feat: Track grab dates as fallback when no imports found

This commit is contained in:
2025-09-08 13:34:48 -04:00
parent 8702c9556d
commit 134fb0c5c3
3 changed files with 111 additions and 25 deletions
+5
View File
@@ -4,6 +4,11 @@ All notable changes to this project will be documented in this file.
## [Unreleased]
## [0.3.7] - 2025-09-08
### Changed
- Improved import date detection by tracking grab dates as fallback
- Added warning when falling back to grab date
## [0.3.6] - 2025-09-08
### Changed
- Improved import detection filtering
+1 -1
View File
@@ -1 +1 @@
0.3.6
0.3.7
+105 -24
View File
@@ -275,9 +275,14 @@ class RadarrClient:
pass
# For import/download events, handle specially
# Grab date of first download/grab event as fallback
if event_type == "grabbed" and not earliest_grab_date:
earliest_grab_date = datetime.fromisoformat(event["date"].replace("Z", "+00:00")).astimezone(timezone.utc).isoformat(timespec="seconds")
_log("DEBUG", f"Found first grab event at {earliest_grab_date}")
if event_type not in ["moviefileimported", "downloadFolderImported"]:
continue
imported_path = None
# Try different sources for the path
if event_type == "moviefileimported":
@@ -431,28 +436,104 @@ class RadarrClient:
return None, "radarr:no_date_found"
def _get_earliest_import_date(self, movie_id: int, movie_info: Dict) -> Optional[str]:
"""Get the earliest import date from Radarr history."""
_log("INFO", f"Finding earliest import for movie_id {movie_id}")
# Get full movie history
earliest_real_import = None
earliest_grab_date = None
page = 1
page_size = 50
total_events = 0
if __name__ == "__main__":
# Test the client
import os
base_url = os.environ.get("RADARR_URL", "")
api_key = os.environ.get("RADARR_API_KEY", "")
if base_url and api_key:
client = RadarrClient(base_url, api_key)
# Test with a known movie
test_imdb = "tt1596343" # Example
movie = client.movie_by_imdb(test_imdb)
if movie:
movie_id = movie.get("id")
print(f"Found movie: {movie.get('title')} (ID: {movie_id})")
while True:
# Get page of history
history_data = self._get_movie_history_page(movie_id, page, page_size)
import_date, source = client.get_movie_import_date(movie_id)
print(f"Import date: {import_date} ({source})")
else:
print(f"Movie not found: {test_imdb}")
else:
print("Please set RADARR_URL and RADARR_API_KEY environment variables for testing")
# Process events on this page
for event in history_data:
event_type = event.get("eventType", "").lower()
# Track earliest grab date as fallback
if event_type == "grabbed" and not earliest_grab_date:
earliest_grab_date = datetime.fromisoformat(event["date"].replace("Z", "+00:00")).astimezone(timezone.utc).isoformat(timespec="seconds")
_log("DEBUG", f"Found first grab event at {earliest_grab_date}")
# Only process import events
if event_type not in ["moviefileimported", "downloadFolderImported"]:
continue
# Try different sources for the path
imported_path = None
if event_type == "moviefileimported":
imported_path = (event.get("data", {}).get("importedPath", "") or
event.get("data", {}).get("path", "") or
event.get("importedPath", "") or
event.get("sourcePath", "")).lower()
elif event_type == "downloadFolderImported":
imported_path = (event.get("data", {}).get("path", "") or
event.get("sourcePath", "") or
event.get("downloadPath", "")).lower()
if not imported_path:
continue
movie_imdb = (movie_info.get("imdbId", "") or "").lower()
movie_title = (movie_info.get("title", "") or "").lower()
movie_year = str(movie_info.get("year", ""))
# First try IMDb ID match
if movie_imdb and (
f"[imdb-{movie_imdb}]" in imported_path or
f"[{movie_imdb}]" in imported_path or
movie_imdb in imported_path
):
_log("INFO", f"Found potential IMDb match in {event_type} event: {imported_path}")
date_iso = datetime.fromisoformat(event["date"].replace("Z", "+00:00")).astimezone(timezone.utc).isoformat(timespec="seconds")
_log("INFO", f"✅ FOUND IMPORT at {date_iso}")
earliest_real_import = date_iso
break
# Then try title/year match with fuzzy path cleaning
if movie_title and movie_year:
# Clean strings for comparison by replacing common separators
clean_title = movie_title.replace(" ", ".").replace(":", ".").replace("-", ".").replace("_", ".").lower()
clean_path = imported_path.replace(" ", ".").replace("-", ".").replace("_", ".").replace("[", "").replace("]", "").lower()
# Remove common words for better matching
for word in ["the", "a", "an"]:
clean_title = clean_title.replace(f".{word}.", ".").replace(f"{word}.", "").replace(f".{word}", "")
clean_path = clean_path.replace(f".{word}.", ".").replace(f"{word}.", "").replace(f".{word}", "")
# Look for both title and year in the path
if clean_title in clean_path and movie_year in clean_path:
date_iso = datetime.fromisoformat(event["date"].replace("Z", "+00:00")).astimezone(timezone.utc).isoformat(timespec="seconds")
_log("INFO", f"Found potential title/year match in {event_type} event: {clean_title} ({movie_year})")
_log("INFO", f"✅ FOUND IMPORT at {date_iso}")
earliest_real_import = date_iso
break
# Break if no more pages
if not history_data:
break
# Break if we found an import
if earliest_real_import:
break
total_events += len(history_data)
page += 1
_log("INFO", f"Processed {total_events} events across {page-1} pages")
# Return earliest real import if found
if earliest_real_import:
return earliest_real_import
# Fall back to earliest grab date if available
if earliest_grab_date:
_log("WARNING", f"⚠️ No real imports found, using grab date: {earliest_grab_date}")
return earliest_grab_date
return None