test_musehub_topics.py
python
| 1 | """Tests for the MuseHub topics/tag browse API endpoints. |
| 2 | |
| 3 | Covers acceptance criteria: |
| 4 | - test_list_topics_empty — no public repos → empty topics list |
| 5 | - test_list_topics_aggregates_counts — counts reflect public repos only |
| 6 | - test_list_topics_excludes_private_repos — private repo tags are not counted |
| 7 | - test_list_topics_sorted_by_count_desc — most popular topic appears first |
| 8 | - test_repos_by_topic_empty — unknown tag → empty list (not 404) |
| 9 | - test_repos_by_topic_returns_tagged_repos — only repos with exact tag returned |
| 10 | - test_repos_by_topic_excludes_private — private repos are hidden |
| 11 | - test_repos_by_topic_sort_by_stars — stars sort returns most-starred first |
| 12 | - test_repos_by_topic_sort_by_updated — updated sort returns most-recently-committed first |
| 13 | - test_repos_by_topic_invalid_sort — invalid sort param returns 422 |
| 14 | - test_repos_by_topic_pagination — page 2 returns different repos |
| 15 | - test_set_topics_requires_auth — POST without JWT returns 401 |
| 16 | - test_set_topics_owner_only — non-owner gets 403 |
| 17 | - test_set_topics_replaces_list — new list replaces old list entirely |
| 18 | - test_set_topics_deduplicates — duplicate slugs are collapsed |
| 19 | - test_set_topics_invalid_slug — bad slug characters return 422 |
| 20 | - test_set_topics_too_many — more than 20 topics returns 422 |
| 21 | - test_set_topics_clears_list — empty body clears all topics |
| 22 | - test_set_topics_repo_not_found — unknown repo_id returns 404 |
| 23 | """ |
| 24 | from __future__ import annotations |
| 25 | |
| 26 | import re |
| 27 | |
| 28 | import pytest |
| 29 | from httpx import AsyncClient |
| 30 | from sqlalchemy.ext.asyncio import AsyncSession |
| 31 | |
| 32 | from musehub.db.musehub_models import MusehubCommit, MusehubRepo, MusehubStar |
| 33 | |
| 34 | |
| 35 | # --------------------------------------------------------------------------- |
| 36 | # Helpers |
| 37 | # --------------------------------------------------------------------------- |
| 38 | |
| 39 | |
| 40 | async def _make_repo( |
| 41 | db_session: AsyncSession, |
| 42 | *, |
| 43 | name: str, |
| 44 | visibility: str = "public", |
| 45 | tags: list[str] | None = None, |
| 46 | owner: str = "testuser", |
| 47 | owner_user_id: str = "test-owner", |
| 48 | ) -> str: |
| 49 | """Seed a repo and return its repo_id.""" |
| 50 | slug = re.sub(r"[^a-z0-9]+", "-", name.lower()).strip("-")[:64].strip("-") or "repo" |
| 51 | repo = MusehubRepo( |
| 52 | name=name, |
| 53 | owner=owner, |
| 54 | slug=f"{slug}-{visibility[:3]}", |
| 55 | visibility=visibility, |
| 56 | owner_user_id=owner_user_id, |
| 57 | description="", |
| 58 | tags=tags or [], |
| 59 | ) |
| 60 | db_session.add(repo) |
| 61 | await db_session.commit() |
| 62 | await db_session.refresh(repo) |
| 63 | return str(repo.repo_id) |
| 64 | |
| 65 | |
| 66 | async def _add_star(db_session: AsyncSession, repo_id: str, user_id: str) -> None: |
| 67 | star = MusehubStar(repo_id=repo_id, user_id=user_id) |
| 68 | db_session.add(star) |
| 69 | await db_session.commit() |
| 70 | |
| 71 | |
| 72 | async def _add_commit( |
| 73 | db_session: AsyncSession, |
| 74 | repo_id: str, |
| 75 | *, |
| 76 | sha: str, |
| 77 | timestamp: str, |
| 78 | ) -> None: |
| 79 | from datetime import datetime, timezone |
| 80 | commit = MusehubCommit( |
| 81 | commit_id=sha, |
| 82 | repo_id=repo_id, |
| 83 | branch="main", |
| 84 | author="tester", |
| 85 | message="test commit", |
| 86 | timestamp=datetime.fromisoformat(timestamp).replace(tzinfo=timezone.utc), |
| 87 | parent_ids=[], |
| 88 | ) |
| 89 | db_session.add(commit) |
| 90 | await db_session.commit() |
| 91 | |
| 92 | |
| 93 | # --------------------------------------------------------------------------- |
| 94 | # GET /api/v1/topics |
| 95 | # --------------------------------------------------------------------------- |
| 96 | |
| 97 | |
| 98 | @pytest.mark.anyio |
| 99 | async def test_list_topics_empty(client: AsyncClient) -> None: |
| 100 | """No public repos → topics list is empty.""" |
| 101 | response = await client.get("/api/v1/topics") |
| 102 | assert response.status_code == 200 |
| 103 | assert response.json() == {"topics": []} |
| 104 | |
| 105 | |
| 106 | @pytest.mark.anyio |
| 107 | async def test_list_topics_aggregates_counts( |
| 108 | client: AsyncClient, db_session: AsyncSession |
| 109 | ) -> None: |
| 110 | """Topics are aggregated across all public repos with correct counts.""" |
| 111 | await _make_repo(db_session, name="repo-a", tags=["jazz", "piano"]) |
| 112 | await _make_repo(db_session, name="repo-b", tags=["jazz", "ambient"]) |
| 113 | await _make_repo(db_session, name="repo-c", tags=["ambient"]) |
| 114 | |
| 115 | response = await client.get("/api/v1/topics") |
| 116 | assert response.status_code == 200 |
| 117 | |
| 118 | topics = {t["name"]: t["repo_count"] for t in response.json()["topics"]} |
| 119 | assert topics["jazz"] == 2 |
| 120 | assert topics["ambient"] == 2 |
| 121 | assert topics["piano"] == 1 |
| 122 | |
| 123 | |
| 124 | @pytest.mark.anyio |
| 125 | async def test_list_topics_excludes_private_repos( |
| 126 | client: AsyncClient, db_session: AsyncSession |
| 127 | ) -> None: |
| 128 | """Private repo tags do not contribute to topic counts.""" |
| 129 | await _make_repo(db_session, name="pub-jazz", tags=["jazz"], visibility="public") |
| 130 | await _make_repo(db_session, name="priv-jazz", tags=["jazz", "secret-tag"], visibility="private") |
| 131 | |
| 132 | response = await client.get("/api/v1/topics") |
| 133 | assert response.status_code == 200 |
| 134 | |
| 135 | topics = {t["name"]: t["repo_count"] for t in response.json()["topics"]} |
| 136 | assert topics.get("jazz") == 1 # only the public repo |
| 137 | assert "secret-tag" not in topics |
| 138 | |
| 139 | |
| 140 | @pytest.mark.anyio |
| 141 | async def test_list_topics_sorted_by_count_desc( |
| 142 | client: AsyncClient, db_session: AsyncSession |
| 143 | ) -> None: |
| 144 | """Topics are sorted by repo_count descending — most popular first.""" |
| 145 | await _make_repo(db_session, name="r1", tags=["baroque"]) |
| 146 | await _make_repo(db_session, name="r2", tags=["jazz", "baroque"]) |
| 147 | await _make_repo(db_session, name="r3", tags=["jazz", "baroque"]) |
| 148 | |
| 149 | response = await client.get("/api/v1/topics") |
| 150 | assert response.status_code == 200 |
| 151 | |
| 152 | topics = response.json()["topics"] |
| 153 | assert topics[0]["name"] == "baroque" # 3 repos |
| 154 | assert topics[1]["name"] == "jazz" # 2 repos |
| 155 | |
| 156 | |
| 157 | # --------------------------------------------------------------------------- |
| 158 | # GET /api/v1/topics/{tag}/repos |
| 159 | # --------------------------------------------------------------------------- |
| 160 | |
| 161 | |
| 162 | @pytest.mark.anyio |
| 163 | async def test_repos_by_topic_empty(client: AsyncClient) -> None: |
| 164 | """Unknown/unused tag → empty repos list, not 404.""" |
| 165 | response = await client.get("/api/v1/topics/nonexistent-tag/repos") |
| 166 | assert response.status_code == 200 |
| 167 | body = response.json() |
| 168 | assert body["repos"] == [] |
| 169 | assert body["total"] == 0 |
| 170 | assert body["tag"] == "nonexistent-tag" |
| 171 | |
| 172 | |
| 173 | @pytest.mark.anyio |
| 174 | async def test_repos_by_topic_returns_tagged_repos( |
| 175 | client: AsyncClient, db_session: AsyncSession |
| 176 | ) -> None: |
| 177 | """Only repos with the exact tag are returned.""" |
| 178 | await _make_repo(db_session, name="jazz-repo", tags=["jazz", "piano"]) |
| 179 | await _make_repo(db_session, name="piano-only-repo", tags=["piano"]) |
| 180 | await _make_repo(db_session, name="unrelated-repo", tags=["ambient"]) |
| 181 | |
| 182 | response = await client.get("/api/v1/topics/jazz/repos") |
| 183 | assert response.status_code == 200 |
| 184 | body = response.json() |
| 185 | assert body["total"] == 1 |
| 186 | assert body["tag"] == "jazz" |
| 187 | assert body["repos"][0]["name"] == "jazz-repo" |
| 188 | |
| 189 | |
| 190 | @pytest.mark.anyio |
| 191 | async def test_repos_by_topic_excludes_private( |
| 192 | client: AsyncClient, db_session: AsyncSession |
| 193 | ) -> None: |
| 194 | """Private repos are not exposed even when they carry the tag.""" |
| 195 | await _make_repo(db_session, name="pub", tags=["classical"], visibility="public") |
| 196 | await _make_repo(db_session, name="priv", tags=["classical"], visibility="private") |
| 197 | |
| 198 | response = await client.get("/api/v1/topics/classical/repos") |
| 199 | assert response.status_code == 200 |
| 200 | assert response.json()["total"] == 1 # only the public repo |
| 201 | |
| 202 | |
| 203 | @pytest.mark.anyio |
| 204 | async def test_repos_by_topic_sort_by_stars( |
| 205 | client: AsyncClient, db_session: AsyncSession |
| 206 | ) -> None: |
| 207 | """sort=stars returns most-starred repo first.""" |
| 208 | id_low = await _make_repo(db_session, name="low-star", tags=["edm"]) |
| 209 | id_high = await _make_repo(db_session, name="high-star", tags=["edm"]) |
| 210 | |
| 211 | await _add_star(db_session, id_high, "user1") |
| 212 | await _add_star(db_session, id_high, "user2") |
| 213 | await _add_star(db_session, id_low, "user3") |
| 214 | |
| 215 | response = await client.get("/api/v1/topics/edm/repos?sort=stars") |
| 216 | assert response.status_code == 200 |
| 217 | names = [r["name"] for r in response.json()["repos"]] |
| 218 | assert names.index("high-star") < names.index("low-star") |
| 219 | |
| 220 | |
| 221 | @pytest.mark.anyio |
| 222 | async def test_repos_by_topic_sort_by_updated( |
| 223 | client: AsyncClient, db_session: AsyncSession |
| 224 | ) -> None: |
| 225 | """sort=updated returns most-recently-committed repo first.""" |
| 226 | id_old = await _make_repo(db_session, name="old-commits", tags=["ambient"]) |
| 227 | id_new = await _make_repo(db_session, name="new-commits", tags=["ambient"]) |
| 228 | |
| 229 | await _add_commit(db_session, id_old, sha="sha-old", timestamp="2023-01-01T00:00:00") |
| 230 | await _add_commit(db_session, id_new, sha="sha-new", timestamp="2024-06-01T00:00:00") |
| 231 | |
| 232 | response = await client.get("/api/v1/topics/ambient/repos?sort=updated") |
| 233 | assert response.status_code == 200 |
| 234 | names = [r["name"] for r in response.json()["repos"]] |
| 235 | assert names.index("new-commits") < names.index("old-commits") |
| 236 | |
| 237 | |
| 238 | @pytest.mark.anyio |
| 239 | async def test_repos_by_topic_invalid_sort(client: AsyncClient, db_session: AsyncSession) -> None: |
| 240 | """Invalid sort parameter returns 422.""" |
| 241 | await _make_repo(db_session, name="any-repo", tags=["jazz"]) |
| 242 | response = await client.get("/api/v1/topics/jazz/repos?sort=invalid") |
| 243 | assert response.status_code == 422 |
| 244 | |
| 245 | |
| 246 | @pytest.mark.anyio |
| 247 | async def test_repos_by_topic_pagination( |
| 248 | client: AsyncClient, db_session: AsyncSession |
| 249 | ) -> None: |
| 250 | """Pagination works: page 2 returns a different set of repos.""" |
| 251 | for i in range(5): |
| 252 | await _make_repo(db_session, name=f"cinematic-{i}", tags=["cinematic"]) |
| 253 | |
| 254 | page1 = await client.get("/api/v1/topics/cinematic/repos?page=1&page_size=2") |
| 255 | page2 = await client.get("/api/v1/topics/cinematic/repos?page=2&page_size=2") |
| 256 | |
| 257 | assert page1.status_code == 200 |
| 258 | assert page2.status_code == 200 |
| 259 | ids1 = {r["repoId"] for r in page1.json()["repos"]} |
| 260 | ids2 = {r["repoId"] for r in page2.json()["repos"]} |
| 261 | assert ids1.isdisjoint(ids2) |
| 262 | assert page1.json()["total"] == 5 |
| 263 | |
| 264 | |
| 265 | # --------------------------------------------------------------------------- |
| 266 | # POST /api/v1/repos/{repo_id}/topics |
| 267 | # --------------------------------------------------------------------------- |
| 268 | |
| 269 | |
| 270 | @pytest.mark.anyio |
| 271 | async def test_set_topics_requires_auth( |
| 272 | client: AsyncClient, db_session: AsyncSession |
| 273 | ) -> None: |
| 274 | """POST without a JWT returns 401.""" |
| 275 | repo_id = await _make_repo(db_session, name="auth-test") |
| 276 | response = await client.post( |
| 277 | f"/api/v1/repos/{repo_id}/topics", |
| 278 | json={"topics": ["jazz"]}, |
| 279 | ) |
| 280 | assert response.status_code == 401 |
| 281 | |
| 282 | |
| 283 | @pytest.mark.anyio |
| 284 | async def test_set_topics_owner_only( |
| 285 | client: AsyncClient, db_session: AsyncSession, auth_headers: dict[str, str] |
| 286 | ) -> None: |
| 287 | """A user who is not the repo owner receives 403.""" |
| 288 | repo_id = await _make_repo(db_session, name="owned-elsewhere", owner_user_id="different-owner") |
| 289 | response = await client.post( |
| 290 | f"/api/v1/repos/{repo_id}/topics", |
| 291 | json={"topics": ["jazz"]}, |
| 292 | headers=auth_headers, |
| 293 | ) |
| 294 | assert response.status_code == 403 |
| 295 | |
| 296 | |
| 297 | @pytest.mark.anyio |
| 298 | async def test_set_topics_replaces_list( |
| 299 | client: AsyncClient, db_session: AsyncSession, auth_headers: dict[str, str] |
| 300 | ) -> None: |
| 301 | """Posting a new list replaces the existing tags entirely.""" |
| 302 | repo_id = await _make_repo( |
| 303 | db_session, |
| 304 | name="replace-me", |
| 305 | tags=["old-tag"], |
| 306 | owner_user_id="550e8400-e29b-41d4-a716-446655440000", |
| 307 | ) |
| 308 | response = await client.post( |
| 309 | f"/api/v1/repos/{repo_id}/topics", |
| 310 | json={"topics": ["jazz", "piano"]}, |
| 311 | headers=auth_headers, |
| 312 | ) |
| 313 | assert response.status_code == 200 |
| 314 | body = response.json() |
| 315 | assert body["repo_id"] == repo_id |
| 316 | assert body["topics"] == ["jazz", "piano"] |
| 317 | |
| 318 | |
| 319 | @pytest.mark.anyio |
| 320 | async def test_set_topics_deduplicates( |
| 321 | client: AsyncClient, db_session: AsyncSession, auth_headers: dict[str, str] |
| 322 | ) -> None: |
| 323 | """Duplicate topic slugs in the request are silently collapsed.""" |
| 324 | repo_id = await _make_repo( |
| 325 | db_session, |
| 326 | name="dedup-test", |
| 327 | owner_user_id="550e8400-e29b-41d4-a716-446655440000", |
| 328 | ) |
| 329 | response = await client.post( |
| 330 | f"/api/v1/repos/{repo_id}/topics", |
| 331 | json={"topics": ["jazz", "jazz", "piano", "jazz"]}, |
| 332 | headers=auth_headers, |
| 333 | ) |
| 334 | assert response.status_code == 200 |
| 335 | assert response.json()["topics"] == ["jazz", "piano"] |
| 336 | |
| 337 | |
| 338 | @pytest.mark.anyio |
| 339 | async def test_set_topics_invalid_slug( |
| 340 | client: AsyncClient, db_session: AsyncSession, auth_headers: dict[str, str] |
| 341 | ) -> None: |
| 342 | """Topic slugs with invalid characters return 422.""" |
| 343 | repo_id = await _make_repo( |
| 344 | db_session, |
| 345 | name="slug-test", |
| 346 | owner_user_id="550e8400-e29b-41d4-a716-446655440000", |
| 347 | ) |
| 348 | response = await client.post( |
| 349 | f"/api/v1/repos/{repo_id}/topics", |
| 350 | json={"topics": ["Valid-slug", "BAD SLUG!", "ok-slug"]}, |
| 351 | headers=auth_headers, |
| 352 | ) |
| 353 | assert response.status_code == 422 |
| 354 | |
| 355 | |
| 356 | @pytest.mark.anyio |
| 357 | async def test_set_topics_too_many( |
| 358 | client: AsyncClient, db_session: AsyncSession, auth_headers: dict[str, str] |
| 359 | ) -> None: |
| 360 | """Submitting more than 20 topics returns 422.""" |
| 361 | repo_id = await _make_repo( |
| 362 | db_session, |
| 363 | name="too-many", |
| 364 | owner_user_id="550e8400-e29b-41d4-a716-446655440000", |
| 365 | ) |
| 366 | many_topics = [f"topic-{i}" for i in range(21)] |
| 367 | response = await client.post( |
| 368 | f"/api/v1/repos/{repo_id}/topics", |
| 369 | json={"topics": many_topics}, |
| 370 | headers=auth_headers, |
| 371 | ) |
| 372 | assert response.status_code == 422 |
| 373 | |
| 374 | |
| 375 | @pytest.mark.anyio |
| 376 | async def test_set_topics_clears_list( |
| 377 | client: AsyncClient, db_session: AsyncSession, auth_headers: dict[str, str] |
| 378 | ) -> None: |
| 379 | """Sending an empty list removes all topics.""" |
| 380 | repo_id = await _make_repo( |
| 381 | db_session, |
| 382 | name="clear-me", |
| 383 | tags=["jazz", "piano"], |
| 384 | owner_user_id="550e8400-e29b-41d4-a716-446655440000", |
| 385 | ) |
| 386 | response = await client.post( |
| 387 | f"/api/v1/repos/{repo_id}/topics", |
| 388 | json={"topics": []}, |
| 389 | headers=auth_headers, |
| 390 | ) |
| 391 | assert response.status_code == 200 |
| 392 | assert response.json()["topics"] == [] |
| 393 | |
| 394 | |
| 395 | @pytest.mark.anyio |
| 396 | async def test_set_topics_repo_not_found( |
| 397 | client: AsyncClient, auth_headers: dict[str, str] |
| 398 | ) -> None: |
| 399 | """Unknown repo_id returns 404.""" |
| 400 | response = await client.post( |
| 401 | "/api/v1/repos/nonexistent-repo-id/topics", |
| 402 | json={"topics": ["jazz"]}, |
| 403 | headers=auth_headers, |
| 404 | ) |
| 405 | assert response.status_code == 404 |