Merge 796447b1a564ac3452274c0a10acd2923d14f028 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

* thx seproDev, yt-dlp/yt-dlp/pull/12694/commits/ae69e3c
* also add documenting comment in `InfoExtractor`

Makefile

@@ -49,7 +49,8 @@ offlinetest: codetest
 		--exclude test_subtitles.py \
 		--exclude test_write_annotations.py \
 		--exclude test_youtube_lists.py \
-		--exclude test_youtube_signature.py
+		--exclude test_youtube_signature.py \
+		--exclude test_post_hooks.py
 
 tar: youtube-dl.tar.gz
 

devscripts/run_tests.bat

@@ -1,7 +1,7 @@
 @echo off
 
 rem Keep this list in sync with the `offlinetest` target in Makefile
-set DOWNLOAD_TESTS="age_restriction^|download^|iqiyi_sdk_interpreter^|socks^|subtitles^|write_annotations^|youtube_lists^|youtube_signature"
+set DOWNLOAD_TESTS="age_restriction^|download^|iqiyi_sdk_interpreter^|socks^|subtitles^|write_annotations^|youtube_lists^|youtube_signature^|post_hooks"
 
 if "%YTDL_TEST_SET%" == "core" (
     set test_set="-I test_("%DOWNLOAD_TESTS%")\.py"

devscripts/run_tests.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 
 # Keep this list in sync with the `offlinetest` target in Makefile
-DOWNLOAD_TESTS="age_restriction|download|iqiyi_sdk_interpreter|socks|subtitles|write_annotations|youtube_lists|youtube_signature"
+DOWNLOAD_TESTS="age_restriction|download|iqiyi_sdk_interpreter|socks|subtitles|write_annotations|youtube_lists|youtube_signature|post_hooks"
 
 test_set=""
 multiprocess_args=""

test/test_post_hooks.py

@@ -0,0 +1,68 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import get_params, try_rm
+import youtube_dl.YoutubeDL
+from youtube_dl.utils import DownloadError
+
+
+class YoutubeDL(youtube_dl.YoutubeDL):
+    def __init__(self, *args, **kwargs):
+        super(YoutubeDL, self).__init__(*args, **kwargs)
+        self.to_stderr = self.to_screen
+
+
+TEST_ID = 'gr51aVj-mLg'
+EXPECTED_NAME = 'gr51aVj-mLg'
+
+
+class TestPostHooks(unittest.TestCase):
+    def setUp(self):
+        self.stored_name_1 = None
+        self.stored_name_2 = None
+        self.params = get_params({
+            'skip_download': False,
+            'writeinfojson': False,
+            'quiet': True,
+            'verbose': False,
+            'cachedir': False,
+        })
+        self.files = []
+
+    def test_post_hooks(self):
+        self.params['post_hooks'] = [self.hook_one, self.hook_two]
+        ydl = YoutubeDL(self.params)
+        ydl.download([TEST_ID])
+        self.assertEqual(self.stored_name_1, EXPECTED_NAME, 'Not the expected name from hook 1')
+        self.assertEqual(self.stored_name_2, EXPECTED_NAME, 'Not the expected name from hook 2')
+
+    def test_post_hook_exception(self):
+        self.params['post_hooks'] = [self.hook_three]
+        ydl = YoutubeDL(self.params)
+        self.assertRaises(DownloadError, ydl.download, [TEST_ID])
+
+    def hook_one(self, filename):
+        self.stored_name_1, _ = os.path.splitext(os.path.basename(filename))
+        self.files.append(filename)
+
+    def hook_two(self, filename):
+        self.stored_name_2, _ = os.path.splitext(os.path.basename(filename))
+        self.files.append(filename)
+
+    def hook_three(self, filename):
+        self.files.append(filename)
+        raise Exception('Test exception for \'%s\'' % filename)
+
+    def tearDown(self):
+        for f in self.files:
+            try_rm(f)
+
+
+if __name__ == '__main__':
+    unittest.main()

youtube_dl/YoutubeDL.py

@@ -264,6 +264,9 @@ class YoutubeDL(object):
                                youtube_dl/postprocessor/__init__.py for a list.
                        as well as any further keyword arguments for the
                        postprocessor.
+    post_hooks:        A list of functions that get called as the final step
+                       for each video file, after all postprocessors have been
+                       called. The filename will be passed as the only argument.
     progress_hooks:    A list of functions that get called on download
                        progress, with a dictionary with the entries
                        * status: One of "downloading", "error", or "finished".
@@ -382,6 +385,7 @@ class YoutubeDL(object):
         self._ies = []
         self._ies_instances = {}
         self._pps = []
+        self._post_hooks = []
         self._progress_hooks = []
         self._download_retcode = 0
         self._num_downloads = 0
@@ -467,6 +471,9 @@ class YoutubeDL(object):
             pp = pp_class(self, **compat_kwargs(pp_def))
             self.add_post_processor(pp)
 
+        for ph in self.params.get('post_hooks', []):
+            self.add_post_hook(ph)
+
         for ph in self.params.get('progress_hooks', []):
             self.add_progress_hook(ph)
 
@@ -519,6 +526,10 @@ class YoutubeDL(object):
         self._pps.append(pp)
         pp.set_downloader(self)
 
+    def add_post_hook(self, ph):
+        """Add the post hook"""
+        self._post_hooks.append(ph)
+
     def add_progress_hook(self, ph):
         """Add the progress hook (currently only for the file downloader)"""
         self._progress_hooks.append(ph)
@@ -2233,6 +2244,12 @@ class YoutubeDL(object):
                 except (PostProcessingError) as err:
                     self.report_error('postprocessing: %s' % error_to_compat_str(err))
                     return
+                try:
+                    for ph in self._post_hooks:
+                        ph(filename)
+                except Exception as err:
+                    self.report_error('post hooks: %s' % str(err))
+                    return
                 self.record_download_archive(info_dict)
                 # avoid possible nugatory search for further items (PR #26638)
                 if self._num_downloads >= max_downloads:

youtube_dl/extractor/bokecc.py

@@ -32,7 +32,7 @@ class BokeCCBaseIE(InfoExtractor):
 
 
 class BokeCCIE(BokeCCBaseIE):
-    _IE_DESC = 'CC视频'
+    IE_DESC = 'CC视频'
     _VALID_URL = r'https?://union\.bokecc\.com/playvideo\.bo\?(?P<query>.*)'
 
     _TESTS = [{

youtube_dl/extractor/cloudy.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec'
+    IE_DESC = 'cloudy.ec'
     _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',

youtube_dl/extractor/common.py

@@ -422,6 +422,8 @@ class InfoExtractor(object):
     _GEO_COUNTRIES = None
     _GEO_IP_BLOCKS = None
     _WORKING = True
+    # supply this in public subclasses: used in supported sites list, etc
+    # IE_DESC = 'short description of IE'
 
     def __init__(self, downloader=None):
         """Constructor. Receives an optional downloader."""

youtube_dl/extractor/itv.py

@@ -35,15 +35,6 @@ from ..utils import (
 
 class ITVBaseIE(InfoExtractor):
 
-    def _search_nextjs_data(self, webpage, video_id, **kw):
-        transform_source = kw.pop('transform_source', None)
-        fatal = kw.pop('fatal', True)
-        return self._parse_json(
-            self._search_regex(
-                r'''<script\b[^>]+\bid=('|")__NEXT_DATA__\1[^>]*>(?P<js>[^<]+)</script>''',
-                webpage, 'next.js data', group='js', fatal=fatal, **kw),
-            video_id, transform_source=transform_source, fatal=fatal)
-
     def __handle_request_webpage_error(self, err, video_id=None, errnote=None, fatal=True):
         if errnote is False:
             return False
@@ -109,7 +100,9 @@ class ITVBaseIE(InfoExtractor):
 
 class ITVIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:(?P<w>watch)|hub)/[^/]+/(?(w)[\w-]+/)(?P<id>\w+)'
-    _IE_DESC = 'ITVX'
+    IE_DESC = 'ITVX'
+    _WORKING = False
+
     _TESTS = [{
         'note': 'Hub URLs redirect to ITVX',
         'url': 'https://www.itv.com/hub/liar/2a4547a0012',
@@ -270,7 +263,7 @@ class ITVIE(ITVBaseIE):
                 'ext': determine_ext(href, 'vtt'),
             })
 
-        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default='{}')
+        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default={})
         video_data.update(traverse_obj(next_data, ('props', 'pageProps', ('title', 'episode')), expected_type=dict)[0] or {})
         title = traverse_obj(video_data, 'headerTitle', 'episodeTitle')
         info = self._og_extract(webpage, require_title=not title)
@@ -323,7 +316,7 @@ class ITVIE(ITVBaseIE):
 
 class ITVBTCCIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?!(?:watch|hub)/)(?:[^/]+/)+(?P<id>[^/?#&]+)'
-    _IE_DESC = 'ITV articles: News, British Touring Car Championship'
+    IE_DESC = 'ITV articles: News, British Touring Car Championship'
     _TESTS = [{
         'note': 'British Touring Car Championship',
         'url': 'https://www.itv.com/btcc/articles/btcc-2018-all-the-action-from-brands-hatch',

youtube_dl/extractor/senateisvp.py

@@ -47,7 +47,7 @@ class SenateISVPIE(InfoExtractor):
         ['vetaff', '76462', 'http://vetaff-f.akamaihd.net'],
         ['arch', '', 'http://ussenate-f.akamaihd.net/']
     ]
-    _IE_NAME = 'senate.gov'
+    IE_NAME = 'senate.gov'
     _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
     _TESTS = [{
         'url': 'http://www.senate.gov/isvp/?comm=judiciary&type=live&stt=&filename=judiciary031715&auto_play=false&wmode=transparent&poster=http%3A%2F%2Fwww.judiciary.senate.gov%2Fthemes%2Fjudiciary%2Fimages%2Fvideo-poster-flash-fit.png',

youtube_dl/jsinterp.py

@@ -686,6 +686,8 @@ class JSInterpreter(object):
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
+        if obj is JS_Undefined:
+            return 'undefined'
         try:
             return json.dumps(obj)
         except TypeError: