Merge b3ab97c05e29bfabfbf6ec889f64e6144730b8d1 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

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

devscripts/zsh-completion.in

@@ -3,26 +3,8 @@
 __youtube_dl() {
     local curcontext="$curcontext" fileopts diropts cur prev
     typeset -A opt_args
-    fileopts="{{fileopts}}"
-    diropts="{{diropts}}"
-    cur=$words[CURRENT]
-    case $cur in
-        :)
-            _arguments '*: :(::ytfavorites ::ytrecommended ::ytsubscriptions ::ytwatchlater ::ythistory)'
-        ;;
-        *)
-            prev=$words[CURRENT-1]
-            if [[ ${prev} =~ ${fileopts} ]]; then
-                _path_files
-            elif [[ ${prev} =~ ${diropts} ]]; then
-                _path_files -/
-            elif [[ ${prev} == "--recode-video" ]]; then
-                _arguments '*: :(mp4 flv ogg webm mkv)'
-            else
-                _arguments '*: :({{flags}})'
-            fi
-        ;;
-    esac
+    _arguments {{args}} \
+      '*: :(::ytfavorites ::ytrecommended ::ytsubscriptions ::ytwatchlater ::ythistory)'
 }
 
-__youtube_dl
+__youtube_dl

devscripts/zsh-completion.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import os
 from os.path import dirname as dirn
 import sys
+import optparse
 
 sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
 import youtube_dl
@@ -17,30 +18,44 @@ ZSH_COMPLETION_TEMPLATE = "devscripts/zsh-completion.in"
 def build_completion(opt_parser):
     opts = [opt for group in opt_parser.option_groups
             for opt in group.option_list]
-    opts_file = [opt for opt in opts if opt.metavar == "FILE"]
-    opts_dir = [opt for opt in opts if opt.metavar == "DIR"]
 
-    fileopts = []
-    for opt in opts_file:
+    # escaping is hard:
+    #  - help may contain colons
+    #  - metavar must have colons : escaped
+    #  - single quotes must be removed
+    #  - help must have square brackets [] escaped
+    def metaparse(opt):
+        if "--recode-video" == opt.get_opt_string():
+            return ":{}:(mp4 flv ogg webm mkv)".format(opt.metavar)
+        if opt.metavar is None:
+            return ""
+        if opt.metavar == "FILE":
+            return ":FILE:_files"
+        if opt.metavar == "DIR":
+            return ":DIR:_directories"
+        else:
+            return ":{}:".format(opt.metavar.replace(":", "\\:"))
+
+    def helpescape(opthelp):
+        if opthelp == optparse.SUPPRESS_HELP:
+            return ""
+        return "[{}]".format(opthelp.replace("'", "\"").replace("]", "\\]").replace("[", "\\["))
+
+    def optionexclude(opt):
+        # When an argument has a long and short version, the arguments entry shall be
+        # _arguments \
+        #   "(-t --thing)"{-t,--thing}"[do things]:WHAT_THING:"
+        # i.e. in parentheses with space the explanation of redundancy and in curly braces
+        # regular shell expansion to create two mostly identical entries.
         if opt._short_opts:
-            fileopts.extend(opt._short_opts)
-        if opt._long_opts:
-            fileopts.extend(opt._long_opts)
+            return "({0} {1})'{{{0},{1}}}'".format(opt._short_opts[0], opt.get_opt_string())
+        return "{}".format(opt.get_opt_string())
 
-    diropts = []
-    for opt in opts_dir:
-        if opt._short_opts:
-            diropts.extend(opt._short_opts)
-        if opt._long_opts:
-            diropts.extend(opt._long_opts)
-
-    flags = [opt.get_opt_string() for opt in opts]
+    mytest = ["'{}{}{}'".format(optionexclude(opt), helpescape(opt.help), metaparse(opt)) for opt in opts]
 
     template = read_file(ZSH_COMPLETION_TEMPLATE)
 
-    template = template.replace("{{fileopts}}", "|".join(fileopts))
-    template = template.replace("{{diropts}}", "|".join(diropts))
-    template = template.replace("{{flags}}", " ".join(flags))
+    template = template.replace("{{args}}", " \\\n      ".join(mytest))
 
     write_file(ZSH_COMPLETION_FILE, template)
 

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: