From 70b261f21dc0a2b8bafb6890c404e3652d7b6b43 Mon Sep 17 00:00:00 2001 From: stefpi Date: Wed, 5 Nov 2025 15:50:34 -0500 Subject: [PATCH 1/3] fix clean_excluded dir parsing --- api/data_pipeline.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/api/data_pipeline.py b/api/data_pipeline.py index 67b2dcbf..5903f52f 100644 --- a/api/data_pipeline.py +++ b/api/data_pipeline.py @@ -278,7 +278,7 @@ def should_process_file(file_path: str, use_inclusion: bool, included_dirs: List # Check if file is in an excluded directory for excluded in excluded_dirs: - clean_excluded = excluded.strip("./").rstrip("/") + clean_excluded = excluded.split("./")[1].rstrip("/") if clean_excluded in file_path_parts: is_excluded = True break From 59abbc430e48e53d00ea77483123060a2f18f03b Mon Sep 17 00:00:00 2001 From: stefpi Date: Wed, 5 Nov 2025 16:00:48 -0500 Subject: [PATCH 2/3] apply fix to included directories as well --- api/data_pipeline.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/api/data_pipeline.py b/api/data_pipeline.py index 5903f52f..c1610bae 100644 --- a/api/data_pipeline.py +++ b/api/data_pipeline.py @@ -249,7 +249,7 @@ def should_process_file(file_path: str, use_inclusion: bool, included_dirs: List # Check if file is in an included directory if included_dirs: for included in included_dirs: - clean_included = included.strip("./").rstrip("/") + clean_included = included.split("./")[1].rstrip("/") if clean_included in file_path_parts: is_included = True break From 659296e57bdfb001bdb30d94124b778d5ae81ef8 Mon Sep 17 00:00:00 2001 From: stefpi Date: Wed, 5 Nov 2025 16:06:01 -0500 Subject: [PATCH 3/3] implement gemeni's suggestion --- api/data_pipeline.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/api/data_pipeline.py b/api/data_pipeline.py index c1610bae..458c8fa3 100644 --- a/api/data_pipeline.py +++ b/api/data_pipeline.py @@ -249,7 +249,7 @@ def should_process_file(file_path: str, use_inclusion: bool, included_dirs: List # Check if file is in an included directory if included_dirs: for included in included_dirs: - clean_included = included.split("./")[1].rstrip("/") + clean_included = included.removeprefix("./").removesuffix("/") if clean_included in file_path_parts: is_included = True break @@ -278,7 +278,7 @@ def should_process_file(file_path: str, use_inclusion: bool, included_dirs: List # Check if file is in an excluded directory for excluded in excluded_dirs: - clean_excluded = excluded.split("./")[1].rstrip("/") + clean_excluded = excluded.removeprefix("./").removesuffix("/") if clean_excluded in file_path_parts: is_excluded = True break