logdetective 2.0.0__py3-none-any.whl → 2.0.2__py3-none-any.whl

logdetective/extractors.py

@@ -20,7 +20,8 @@ class DrainExtractor:
         context: bool = False,
         max_clusters=8,
         skip_snippets: SkipSnippets = SkipSnippets({}),
-    ):
+        max_snippet_len: int = 2000
+    ):  # pylint: disable=R0913,R0917
         config = TemplateMinerConfig()
         config.load(f"{os.path.dirname(__file__)}/drain3.ini")
         config.profiling_enabled = verbose
@@ -29,11 +30,12 @@ class DrainExtractor:
         self.verbose = verbose
         self.context = context
         self.skip_snippets = skip_snippets
+        self.max_snippet_len = max_snippet_len
 
     def __call__(self, log: str) -> list[Tuple[int, str]]:
         out = []
         # Create chunks
-        chunks = list(get_chunks(log))
+        chunks = list(get_chunks(log, self.max_snippet_len))
         # Keep only chunks that don't match any of the excluded patterns
         chunks = [
             (_, chunk)

logdetective/server/models.py

@@ -244,18 +244,18 @@ class InferenceConfig(BaseModel):  # pylint: disable=too-many-instance-attribute
 class ExtractorConfig(BaseModel):
     """Model for extractor configuration of logdetective server."""
 
-    context: bool = True
     max_clusters: int = 8
     verbose: bool = False
+    max_snippet_len: int = 2000
 
     def __init__(self, data: Optional[dict] = None):
         super().__init__()
         if data is None:
             return
 
-        self.context = data.get("context", True)
         self.max_clusters = data.get("max_clusters", 8)
         self.verbose = data.get("verbose", False)
+        self.max_snippet_len = data.get("max_snippet_len", 2000)
 
 
 class GitLabInstanceConfig(BaseModel):  # pylint: disable=too-many-instance-attributes

logdetective/server/server.py

@@ -117,23 +117,25 @@ def requires_token_when_set(authentication: Annotated[str | None, Header()] = No
         LOG.info("LOGDETECTIVE_TOKEN env var not set, authentication disabled")
         # no token required, means local dev environment
         return
-    token = None
     if authentication:
         try:
             token = authentication.split(" ", 1)[1]
-        except (ValueError, IndexError):
+        except (ValueError, IndexError) as ex:
             LOG.warning(
-                "Authentication header has invalid structure (%s), it should be 'Bearer TOKEN'",
+                "Authentication header has invalid structure '%s', it should be 'Bearer TOKEN'",
                 authentication,
             )
             # eat the exception and raise 401 below
-            token = None
+            raise HTTPException(
+                status_code=401,
+                detail=f"Invalid authentication, HEADER '{authentication}' not valid.",
+            ) from ex
         if token == API_TOKEN:
             return
-    LOG.info(
-        "LOGDETECTIVE_TOKEN env var is set (%s), clien token = %s", API_TOKEN, token
-    )
-    raise HTTPException(status_code=401, detail=f"Token {token} not valid.")
+        LOG.info("Provided token '%s' does not match expected value.", token)
+        raise HTTPException(status_code=401, detail=f"Token '{token}' not valid.")
+    LOG.error("No authentication header provided but LOGDETECTIVE_TOKEN env var is set")
+    raise HTTPException(status_code=401, detail="No token provided.")
 
 
 app = FastAPI(dependencies=[Depends(requires_token_when_set)], lifespan=lifespan)

logdetective/server/templates/gitlab_full_comment.md.j2

@@ -12,7 +12,7 @@ In this case, we are {{ "%.2f" | format(certainty) }}% certain of the response {
 {% for snippet in snippets %}
 <li>
 <b>Line {{ snippet.line_number }}:</b> <code>{{ snippet.text }}</code>
-{{ snippet.explanation }}
+{{ snippet.explanation.text }}
 </li>
 {% endfor %}
 </ul>

logdetective/server/utils.py

@@ -29,6 +29,7 @@ def mine_logs(log: str) -> List[Tuple[int, str]]:
         context=True,
         max_clusters=SERVER_CONFIG.extractor.max_clusters,
         skip_snippets=SKIP_SNIPPETS_CONFIG,
+        max_snippet_len=SERVER_CONFIG.extractor.max_snippet_len
     )
 
     LOG.info("Getting summary")

logdetective/utils.py

@@ -39,7 +39,7 @@ def chunk_continues(text: str, index: int) -> bool:
     return False
 
 
-def get_chunks(text: str) -> Generator[Tuple[int, str], None, None]:
+def get_chunks(text: str, max_len: int = 2000) -> Generator[Tuple[int, str], None, None]:
     """Split log into chunks according to heuristic
     based on whitespace and backslash presence.
     """
@@ -54,7 +54,7 @@ def get_chunks(text: str) -> Generator[Tuple[int, str], None, None]:
         chunk += text[i]
         if text[i] == "\n":
             next_line_number += 1
-            if i + 1 < text_len and chunk_continues(text, i):
+            if i + 1 < text_len and chunk_continues(text, i) and i + 1 < max_len:
                 i += 1
                 continue
             yield (original_line_number, chunk)

{logdetective-2.0.0.dist-info → logdetective-2.0.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: logdetective
-Version: 2.0.0
+Version: 2.0.2
 Summary: Log using LLM AI to search for build/test failures and provide ideas for fixing these.
 License: Apache-2.0
 Author: Jiri Podivin
@@ -20,23 +20,30 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development :: Debuggers
 Provides-Extra: server
 Provides-Extra: server-testing
-Requires-Dist: aiohttp (>=3.7.4)
+Provides-Extra: testing
+Requires-Dist: aiohttp (>=3.7.4,<4.0.0)
 Requires-Dist: aiolimiter (>=1.0.0,<2.0.0) ; extra == "server"
+Requires-Dist: aioresponses (>=0.7.8,<0.8.0) ; extra == "testing"
 Requires-Dist: alembic (>=1.13.3,<2.0.0) ; extra == "server" or extra == "server-testing"
 Requires-Dist: backoff (==2.2.1) ; extra == "server" or extra == "server-testing"
 Requires-Dist: drain3 (>=0.9.11,<0.10.0)
-Requires-Dist: fastapi (>=0.111.1) ; extra == "server" or extra == "server-testing"
-Requires-Dist: huggingface-hub (>0.23.2)
+Requires-Dist: fastapi (>=0.111.1,<1.0.0) ; extra == "server" or extra == "server-testing"
+Requires-Dist: flexmock (>=0.12.2,<0.13.0) ; extra == "testing"
+Requires-Dist: huggingface-hub (>0.23.2,<0.35.0)
 Requires-Dist: koji (>=1.35.0,<2.0.0) ; extra == "server" or extra == "server-testing"
-Requires-Dist: llama-cpp-python (>0.2.56,!=0.2.86)
+Requires-Dist: llama-cpp-python (>0.2.56,!=0.2.86,<1.0.0)
 Requires-Dist: matplotlib (>=3.8.4,<4.0.0) ; extra == "server" or extra == "server-testing"
 Requires-Dist: numpy (>=1.26.0)
 Requires-Dist: openai (>=1.82.1,<2.0.0) ; extra == "server" or extra == "server-testing"
 Requires-Dist: psycopg2 (>=2.9.9,<3.0.0) ; extra == "server"
 Requires-Dist: psycopg2-binary (>=2.9.9,<3.0.0) ; extra == "server-testing"
 Requires-Dist: pydantic (>=2.8.2,<3.0.0)
+Requires-Dist: pytest (>=8.4.1,<9.0.0) ; extra == "testing"
+Requires-Dist: pytest-asyncio (>=1.1.0,<2.0.0) ; extra == "testing"
+Requires-Dist: pytest-mock (>=3.14.1,<4.0.0) ; extra == "server-testing"
 Requires-Dist: python-gitlab (>=4.4.0)
 Requires-Dist: pyyaml (>=6.0.1,<7.0.0)
+Requires-Dist: responses (>=0.25.7,<0.26.0) ; extra == "server-testing"
 Requires-Dist: sentry-sdk[fastapi] (>=2.17.0,<3.0.0) ; extra == "server" or extra == "server-testing"
 Requires-Dist: sqlalchemy (>=2.0.36,<3.0.0) ; extra == "server" or extra == "server-testing"
 Project-URL: homepage, https://github.com/fedora-copr/logdetective

{logdetective-2.0.0.dist-info → logdetective-2.0.2.dist-info}/RECORD

@@ -1,7 +1,7 @@
 logdetective/__init__.py,sha256=VqRngDcuFT7JWms8Qc_MsOvajoXVOKPr-S1kqY3Pqhc,59
 logdetective/constants.py,sha256=aCwrkBrDdS_kbNESK-Z-ewg--DSzodV2OMgwEq3UE38,2456
 logdetective/drain3.ini,sha256=ni91eCT1TwTznZwcqWoOVMQcGEnWhEDNCoTPF7cfGfY,1360
-logdetective/extractors.py,sha256=BkQe7FMLDoKVWitP85Vpv1qle1Fo1FeupKm0wVlcALI,1859
+logdetective/extractors.py,sha256=gRiEcS1M_0Otu9zzT-tGm2ez3DoCKHYrQwJXZAZS0lk,1995
 logdetective/logdetective.py,sha256=DECG4qnmYHlCcQ5Waj3Esr4wSb6LtM4m7qqtmZqYDX0,6151
 logdetective/models.py,sha256=h01nagxgb8sR9Js_00DMoZv6GvwHjcOk0MeKttftDHk,2460
 logdetective/prompts-summary-first.yml,sha256=3Zfp4NNOfaFYq5xBlBjeQa5PdjYfS4v17OtJqQ-DRpU,821
@@ -24,16 +24,16 @@ logdetective/server/gitlab.py,sha256=MrAprXLTN6Q15qBC_Y2y42iKdtmIfed_pfjEt0gABvc
 logdetective/server/koji.py,sha256=LG1pRiKUFvYFRKzgQoUG3pUHfcEwMoaMNjUSMKw_pBA,5640
 logdetective/server/llm.py,sha256=_JRbEpNHzXWy0BuwjFhhuwLdPzFi4yKikZHwoibWiek,9736
 logdetective/server/metric.py,sha256=QrrX1FmMa7sc57av0P9UFOiCIFYVLs1opOWV3ObYo0s,4086
-logdetective/server/models.py,sha256=ogMh00LAPUD8f5HjMWBfOsRxhcTvxbAZ6-MzmG4PhzE,19244
+logdetective/server/models.py,sha256=SY3wdYgyNcggdCcslDX6QM-aMTTtI4OAG5UQtWrSIWM,19267
 logdetective/server/plot.py,sha256=C98U9prGoPkp8_t4v2dovdZuwOhSbxXSeB_K9Q2r3NE,14607
-logdetective/server/server.py,sha256=rpl5WtnesxPz_Zahy2bKN5wMIjbvdE1sM1ulvxvDiGw,24416
-logdetective/server/templates/gitlab_full_comment.md.j2,sha256=mD98HA-wQlv-H98M9Bb63lCROpcF4TkSpoat90c7eNY,1942
+logdetective/server/server.py,sha256=zap8Mz3NTFvaDJMNQDATbPYk6MhQ9o1J9gJECnGWvuQ,24694
+logdetective/server/templates/gitlab_full_comment.md.j2,sha256=H4NPjm3l8X5d0TNtfyZZZj_gHY1Y7hWEqY6RaVA8qt0,1947
 logdetective/server/templates/gitlab_short_comment.md.j2,sha256=vPisU1c98LPKEwlKtMrtlqnEOlbykPZK96MpHAf-o88,1758
-logdetective/server/utils.py,sha256=6RnEDsUgx6S3m6PzMcAO_Smsu_U-MHoP6FGKiaERPcY,4224
+logdetective/server/utils.py,sha256=ixEXSc93Zdwy94YkExdNJ1vRCe90dNP7DQpy8XlFYRU,4288
 logdetective/skip_snippets.yml,sha256=reGlhPPCo06nNUJWiC2LY-OJOoPdcyOB7QBTSMeh0eg,487
-logdetective/utils.py,sha256=UAhPWbOGdTR7PWc1dEQk8FSxBSsO0UgfdyY8AKGfMJY,7781
-logdetective-2.0.0.dist-info/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
-logdetective-2.0.0.dist-info/METADATA,sha256=so6qpVEedLsX8ZvyQbiwAF4AOOWBlWZ0pV5At4xvPSg,20104
-logdetective-2.0.0.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-logdetective-2.0.0.dist-info/entry_points.txt,sha256=3K_vXja6PmcA8sNdUi63WdImeiNhVZcEGPTaoJmltfA,63
-logdetective-2.0.0.dist-info/RECORD,,
+logdetective/utils.py,sha256=Ur4EPCAeuvymFcUO10SpbCBI2tdcZ33s-fjCn9pfgUY,7822
+logdetective-2.0.2.dist-info/LICENSE,sha256=z8d0m5b2O9McPEK1xHG_dWgUBT6EfBDz6wA0F7xSPTA,11358
+logdetective-2.0.2.dist-info/METADATA,sha256=nDKMJPHfVQkCNGBszgka3b0Y0MBAfGrU_-aZRIKmOqM,20560
+logdetective-2.0.2.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+logdetective-2.0.2.dist-info/entry_points.txt,sha256=3K_vXja6PmcA8sNdUi63WdImeiNhVZcEGPTaoJmltfA,63
+logdetective-2.0.2.dist-info/RECORD,,