logdetective 2.0.0__tar.gz → 2.0.1__tar.gz

{logdetective-2.0.0 → logdetective-2.0.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: logdetective
-Version: 2.0.0
+Version: 2.0.1
 Summary: Log using LLM AI to search for build/test failures and provide ideas for fixing these.
 License: Apache-2.0
 Author: Jiri Podivin

{logdetective-2.0.0 → logdetective-2.0.1}/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-2.0.0 → logdetective-2.0.1}/logdetective/server/models.py

@@ -247,6 +247,7 @@ class ExtractorConfig(BaseModel):
     context: bool = True
     max_clusters: int = 8
     verbose: bool = False
+    max_snippet_len: int = 2000
 
     def __init__(self, data: Optional[dict] = None):
         super().__init__()
@@ -256,6 +257,7 @@ class ExtractorConfig(BaseModel):
         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-2.0.0 → logdetective-2.0.1}/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-2.0.0 → logdetective-2.0.1}/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-2.0.0 → logdetective-2.0.1}/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 → logdetective-2.0.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "logdetective"
-version = "2.0.0"
+version = "2.0.1"
 description = "Log using LLM AI to search for build/test failures and provide ideas for fixing these."
 authors = ["Jiri Podivin <jpodivin@gmail.com>"]
 license = "Apache-2.0"