test: improve unit test coverage toward 80% target

.github/workflows/release.yml

@@ -2,7 +2,7 @@ name: Distribution
 
 on:
   push:
-    branches-ignore: [renovate/**]
+    branches: [main, dev]
     tags: [ontokit-*]
   pull_request:
 
@@ -23,6 +23,32 @@ jobs:
     runs-on: ubuntu-latest
     permissions:
       contents: read
+    services:
+      postgres:
+        image: pgvector/pgvector:pg17
+        env:
+          POSTGRES_USER: ontokit_test
+          POSTGRES_PASSWORD: ontokit_test
+          POSTGRES_DB: ontokit_test
+        ports:
+          - 5432:5432
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+      redis:
+        image: redis:7
+        ports:
+          - 6379:6379
+        options: >-
+          --health-cmd "redis-cli ping"
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+    env:
+      DATABASE_URL: postgresql+asyncpg://ontokit_test:ontokit_test@localhost:5432/ontokit_test
+      REDIS_URL: redis://localhost:6379/0
     steps:
       - uses: actions/checkout@v4
       - uses: astral-sh/setup-uv@v4

.pre-commit-config.yaml

@@ -32,3 +32,4 @@ repos:
           - sentence-transformers>=3.0.0
           - cryptography>=42.0.0
           - pytest>=8.0.0
+          - pytest-asyncio>=0.24.0

docs/coverage-plan.md

@@ -0,0 +1,62 @@
+# Test Coverage Plan: 78% → 80%
+
+**Created:** 2026-04-08
+**Updated:** 2026-04-08
+**Baseline:** 78% (7502/9571 statements covered, 983 tests)
+**Target:** 80% (7657 statements covered, ~155 more needed)
+
+## Completed
+
+The following Phase 1 items have been completed:
+
+| File | Before | After | Tests Added |
+|------|--------|-------|-------------|
+| `services/project_service.py` | 55% | 94% | ~40 |
+| `services/suggestion_service.py` | 39% | 96% | ~51 |
+| `services/embedding_service.py` | 33% | 99% | ~33 |
+
+## Phase 1 — Remaining (~170 statements recoverable)
+
+| File | Current | Missed | Target | To Recover |
+|------|---------|--------|--------|------------|
+| `services/pull_request_service.py` | 56% | 305 | 80% | ~170 |
+
+### pull_request_service.py (56% → 80%)
+- [ ] `create_pull_request()` — creation with validation
+- [ ] `merge_pull_request()` — merge strategies
+- [ ] `close_pull_request()`, `reopen_pull_request()`
+- [ ] Review CRUD: `create_review()`, `list_reviews()`
+- [ ] Comment CRUD: `create_comment()`, `list_comments()`, `update_comment()`, `delete_comment()`
+- [ ] Branch management: `list_branches()`, `create_branch()`
+- [ ] GitHub integration: `create_github_integration()`, `update_github_integration()`, `delete_github_integration()`
+- [ ] Webhook handlers: `handle_github_pr_webhook()`, `handle_github_review_webhook()`, `handle_github_push_webhook()`
+- [ ] PR settings: `get_pr_settings()`, `update_pr_settings()`
+
+Covering ~155 of the 305 missed statements reaches 80% overall.
+
+## Phase 2 — Medium Impact (~250 statements)
+
+| File | Current | Missed | Target | To Recover |
+|------|---------|--------|--------|------------|
+| `git/bare_repository.py` | 70% | 150 | 80% | ~55 |
+| `worker.py` | 70% | 111 | 80% | ~40 |
+| `services/ontology_extractor.py` | 64% | 93 | 80% | ~45 |
+| `services/ontology_index.py` | 75% | 89 | 80% | ~25 |
+| `services/github_sync.py` | 61% | 46 | 80% | ~25 |
+| `services/indexed_ontology.py` | 44% | 50 | 80% | ~30 |
+| `services/normalization_service.py` | 73% | 25 | 80% | ~10 |
+| `services/embedding_providers/*` | 0-75% | ~108 | 80% | ~20 |
+
+## Phase 3 — Diminishing Returns
+
+| File | Current | Notes |
+|------|---------|-------|
+| `main.py` | 54% | Startup/lifespan — hard to unit test |
+| `runner.py` | 0% | 6 lines, CLI entry point |
+| `services/ontology.py` | 82% | Already above target |
+| `services/linter.py` | 80% | Already at target |
+
+## Execution Order
+
+1. `pull_request_service.py` — the only Phase 1 item remaining; ~155 statements gets us to 80%
+2. Phase 2 files as needed to build further margin

ontokit/api/routes/projects.py

@@ -540,6 +540,11 @@ async def _ensure_ontology_loaded(
         if git is not None and git.repository_exists(project_id):
             try:
                 await ontology.load_from_git(project_id, branch, filename, git)
+            except ValueError as e:
+                raise HTTPException(
+                    status_code=status.HTTP_422_UNPROCESSABLE_CONTENT,
+                    detail=str(e),
+                ) from e
             except Exception as e:
                 raise HTTPException(
                     status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
@@ -556,7 +561,7 @@ async def _ensure_ontology_loaded(
                 ) from e
             except ValueError as e:
                 raise HTTPException(
-                    status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
+                    status_code=status.HTTP_422_UNPROCESSABLE_CONTENT,
                     detail=str(e),
                 ) from e
 
@@ -1274,7 +1279,7 @@ async def save_source_content(
         g.parse(data=data.content, format="turtle")
     except Exception as e:
         raise HTTPException(
-            status_code=status.HTTP_422_UNPROCESSABLE_ENTITY,
+            status_code=status.HTTP_422_UNPROCESSABLE_CONTENT,
             detail=f"Invalid Turtle syntax: {e}",
         ) from e
 

ontokit/collab/presence.py

@@ -1,6 +1,6 @@
 """User presence tracking for collaboration sessions."""
 
-from datetime import datetime, timedelta
+from datetime import UTC, datetime, timedelta
 
 from ontokit.collab.protocol import User
 
@@ -37,7 +37,7 @@ def join(self, room: str, user: User) -> list[User]:
         user.color = self._colors[user_count % len(self._colors)]
 
         self._rooms[room][user.user_id] = user
-        self._last_seen[user.user_id] = datetime.utcnow()
+        self._last_seen[user.user_id] = datetime.now(tz=UTC)
 
         return list(self._rooms[room].values())
 
@@ -57,23 +57,23 @@ def update_cursor(self, room: str, user_id: str, path: str) -> None:
         """Update user's cursor position."""
         if room in self._rooms and user_id in self._rooms[room]:
             self._rooms[room][user_id].cursor_path = path
-            self._last_seen[user_id] = datetime.utcnow()
+            self._last_seen[user_id] = datetime.now(tz=UTC)
 
     def get_users(self, room: str) -> list[User]:
         """Get all users in a room."""
         return list(self._rooms.get(room, {}).values())
 
     def heartbeat(self, user_id: str) -> None:
         """Update last seen timestamp for a user."""
-        self._last_seen[user_id] = datetime.utcnow()
+        self._last_seen[user_id] = datetime.now(tz=UTC)
 
     def cleanup_stale(self, timeout_minutes: int = 5) -> list[tuple[str, str]]:
         """
         Remove users who haven't been seen recently.
 
         Returns list of (room, user_id) tuples for removed users.
         """
-        cutoff = datetime.utcnow() - timedelta(minutes=timeout_minutes)
+        cutoff = datetime.now(tz=UTC) - timedelta(minutes=timeout_minutes)
         removed = []
 
         for room, users in list(self._rooms.items()):

ontokit/git/bare_repository.py

@@ -156,7 +156,7 @@ def _resolve_ref(self, ref: str) -> pygit2.Commit:
 
         # Try as partial hash
         try:
-            for commit in self.repo.walk(self.repo.head.target, pygit2.GIT_SORT_TIME):  # type: ignore[arg-type]
+            for commit in self.repo.walk(self.repo.head.target, pygit2.enums.SortMode.TIME):
                 if str(commit.id).startswith(ref):
                     return commit
         except Exception:
@@ -361,7 +361,10 @@ def get_history(
                 for ref_name in self.repo.references:
                     if ref_name.startswith("refs/heads/"):
                         ref = self.repo.references[ref_name]
-                        for commit in self.repo.walk(ref.target, pygit2.GIT_SORT_TIME):  # type: ignore[arg-type]
+                        for commit in self.repo.walk(
+                            ref.target,
+                            pygit2.enums.SortMode.TIME | pygit2.enums.SortMode.TOPOLOGICAL,
+                        ):
                             commit_hash = str(commit.id)
                             if commit_hash not in seen_hashes:
                                 seen_hashes.add(commit_hash)
@@ -379,7 +382,12 @@ def get_history(
                     target = self.repo.head.target
 
                 commit_iter = []
-                for count, commit in enumerate(self.repo.walk(target, pygit2.GIT_SORT_TIME)):  # type: ignore[arg-type]
+                for count, commit in enumerate(
+                    self.repo.walk(
+                        target,
+                        pygit2.enums.SortMode.TIME | pygit2.enums.SortMode.TOPOLOGICAL,
+                    )
+                ):
                     commit_iter.append(commit)
                     if count + 1 >= limit:
                         break
@@ -746,12 +754,12 @@ def get_commits_between(self, from_ref: str, to_ref: str = "HEAD") -> list[Commi
 
             # Get commits reachable from to_ref but not from from_ref
             from_ancestors = set()
-            for commit in self.repo.walk(from_commit.id, pygit2.GIT_SORT_TIME):  # type: ignore[arg-type]
+            for commit in self.repo.walk(from_commit.id, pygit2.enums.SortMode.TIME):
                 from_ancestors.add(str(commit.id))
 
-            for commit in self.repo.walk(to_commit.id, pygit2.GIT_SORT_TIME):  # type: ignore[arg-type]
+            for commit in self.repo.walk(to_commit.id, pygit2.enums.SortMode.TIME):
                 if str(commit.id) in from_ancestors:
-                    break
+                    continue
                 commits.append(self._commit_to_info(commit))
 
         except Exception: