Linting fixes

evekhm · Jul 19, 2024 · fc2b90a · fc2b90a
1 parent b8ca445
commit fc2b90a
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 3 deletions.
diff --git a/classify-split-extract-workflow/classify-job/Dockerfile b/classify-split-extract-workflow/classify-job/Dockerfile
@@ -9,6 +9,5 @@ WORKDIR $APP_HOME
 COPY . ./
 
 # Install production dependencies.
-RUN pip install -r requirements.txt
-
+RUN pip install --no-cache-dir -r requirements.txt
 CMD ["/usr/local/bin/python3", "classify.py"]
diff --git a/classify-split-extract-workflow/classify-job/config.py b/classify-split-extract-workflow/classify-job/config.py
@@ -210,7 +210,7 @@ def load_config(bucket_name: str, filename: str) -> Optional[Dict[Any, Any]]:
 
     last_modified_time = blob.updated
     global LAST_MODIFIED_TIME_OF_CONFIG
-    if last_modified_time == LAST_MODIFIED_TIME_OF_CONFIG:
+    if LAST_MODIFIED_TIME_OF_CONFIG == last_modified_time:
         return CONFIG_DATA
 
     logger.info(f"Reloading config from: {filename}")

diff --git a/classify-split-extract-workflow/classify-job/split_and_classify.py b/classify-split-extract-workflow/classify-job/split_and_classify.py
@@ -324,6 +324,9 @@ def split_pdf(gcs_uri: str, entities: List[Document.Entity]) -> Dict:
                 if page_refs:
                     start_page = int(page_refs[0].page)
                     end_page = int(page_refs[-1].page)
+                else:
+                    logger.warning(f"Skipping {pdf_path} entity due to no page refs, no splitting")
+                    continue
                 page_range = (
                     f"pg{start_page + 1}"
                     if start_page == end_page