First format and fixes from ruff

api/all_tasks.py

@@ -9,7 +9,7 @@
 from django.utils import timezone
 from django_q.tasks import AsyncTask, schedule
 
-import api.util as util
+from api import util
 from api.models.long_running_job import LongRunningJob
 
 
@@ -62,7 +62,7 @@ def zip_photos_task(job_id, user, photos, filename):
             output_file.write(mf.getvalue())
 
     except Exception as e:
-        util.logger.error("Error while converting files to zip: {}".format(e))
+        util.logger.error(f"Error while converting files to zip: {e}")
 
     lrj.finished_at = datetime.now().replace(tzinfo=pytz.utc)
     lrj.finished = True
@@ -77,15 +77,13 @@ def delete_zip_file(filename):
     file_path = os.path.join(settings.MEDIA_ROOT, "zip", filename)
     try:
         if not os.path.exists(file_path):
-            util.logger.error(
-                "Error while deleting file not found at : {}".format(file_path)
-            )
+            util.logger.error(f"Error while deleting file not found at : {file_path}")
             return
         else:
             os.remove(file_path)
-            util.logger.info("file deleted sucessfully at path : {}".format(file_path))
+            util.logger.info(f"file deleted sucessfully at path : {file_path}")
             return
 
     except Exception as e:
-        util.logger.error("Error while deleting file: {}".format(e))
+        util.logger.error(f"Error while deleting file: {e}")
         return e

api/api_util.py

@@ -297,7 +297,7 @@ def median_value(queryset, term):
 def calc_megabytes(bytes):
     if bytes == 0 or bytes is None:
         return 0
-    return round(((bytes / 1024) / 1024))
+    return round((bytes / 1024) / 1024)
 
 
 def get_server_stats():
@@ -842,18 +842,14 @@ def get_photo_month_counts(user):
 def get_searchterms_wordcloud(user):
     query = {}
     out = {"captions": [], "people": [], "locations": []}
-    query[
-        "captions"
-    ] = """
+    query["captions"] = """
         with captionList as (
             select unnest(regexp_split_to_array(search_captions,' , ')) caption
             from api_photo where owner_id = %(userid)s
         )
         select caption, count(*) from captionList group by caption order by count(*) desc limit 100;
     """
-    query[
-        "people"
-    ] = """
+    query["people"] = """
         with NameList as (
             select api_person.name
             from api_photo join api_face on image_hash = api_face.photo_id
@@ -862,9 +858,7 @@ def get_searchterms_wordcloud(user):
         )
         select name, count(*) from NameList group by name order by count(*) desc limit 100;
     """
-    query[
-        "locations"
-    ] = """
+    query["locations"] = """
          with arrayloctable as (
             select jsonb_array_elements(jsonb_extract_path(api_photo.geolocation_json,  'features')::jsonb) arrayloc , image_hash
             from api_photo where owner_id = %(userid)s

api/autoalbum.py

@@ -36,7 +36,7 @@ def regenerate_event_titles(user, job_id):
         aus = AlbumAuto.objects.filter(owner=user).prefetch_related("photos")
         target_count = len(aus)
         for idx, au in enumerate(aus):
-            logger.info("job {}: {}".format(job_id, idx))
+            logger.info(f"job {job_id}: {idx}")
             au._generate_title()
             au.save()
 
@@ -47,7 +47,7 @@ def regenerate_event_titles(user, job_id):
         lrj.finished = True
         lrj.finished_at = datetime.now().replace(tzinfo=pytz.utc)
         lrj.save()
-        logger.info("job {}: updated lrj entry to db".format(job_id))
+        logger.info(f"job {job_id}: updated lrj entry to db")
 
     except Exception:
         logger.exception("An error occurred")
@@ -88,24 +88,21 @@ def group(photos, dt=timedelta(hours=6)):
                 if len(groups) == 0:
                     groups.append([])
                     groups[-1].append(photo)
+                # Photos are sorted by timestamp, so we can just check the last photo of the last group
+                # to see if it is within the time delta
+                elif photo.exif_timestamp - groups[-1][-1].exif_timestamp < dt:
+                    groups[-1].append(photo)
+                # If the photo is not within the time delta, we create a new group
                 else:
-                    # Photos are sorted by timestamp, so we can just check the last photo of the last group
-                    # to see if it is within the time delta
-                    if photo.exif_timestamp - groups[-1][-1].exif_timestamp < dt:
-                        groups[-1].append(photo)
-                    # If the photo is not within the time delta, we create a new group
-                    else:
-                        groups.append([])
-                        groups[-1].append(photo)
+                    groups.append([])
+                    groups[-1].append(photo)
             return groups
 
         # Group images that are on the same 1 day and 12 hours interval
         groups = group(photos, dt=timedelta(days=1, hours=12))
         target_count = len(groups)
         logger.info(
-            "job {}: made {} groups out of {} images".format(
-                job_id, target_count, len(photos)
-            )
+            f"job {job_id}: made {target_count} groups out of {len(photos)} images"
         )
 
         album_locations = []
@@ -116,7 +113,7 @@ def group(photos, dt=timedelta(hours=6)):
             lastKey = group[-1].exif_timestamp + timedelta(hours=11, minutes=59)
             logger.info(str(key.date) + " - " + str(lastKey.date))
             logger.info(
-                "job {}: processing auto album with date: ".format(job_id)
+                f"job {job_id}: processing auto album with date: "
                 + key.strftime(date_format)
                 + " to "
                 + lastKey.strftime(date_format)
@@ -134,9 +131,7 @@ def group(photos, dt=timedelta(hours=6)):
                     album.timestamp = key
                     album.save()
 
-                    logger.info(
-                        "job {}: generate auto album {}".format(job_id, album.id)
-                    )
+                    logger.info(f"job {job_id}: generate auto album {album.id}")
                     locs = []
                     for item in items:
                         album.photos.add(item)
@@ -155,7 +150,7 @@ def group(photos, dt=timedelta(hours=6)):
                     continue
                 if qs.count() == 1:
                     album = qs.first()
-                    logger.info("job {}: update auto album {}".format(job_id, album.id))
+                    logger.info(f"job {job_id}: update auto album {album.id}")
                     for item in items:
                         if item in album.photos.all():
                             continue
@@ -167,9 +162,7 @@ def group(photos, dt=timedelta(hours=6)):
                 if qs.count() > 1:
                     # To-Do: Merge both auto albums
                     logger.info(
-                        "job {}: found multiple auto albums for date {}".format(
-                            job_id, key.strftime(date_format)
-                        )
+                        f"job {job_id}: found multiple auto albums for date {key.strftime(date_format)}"
                     )
                     continue
 

api/batch_jobs.py

@@ -5,7 +5,7 @@
 import pytz
 from django.db.models import Q
 
-import api.util as util
+from api import util
 from api.image_similarity import build_image_similarity_index
 from api.models.long_running_job import LongRunningJob
 from api.models.photo import Photo
@@ -37,7 +37,7 @@ def batch_calculate_clip_embedding(user):
         torch.multiprocessing.set_start_method("spawn", force=True)
 
     BATCH_SIZE = 64
-    util.logger.info("Using threads: {}".format(torch.get_num_threads()))
+    util.logger.info(f"Using threads: {torch.get_num_threads()}")
 
     done_count = 0
     while done_count < count:
@@ -67,7 +67,7 @@ def batch_calculate_clip_embedding(user):
                 obj.clip_embeddings_magnitude = magnitude
                 obj.save()
         except Exception as e:
-            util.logger.error("Error calculating clip embeddings: {}".format(e))
+            util.logger.error(f"Error calculating clip embeddings: {e}")
 
         lrj.progress_current = done_count
         lrj.progress_target = count

api/cluster_manager.py

@@ -99,9 +99,9 @@ def try_add_cluster(
             # Set initial metadata on the split clusters based on known faces
             for new_cluster in added_clusters:
                 new_cluster.set_metadata(encoding_by_person[new_cluster.person.id])
-                mean_encoding_by_cluster[
-                    new_cluster.id
-                ] = new_cluster.get_mean_encoding_array()
+                mean_encoding_by_cluster[new_cluster.id] = (
+                    new_cluster.get_mean_encoding_array()
+                )
 
             # Clear the face IDs list to prepare for processing the unknown faces
             for new_cluster in added_clusters:

api/date_time_extractor.py

@@ -117,8 +117,7 @@ class RuleTypes:
 
 
 class TimeExtractionRule:
-    """
-    The goal is to help extract local time, but for historical reason it is expected the returned
+    """The goal is to help extract local time, but for historical reason it is expected the returned
     datetime will have timezone to be set to pytz.utc (so local time + timezone equal to UTC)..
 
     Some sources of data might give us very rich information, e.g. timestamp + timezone,
@@ -303,8 +302,7 @@ def apply(
             raise ValueError(f"Unknown rule type {self.rule_type}")
 
     def _get_tz(self, description, gps_lat, gps_lon, user_default_tz):
-        """
-        None is a valid timezone returned here (meaning that we want to use server local time).
+        """None is a valid timezone returned here (meaning that we want to use server local time).
         This is why this function returns a tuple with the first element specifying success of
         determining the timezone, and the second element - the timezone itself.
         """