fix(integrations): Cache missing repo tree lookups

Cache GitHub repo tree 404 responses as empty results to avoid repeated
failing requests from auto source code config. Keep non-404 API errors
raising normally and cover both behaviors with tests.

Fixes SENTRY-5K7G
Co-Authored-By: Codex <noreply@openai.com>

Made-with: Cursor

Author: armenzg

src/sentry/integrations/source_code_management/repo_trees.py

@@ -35,6 +35,7 @@ class RepoTree(NamedTuple):
 # When the number of remaining API requests is less than this value, it will
 # fall back to the cache.
 MINIMUM_REQUESTS_REMAINING = 200
+NOT_FOUND_CACHE_SECONDS = 3600 * 24
 
 
 class RepoTreesIntegration(ABC):
@@ -211,6 +212,16 @@ def get_cached_repo_files(
                 )
                 cache.set(key, [], self.CACHE_SECONDS + shifted_seconds)
                 tree = None
+            except ApiError as error:
+                if _is_not_found_error(error):
+                    logger.info(
+                        "Caching empty files result for missing repo or ref",
+                        extra={"repo": repo_full_name},
+                    )
+                    cache.set(key, [], NOT_FOUND_CACHE_SECONDS)
+                    tree = None
+                else:
+                    raise
             if tree:
                 # Keep files; discard directories
                 repo_files = [node["path"] for node in tree if node["type"] == "blob"]
@@ -296,3 +307,11 @@ def should_include(file_path: str) -> bool:
     if any(file_path.startswith(path) for path in EXCLUDED_PATHS):
         return False
     return True
+
+
+def _is_not_found_error(error: ApiError) -> bool:
+    if error.code == 404:
+        return True
+
+    error_message = error.json.get("message") if error.json else error.text
+    return error_message in ("Not Found", "Not Found.")

tests/sentry/integrations/github/test_client.py

@@ -379,6 +379,38 @@ def test_get_cached_repo_files_with_all_files(self) -> None:
             files = self.install.get_cached_repo_files(self.repo.name, "master", 0)
             assert files == ["src/foo.py"]
 
+    @responses.activate
+    def test_get_cached_repo_files_caches_not_found(self) -> None:
+        responses.add(
+            method=responses.GET,
+            url=f"https://api.github.com/repos/{self.repo.name}/git/trees/master?recursive=1",
+            status=404,
+            json={"message": "Not Found"},
+        )
+        repo_key = f"github:repo:{self.repo.name}:source-code"
+        assert cache.get(repo_key) is None
+
+        files = self.install.get_cached_repo_files(self.repo.name, "master", 0)
+        assert files == []
+        assert cache.get(repo_key) == []
+
+        # Negative-cache hit should avoid an additional API request.
+        files = self.install.get_cached_repo_files(self.repo.name, "master", 0)
+        assert files == []
+        assert len(responses.calls) == 1
+
+    @responses.activate
+    def test_get_cached_repo_files_raises_non_not_found_api_error(self) -> None:
+        responses.add(
+            method=responses.GET,
+            url=f"https://api.github.com/repos/{self.repo.name}/git/trees/master?recursive=1",
+            status=500,
+            json={"message": "Server Error"},
+        )
+
+        with pytest.raises(ApiError):
+            self.install.get_cached_repo_files(self.repo.name, "master", 0)
+
     @mock.patch("sentry.integrations.github.client.get_jwt", return_value="jwt_token_1")
     @responses.activate
     def test_update_comment(self, get_jwt) -> None: