Add support for live photos

Look for MediaGroupUUID and ContentIdentifier EXIF tags. These are present in media files which are part of live photo pairs. If found, any future files with the same UUID are named the same as the first one and placed in the same directory. With this change, live photos and their associated .mov files are moved together.
ivandokov · Aug 30, 2023 · 49173d8 · 49173d8
1 parent f94aa2e
commit 49173d8
Show file tree

Hide file tree

Showing 2 changed files with 34 additions and 16 deletions.
diff --git a/src/exif.py b/src/exif.py
@@ -28,5 +28,5 @@ def data(self):
     def get_exif_command(filename):
         # Handle all platform variations
         if sys.platform == 'win32':
-            return f'exiftool -time:all -mimetype -j "{filename}"'
-        return f'exiftool -time:all -mimetype -j {shlex.quote(filename)}'
+            return f'exiftool -MediaGroupUUID -ContentIdentifier -time:all -mimetype -j "{filename}"'
+        return f'exiftool -MediaGroupUUID -ContentIdentifier -time:all -mimetype -j {shlex.quote(filename)}'
diff --git a/src/phockup.py b/src/phockup.py
@@ -12,6 +12,7 @@
 
 from src.date import Date
 from src.exif import Exif
+from collections import defaultdict
 
 logger = logging.getLogger('phockup')
 ignored_files = ('.DS_Store', 'Thumbs.db')
@@ -28,6 +29,7 @@ def __init__(self, input_dir, output_dir, **args):
         self.unknown_found = 0
         self.files_moved = 0
         self.files_copied = 0
+        self.content_ids = defaultdict(list)
 
         input_dir = os.path.expanduser(input_dir)
         output_dir = os.path.expanduser(output_dir)
@@ -56,7 +58,7 @@ def __init__(self, input_dir, output_dir, **args):
         # default to concurrency of one to retain existing behavior
         self.max_concurrency = args.get("max_concurrency", 1)
         if self.max_concurrency > 1:
-            logger.info(f"Using {self.max_concurrency} workers to process files.")
+            logger.debug(f"Using {self.max_concurrency} workers to process files.")
 
         self.stop_depth = self.input_dir.count(os.sep) + self.max_depth \
             if self.max_depth > -1 else sys.maxsize
@@ -85,21 +87,21 @@ def __init__(self, input_dir, output_dir, **args):
             self.print_action_report(run_time)
 
     def print_action_report(self, run_time):
-        logger.info(f"Processed {self.files_processed} files in {run_time:.2f} seconds. Average Throughput: {self.files_processed/run_time:.2f} files/second")
+        logger.debug(f"Processed {self.files_processed} files in {run_time:.2f} seconds. Average Throughput: {self.files_processed/run_time:.2f} files/second")
         if self.unknown_found:
-            logger.info(f"Found {self.unknown_found} files without EXIF date data.")
+            logger.debug(f"Found {self.unknown_found} files without EXIF date data.")
         if self.duplicates_found:
-            logger.info(f"Found {self.duplicates_found} duplicate files.")
+            logger.debug(f"Found {self.duplicates_found} duplicate files.")
         if self.files_copied:
             if self.dry_run:
-                logger.info(f"Would have copied {self.files_copied} files.")
+                logger.debug(f"Would have copied {self.files_copied} files.")
             else:
-                logger.info(f"Copied {self.files_copied} files.")
+                logger.debug(f"Copied {self.files_copied} files.")
         if self.files_moved:
             if self.dry_run:
-                logger.info(f"Would have moved {self.files_moved} files.")
+                logger.debug(f"Would have moved {self.files_moved} files.")
             else:
-                logger.info(f"Moved {self.files_moved} files.")
+                logger.debug(f"Moved {self.files_moved} files.")
 
     def check_directories(self):
         """
@@ -258,16 +260,16 @@ def process_file(self, filename):
                     and self.file_type != target_file_type:
                 progress = f"{progress} => skipped, file is '{target_file_type}' \
 but looking for '{self.file_type}'"
-                logger.info(progress)
+                logger.debug(progress)
                 break
 
             if self.skip_unknown and output.endswith(self.no_date_dir):
                 # Skip files that didn't generate a path from EXIF data
-                progress = f"{progress} => skipped, unknown date EXIF information for '{target_file_name}'"
+                progress = f"{progress} => skipped, unknown date EXIF debugrmation for '{target_file_name}'"
                 self.unknown_found += 1
                 if self.progress:
                     self.pbar.write(progress)
-                logger.info(progress)
+                logger.debug(progress)
                 break
 
             if os.path.isfile(target_file):
@@ -276,7 +278,7 @@ def process_file(self, filename):
                     self.duplicates_found += 1
                     if self.progress:
                         self.pbar.write(progress)
-                    logger.info(progress)
+                    logger.debug(progress)
                     break
             else:
                 if self.move:
@@ -307,7 +309,7 @@ def process_file(self, filename):
                 progress = f'{progress} => {target_file}'
                 if self.progress:
                     self.pbar.write(progress)
-                logger.info(progress)
+                logger.debug(progress)
 
                 self.process_xmp(filename, target_file_name, suffix, output)
                 break
@@ -342,6 +344,22 @@ def get_file_name_and_path(self, filename):
             target_file_name = os.path.basename(filename)
 
         target_file_path = os.path.sep.join([output, target_file_name])
+
+        if exif_data and ('ContentIdentifier' in exif_data or 'MediaGroupUUID' in exif_data):
+            uuid = exif_data['ContentIdentifier'] if 'ContentIdentifier' in exif_data else exif_data['MediaGroupUUID']
+            logger.debug(f"{filename} has media group ID: {uuid}")
+            cids = self.content_ids[uuid]
+
+            # If a file with this same ContentId UUID exists already, use its
+            # name and put it in the same directory
+            if cids:
+                logger.debug(f"Found existing files with same ID: {cids}")
+                logger.debug(f"Previous target for current file: {target_file_path}")
+                target_file_name = os.path.splitext(os.path.basename(cids[0]))[0] + os.path.splitext(filename)[1]
+                target_file_path = os.path.sep.join([os.path.split(cids[0])[0], target_file_name])
+                logger.debug(f"New target for current file: {target_file_path}")
+            cids.append(target_file_path)
+
         return output, target_file_name, target_file_path, target_file_type
 
     def process_xmp(self, original_filename, file_name, suffix, output):
@@ -364,7 +382,7 @@ def process_xmp(self, original_filename, file_name, suffix, output):
 
         for original, target in xmp_files.items():
             xmp_path = os.path.sep.join([output, target])
-            logger.info(f'{original} => {xmp_path}')
+            logger.debug(f'{original} => {xmp_path}')
 
             if not self.dry_run:
                 if self.move: