bellingcat/auto-archiver
1
0
Fork 0
mirror of https://github.com/bellingcat/auto-archiver.git synced 2026-06-12 13:18:28 +03:00
Code Issues Packages Projects Releases Wiki Activity

improve missing col behaviour to error log

Browse Source
This commit is contained in:
msramalho
2023-06-26 17:27:37 +01:00
parent d80145002d
commit a7a6bda1c2
1 changed files with 15 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
src/auto_archiver/databases/gsheet_db.py
View File

@@ -52,8 +52,11 @@ class GsheetsDb(Database):
def batch_if_valid(col, val, final_value=None):
final_value = final_value or val
if val and gw.col_exists(col) and gw.get_cell(row_values, col) == '':
cell_updates.append((row, col, final_value))
try:
if val and gw.col_exists(col) and gw.get_cell(row_values, col) == '':
cell_updates.append((row, col, final_value))
except Exception as e:
logger.error(f"Unable to batch {col}={final_value} due to {e}")
cell_updates.append((row, 'status', item.status))
@@ -65,6 +68,16 @@ class GsheetsDb(Database):
batch_if_valid('text', item.get("content", ""))
batch_if_valid('timestamp', item.get_timestamp())
batch_if_valid('hash', media.get("hash", "not-calculated"))
# merge all pdq hashes into a single string, if present
pdq_hashes = []
all_media = item.get_all_media()
for m in all_media:
if pdq := m.get("pdq_hash"):
pdq_hashes.append(pdq)
if len(pdq_hashes):
batch_if_valid('pdq_hash', ",".join(pdq_hashes))
if (screenshot := item.get_media_by_id("screenshot")) and hasattr(screenshot, "urls"):
batch_if_valid('screenshot', "\n".join(screenshot.urls))