Either, io::BufReader as BufReaderAsync, stream::TryStreamExt};
 use js_sys::Uint8Array;
 use prost::Message;
@@ -175,3 +175,27 @@ pub async fn parse_internal_bin_and_meta_from_tarball(
         versioned_bundle,
     })
 }
+
+#[wasm_bindgen]
+// trunk-ignore(clippy/too_many_arguments)
+pub fn gen_info_id(
+    org_url_slug: String,
+    repo_full_name: String,
+    file: Option<String>,
+    classname: Option<String>,
+    parent_fact_path: Option<String>,
+    name: Option<String>,
+    info_id: Option<String>,
+    variant: String,
+) -> String {
+    gen_info_id_impl(
+        &org_url_slug,
+        &repo_full_name,
+        file.as_deref(),
+        classname.as_deref(),
+        parent_fact_path.as_deref(),
+        name.as_deref(),
+        info_id.as_deref(),
+        &variant,
+    )
+}

context-js/tests/meta.test.ts

@@ -0,0 +1,97 @@
+import { describe, expect, it } from "vitest";
+import { gen_info_id } from "../pkg/context_js";
+
+describe("context-js", () => {
+  // These tests match the tests in context/src/meta/id.rs.
+  // While they don't need to match, it proves both the bindings and
+  // rust code are generating the same IDs.
+  describe("gen_info_id", () => {
+    it("generates ID properly for trunk", () => {
+      expect.hasAssertions();
+
+      const generateIdForTest = () =>
+        gen_info_id(
+          "example_org",
+          "example_repo",
+          "src/lib.rs",
+          "ExampleClass",
+          "parent/fact/path",
+          "example_name",
+          "trunk:12345",
+          "unix",
+        );
+
+      const result = generateIdForTest();
+
+      expect(result).toBe("4392f63c-8dc9-5cec-bbdc-e7b90c2e5a6b");
+
+      // Generate again to ensure it is consistent
+      const result2 = generateIdForTest();
+
+      expect(result2).toBe(result);
+    });
+
+    it("works properly with existing v5 UUID", () => {
+      expect.hasAssertions();
+
+      const existingInfoId = "a6e84936-3ee9-57d5-b041-ae124896f654";
+      const generateIdForTest = ({ variant = "" }: { variant?: string }) =>
+        gen_info_id(
+          "example_org",
+          "example_repo",
+          "src/lib.rs",
+          "ExampleClass",
+          "parent/fact/path",
+          "example_name",
+          existingInfoId,
+          variant,
+        );
+
+      const result = generateIdForTest({});
+
+      expect(result).toBe(existingInfoId);
+
+      // Generate again to ensure it is consistent
+      const result2 = generateIdForTest({});
+
+      expect(result2).toBe(result);
+
+      // Adding a variant changs the ID.
+      const resultWithVariant = generateIdForTest({ variant: "unix" });
+
+      expect(resultWithVariant).toBe("8057218b-95e4-5373-afbe-c366d4058615");
+    });
+
+    it("works properly without existing v5 UUID", () => {
+      expect.hasAssertions();
+
+      const generateIdForTest = ({ infoId }: { infoId?: string }) =>
+        gen_info_id(
+          "example_org",
+          "example_repo",
+          "src/lib.rs",
+          "ExampleClass",
+          "parent/fact/path",
+          "example_name",
+          infoId,
+          "unix",
+        );
+
+      const result = generateIdForTest({});
+
+      expect(result).toBe("c869cb93-66e2-516d-a0ea-15ff4b413c3f");
+
+      // Generate again to ensure it is consistent
+      const result2 = generateIdForTest({});
+
+      expect(result2).toBe(result);
+
+      // Existing UUID is ignored if it isn't V5
+      const resultForV4Uuid = generateIdForTest({
+        infoId: "08e1c642-3a55-45cf-8bf9-b9d0b21785dd", // V4
+      });
+
+      expect(resultForV4Uuid).toBe(result);
+    });
+  });
+});

context-py/src/lib.rs

@@ -10,7 +10,8 @@ use codeowners::{
 use context::{
     env,
     junit::{self, junit_path::TestRunnerReport},
-    meta, repo,
+    meta::{bindings, id, validator},
+    repo,
 };
 use prost::Message;
 use pyo3::{exceptions::PyTypeError, prelude::*};
@@ -209,16 +210,14 @@ pub fn parse_meta(meta_bytes: Vec<u8>) -> PyResult<bundle::BindingsVersionedBund
 
 #[gen_stub_pyfunction]
 #[pyfunction]
-fn meta_validate(
-    meta_context: meta::bindings::BindingsMetaContext,
-) -> meta::validator::MetaValidation {
-    meta::validator::validate(&meta::bindings::BindingsMetaContext::into(meta_context))
+fn meta_validate(meta_context: bindings::BindingsMetaContext) -> validator::MetaValidation {
+    validator::validate(&bindings::BindingsMetaContext::into(meta_context))
 }
 
 #[gen_stub_pyfunction]
 #[pyfunction]
 fn meta_validation_level_to_string(
-    meta_validation_level: meta::validator::MetaValidationLevel,
+    meta_validation_level: validator::MetaValidationLevel,
 ) -> String {
     meta_validation_level.to_string()
 }
@@ -362,6 +361,32 @@ fn associate_codeowners_multithreaded_impl(
     Ok(results)
 }
 
+#[gen_stub_pyfunction]
+#[pyfunction]
+// trunk-ignore(clippy/too_many_arguments)
+// trunk-ignore(clippy/deprecated)
+pub fn gen_info_id(
+    org_url_slug: String,
+    repo_full_name: String,
+    variant: String,
+    file: Option<String>,
+    classname: Option<String>,
+    parent_fact_path: Option<String>,
+    name: Option<String>,
+    info_id: Option<String>,
+) -> String {
+    id::gen_info_id(
+        &org_url_slug,
+        &repo_full_name,
+        file.as_deref(),
+        classname.as_deref(),
+        parent_fact_path.as_deref(),
+        name.as_deref(),
+        info_id.as_deref(),
+        &variant,
+    )
+}
+
 #[pymodule]
 fn context_py(m: &Bound<'_, PyModule>) -> PyResult<()> {
     m.add_class::<env::parser::CIInfo>()?;
@@ -402,9 +427,9 @@ fn context_py(m: &Bound<'_, PyModule>) -> PyResult<()> {
     m.add_function(wrap_pyfunction!(repo_validate, m)?)?;
     m.add_function(wrap_pyfunction!(repo_validation_level_to_string, m)?)?;
 
-    m.add_class::<meta::bindings::BindingsMetaContext>()?;
-    m.add_class::<meta::validator::MetaValidation>()?;
-    m.add_class::<meta::validator::MetaValidationLevel>()?;
+    m.add_class::<bindings::BindingsMetaContext>()?;
+    m.add_class::<validator::MetaValidation>()?;
+    m.add_class::<validator::MetaValidationLevel>()?;
     m.add_function(wrap_pyfunction!(parse_meta_from_tarball, m)?)?;
     m.add_function(wrap_pyfunction!(parse_meta, m)?)?;
     m.add_class::<bundle::BindingsVersionedBundle>()?;
@@ -417,6 +442,7 @@ fn context_py(m: &Bound<'_, PyModule>) -> PyResult<()> {
     m.add_function(wrap_pyfunction!(meta_validate, m)?)?;
     m.add_function(wrap_pyfunction!(meta_validation_level_to_string, m)?)?;
     m.add_function(wrap_pyfunction!(parse_internal_bin_from_tarball, m)?)?;
+    m.add_function(wrap_pyfunction!(gen_info_id, m)?)?;
 
     m.add_class::<codeowners::BindingsOwners>()?;
     m.add_function(wrap_pyfunction!(codeowners_parse, m)?)?;

context-py/tests/test_meta.py

@@ -0,0 +1,76 @@
+from context_py import gen_info_id
+
+
+def test_generates_id_properly_for_trunk():
+    def generate_id_for_test():
+        return gen_info_id(
+            "example_org",
+            "example_repo",
+            "unix",
+            "src/lib.rs",
+            "ExampleClass",
+            "parent/fact/path",
+            "example_name",
+            "trunk:12345",
+        )
+
+    result = generate_id_for_test()
+    assert result == "4392f63c-8dc9-5cec-bbdc-e7b90c2e5a6b"
+
+    # Generate again to ensure it is consistent
+    result2 = generate_id_for_test()
+    assert result2 == result
+
+
+def test_works_properly_with_existing_v5_uuid():
+    existing_info_id = "a6e84936-3ee9-57d5-b041-ae124896f654"
+
+    def generate_id_for_test(variant: str = ""):
+        return gen_info_id(
+            "example_org",
+            "example_repo",
+            variant,
+            "src/lib.rs",
+            "ExampleClass",
+            "parent/fact/path",
+            "example_name",
+            existing_info_id,
+        )
+
+    result = generate_id_for_test()
+    assert result == existing_info_id
+
+    # Generate again to ensure it is consistent
+    result2 = generate_id_for_test()
+    assert result2 == result
+
+    # Adding a variant changes the ID
+    result_with_variant = generate_id_for_test(variant="unix")
+    assert result_with_variant == "8057218b-95e4-5373-afbe-c366d4058615"
+
+
+def test_works_properly_without_existing_v5_uuid():
+    def generate_id_for_test(info_id: str | None = None):
+        return gen_info_id(
+            "example_org",
+            "example_repo",
+            "unix",
+            "src/lib.rs",
+            "ExampleClass",
+            "parent/fact/path",
+            "example_name",
+            info_id,
+        )
+
+    result = generate_id_for_test()
+    assert result == "c869cb93-66e2-516d-a0ea-15ff4b413c3f"
+
+    # Generate again to ensure it is consistent
+    result2 = generate_id_for_test()
+    assert result2 == result
+
+    # Existing UUID is ignored if it isn't V5
+    result_for_v4_uuid = generate_id_for_test(
+        info_id="08e1c642-3a55-45cf-8bf9-b9d0b21785dd"
+    )  # V4
+    assert result_for_v4_uuid == result