From 11d3deb66f87c0d71ec62b0f8b4c92fbcebbf351 Mon Sep 17 00:00:00 2001
From: Benedikt Kristinsson <benedikt@inventati.org>
Date: Sat, 16 May 2020 00:21:40 +0200
Subject: [PATCH] sorting mov files that are the iphone live photos. shitty
 code but comitting anyway

---
 movheaders.py  | 59 ++++++++++++++++++++++++++++++++++++++++++++++++++
 sort-photos.py | 48 ++++++++++++++++++++++++++++++++--------
 2 files changed, 98 insertions(+), 9 deletions(-)
 create mode 100644 movheaders.py
 mode change 100644 => 100755 sort-photos.py

diff --git a/movheaders.py b/movheaders.py
new file mode 100644
index 0000000..c5c9d9c
--- /dev/null
+++ b/movheaders.py
@@ -0,0 +1,59 @@
+#!/usr/bin/python
+
+
+from datetime import datetime
+import struct
+import sys
+
+def get_mov_dates(filename):
+    # borrowed from stackoverflow
+    # https://stackoverflow.com/questions/21355316/getting-metadata-for-mov-video
+    ATOM_HEADER_SIZE = 8
+    # difference between Unix epoch and QuickTime epoch, in seconds
+    EPOCH_ADJUSTER = 2082844800
+
+    # open file and search for moov item
+    f = open(filename, "rb")
+    while 1:
+        atom_header = f.read(ATOM_HEADER_SIZE)
+        if atom_header[4:8] == b'moov':
+            break
+        else:
+            try:
+                atom_size = struct.unpack(">I", atom_header[0:4])[0]
+                f.seek(atom_size - 8, 1)
+            except struct.error:
+                  raise ValueError("no 'moov' header found in {}".format(filename))
+
+    # found 'moov', look for 'mvhd' and timestamps
+    atom_header = f.read(ATOM_HEADER_SIZE)
+    if atom_header[4:8] == b'cmov':
+        print("moov atom is compressed")
+    elif atom_header[4:8] != b'mvhd':
+        print("expected to find 'mvhd' header")
+    else:
+        f.seek(4, 1)
+        creation_date = struct.unpack(">I", f.read(4))[0]
+        modification_date = struct.unpack(">I", f.read(4))[0]
+
+        created = datetime.utcfromtimestamp(creation_date - EPOCH_ADJUSTER)
+        #modified = datetime.utcfromtimestamp(modification_date - EPOCH_ADJUSTER)
+        # print(created)
+        #return modified
+        return created
+
+def get_date(path):
+    """Returns a tuple of strings (y, m, d) where m and d are zero-padded
+
+    what sort-photos.py expects"""
+
+    date = get_mov_dates(path)
+    y, m, d = [str(a).zfill(2) for a in [date.year, date.month, date.day]]
+
+    return (y, m, d)
+
+
+if __name__ == "__main__":
+    path = sys.argv[1]
+    date = get_date(path)
+    print(date)
diff --git a/sort-photos.py b/sort-photos.py
old mode 100644
new mode 100755
index 3fe65e6..1117a0d
--- a/sort-photos.py
+++ b/sort-photos.py
@@ -7,6 +7,8 @@ import shutil
 
 import exifread
 
+import movheaders
+
 HOME = os.getenv("HOME")
 NEXTCLOUD_DIR = os.path.join(HOME, "Nextcloud")
 PHOTOS_DIR = os.path.join(NEXTCLOUD_DIR, "Photos")
@@ -18,6 +20,7 @@ parser.add_argument("--src", required=True)
 parser.add_argument("--dst", default=PHOTOS_DIR, required=False)
 parser.add_argument("--debug", action="store_true")
 parser.add_argument("--dry-run", action="store_true")
+parser.add_argument("--fileext")
 args = parser.parse_args()
 
 def debug(s):
@@ -33,29 +36,56 @@ def mkdir(path):
         os.makedirs(path, exist_ok=True)
 
 def get_date(path):
-    with open(path, 'rb') as f:
-        tags = exifread.process_file(f, stop_tag="EXIF DateTimeOriginal")
+    if os.path.splitext(path)[1].lower() == ".mov":
+        return movheaders.get_date(path)
+    else:
+        with open(path, 'rb') as f:
+            tags = exifread.process_file(f, stop_tag="EXIF DateTimeOriginal")
+        y, m, d = tags["EXIF DateTimeOriginal"].values.split(" ")[0].split(":")
+        return (y, m, d)
 
-    y, m, d = tags["EXIF DateTimeOriginal"].values.split(" ")[0].split(":")
-    return (y, m, d)
 
 
 if __name__ == "__main__":
-    g = glob.glob(args.src)
+    #g = glob.glob(args.src)
+    # just one dir at a time
+    g = os.listdir(args.src)
+    # print(g)
+    # raise SystemExit
     for a in g:
         try:
-            y, m, _ = get_date(a)
+            # im gonna have fun with this next time i look at it
+            a_ext = os.path.splitext(a)[1].lower()
+            if args.fileext and a_ext !=  args.fileext:
+                continue
+            _path = os.path.join(args.src, a)
+            a_mb = os.path.getsize(_path) >> 20
+            #print("{}, {} mb".format(a, a_mb))
+            y, m, d = get_date(_path)
+            if args.fileext and a_ext == ".mov":
+                # big videos (actual videos ive taken) have already
+                # been uploaded by nextcloud. Trying to single out
+                # the live videos around photos
+                if a_mb > 5:
+                    debug("Too big: {}, {} mb, {}/{}/{}".format(_path, a_mb, y, m, d))
+                    # ignore completely, no wasting space on nextcloud
+                    continue
+
             ymdir = os.path.join(PHOTOS_DIR, y, m)
             mkdir(ymdir)
             debug("{} -> {}".format(a, ymdir))
-            move(a, ymdir)
+            move(_path, ymdir)
+        except ValueError:
+            mkdir(UNSORTED_DIR)
+            print("{} -> {}".format(a, UNSORTED_DIR))
+            move(os.path.join(args.src, a), UNSORTED_DIR)
         except KeyError:
             # Could not parse date
             mkdir(UNSORTED_DIR)
             print("{} -> {}".format(a, UNSORTED_DIR))
-            move(a, UNSORTED_DIR)
+            move(os.path.join(args.src, a), UNSORTED_DIR)
         except IsADirectoryError:
             # leave directories, happens if you didn't give the
             # correct full path
-            print("Skipping directory: {}".format(a))
+            print("Skipping directory: {}".format(os.path.join(args.src, a)))
             continue