From 07bd967f59e87881187655dad61cf10b29555fce Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mike=20F=C3=A4hrmann?= <mike_faehrmann@web.de>
Date: Thu, 5 Sep 2024 18:43:06 +0200
Subject: [PATCH] [pp:ugoira] update (#6056)

- introduce '_ugoira_frame_index' metadata field
- store Ugoira file exts separately
- add 'skip' option
---
 docs/configuration.rst             | 33 ++++++++++++++++++++++++++++++
 gallery_dl/extractor/pixiv.py      |  2 +-
 gallery_dl/postprocessor/ugoira.py | 30 +++++++++++++++++----------
 3 files changed, 53 insertions(+), 12 deletions(-)

diff --git a/docs/configuration.rst b/docs/configuration.rst
index dc6df5f1..3fefc911 100644
--- a/docs/configuration.rst
+++ b/docs/configuration.rst
@@ -5643,6 +5643,28 @@ Example
 Description
     Hash digests to compute.
 
+    For a list of available hash algorithms, run
+
+    .. code::
+
+        python -c "import hashlib; print('\n'.join(hashlib.algorithms_available))"
+
+    or see `python/hashlib <https://docs.python.org/3/library/hashlib.html>`__.
+
+    * If this is a ``string``,
+      it is parsed as a a comma-separated list of algorthm-fieldname pairs:
+
+      .. code::
+
+          [<hash algorithm> ":"] <field name> ["," ...]
+
+      When ``<hash algorithm>`` is omitted,
+      ``<field name>`` is used as algorithm name.
+
+    * If this is an ``object``,
+      it is a ``<field name>`` to ``<algorithm name>`` mapping
+      for hash digests to compute.
+
 
 metadata.mode
 -------------
@@ -6259,6 +6281,16 @@ Description
     to prevent it from only being displayed for a very short amount of time.
 
 
+ugoira.skip
+-----------
+Type
+    ``bool``
+Default
+    ``true``
+Description
+    Do not convert frames if target file already exists.
+
+
 zip.compression
 ---------------
 Type
@@ -6270,6 +6302,7 @@ Description
 
     Possible values are ``"store"``, ``"zip"``, ``"bzip2"``, ``"lzma"``.
 
+
 zip.extension
 -------------
 Type
diff --git a/gallery_dl/extractor/pixiv.py b/gallery_dl/extractor/pixiv.py
index eda0405c..3479b88f 100644
--- a/gallery_dl/extractor/pixiv.py
+++ b/gallery_dl/extractor/pixiv.py
@@ -119,7 +119,7 @@ class PixivExtractor(Extractor):
 
                     for num, frame in enumerate(frames):
                         url = ("{}{}.{}".format(base, num, ext))
-                        work["num"] = num
+                        work["num"] = work["_ugoira_frame_index"] = num
                         work["suffix"] = "_p{:02}".format(num)
                         text.nameext_from_url(url, work)
                         yield Message.Url, url, work
diff --git a/gallery_dl/postprocessor/ugoira.py b/gallery_dl/postprocessor/ugoira.py
index 0ab58b2a..f053afa0 100644
--- a/gallery_dl/postprocessor/ugoira.py
+++ b/gallery_dl/postprocessor/ugoira.py
@@ -36,6 +36,7 @@ class UgoiraPP(PostProcessor):
         self.delete = not options.get("keep-files", False)
         self.repeat = options.get("repeat-last-frame", True)
         self.mtime = options.get("mtime", True)
+        self.skip = options.get("skip", True)
         self.uniform = self._convert_zip = self._convert_files = False
 
         ffmpeg = options.get("ffmpeg-location")
@@ -109,12 +110,17 @@ class UgoiraPP(PostProcessor):
                 pathfmt.build_path()
         else:
             pathfmt.build_path()
-            num = pathfmt.kwdict["num"]
-            if not num:
-                self._files = [pathfmt.realpath]
+            index = pathfmt.kwdict["_ugoira_frame_index"]
+            frame = self._frames[index].copy()
+            frame["index"] = index
+            frame["path"] = pathfmt.realpath
+            frame["ext"] = pathfmt.kwdict["extension"]
+
+            if not index:
+                self._files = [frame]
             else:
-                self._files.append(pathfmt.realpath)
-                if num+1 >= len(self._frames):
+                self._files.append(frame)
+                if len(self._files) >= len(self._frames):
                     self._convert_files = True
 
     def convert_zip(self, pathfmt):
@@ -144,32 +150,34 @@ class UgoiraPP(PostProcessor):
         self._convert_files = False
 
         with tempfile.TemporaryDirectory() as tempdir:
-            for frame, path in zip(self._frames, self._files):
+            for frame in self._files:
 
                 # update frame filename extension
                 frame["file"] = name = "{}.{}".format(
-                    frame["file"].partition(".")[0],
-                    path.rpartition(".")[2])
+                    frame["file"].partition(".")[0], frame["ext"])
 
                 # move frame into tempdir
                 try:
-                    self._copy_file(path, tempdir + "/" + name)
+                    self._copy_file(frame["path"], tempdir + "/" + name)
                 except OSError as exc:
                     self.log.debug("Unable to copy frame %s (%s: %s)",
                                    name, exc.__class__.__name__, exc)
                     return
 
             pathfmt.kwdict["num"] = 0
+            self._frames = self._files
             if self.convert(pathfmt, tempdir):
                 self.log.info(pathfmt.filename)
                 if self.delete:
                     self.log.debug("Deleting frames")
-                    for path in self._files:
-                        util.remove_file(path)
+                    for frame in self._files:
+                        util.remove_file(frame["path"])
 
     def convert(self, pathfmt, tempdir):
         pathfmt.set_extension(self.extension)
         pathfmt.build_path()
+        if self.skip and pathfmt.exists():
+            return True
 
         # process frames and collect command-line arguments
         args = self._process(pathfmt, tempdir)