Add option --print-to-file

Closes yt-dlp#2372

Author: pukkandan

README.md

@@ -686,6 +686,12 @@ You can also fork the project on github and run your fork's [build workflow](.gi
                                      Implies --quiet and --simulate (unless
                                      --no-simulate is used). This option can be
                                      used multiple times
+    --print-to-file [WHEN:]TEMPLATE FILE
+                                     Append given template to the file. The
+                                     values of WHEN and TEMPLATE are same as
+                                     that of --print. FILE uses the same syntax
+                                     as the output template. This option can be
+                                     used multiple times
     -j, --dump-json                  Quiet, but print JSON information for each
                                      video. Simulate unless --no-simulate is
                                      used. See "OUTPUT TEMPLATE" for a

yt_dlp/YoutubeDL.py

@@ -201,9 +201,12 @@ class YoutubeDL(object):
     verbose:           Print additional info to stdout.
     quiet:             Do not print messages to stdout.
     no_warnings:       Do not print out anything for warnings.
-    forceprint:        A dict with keys video/playlist mapped to
-                       a list of templates to force print to stdout
+    forceprint:        A dict with keys WHEN mapped to a list of templates to
+                       print to stdout. The allowed keys are video or any of the
+                       items in utils.POSTPROCESS_WHEN.
                        For compatibility, a single list is also accepted
+    print_to_file:     A dict with keys WHEN (same as forceprint) mapped to
+                       a list of tuples with (template, filename)
     forceurl:          Force printing final URL. (Deprecated)
     forcetitle:        Force printing title. (Deprecated)
     forceid:           Force printing ID. (Deprecated)
@@ -349,8 +352,8 @@ class YoutubeDL(object):
     postprocessors:    A list of dictionaries, each with an entry
                        * key:  The name of the postprocessor. See
                                yt_dlp/postprocessor/__init__.py for a list.
-                       * when: When to run the postprocessor. Can be one of
-                               pre_process|before_dl|post_process|after_move.
+                       * when: When to run the postprocessor. Allowed values are
+                               the entries of utils.POSTPROCESS_WHEN
                                Assumed to be 'post_process' if not given
     post_hooks:        Deprecated - Register a custom postprocessor instead
                        A list of functions that get called as the final step
@@ -592,8 +595,10 @@ def check_deprecated(param, option, suggestion):
         else:
             self.params['nooverwrites'] = not self.params['overwrites']
 
-        # Compatibility with older syntax
         params.setdefault('forceprint', {})
+        params.setdefault('print_to_file', {})
+
+        # Compatibility with older syntax
         if not isinstance(params['forceprint'], dict):
             params['forceprint'] = {'video': params['forceprint']}
 
@@ -2683,19 +2688,32 @@ def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
             subs[lang] = f
         return subs
 
-    def _forceprint(self, tmpl, info_dict):
-        mobj = re.match(r'\w+(=?)$', tmpl)
-        if mobj and mobj.group(1):
-            tmpl = f'{tmpl[:-1]} = %({tmpl[:-1]})r'
-        elif mobj:
-            tmpl = '%({})s'.format(tmpl)
+    def _forceprint(self, key, info_dict):
+        if info_dict is None:
+            return
+        info_copy = info_dict.copy()
+        info_copy['formats_table'] = self.render_formats_table(info_dict)
+        info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
+        info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
+        info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
+
+        def format_tmpl(tmpl):
+            mobj = re.match(r'\w+(=?)$', tmpl)
+            if mobj and mobj.group(1):
+                return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
+            elif mobj:
+                return f'%({tmpl})s'
+            return tmpl
 
-        info_dict = info_dict.copy()
-        info_dict['formats_table'] = self.render_formats_table(info_dict)
-        info_dict['thumbnails_table'] = self.render_thumbnails_table(info_dict)
-        info_dict['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
-        info_dict['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
-        self.to_stdout(self.evaluate_outtmpl(tmpl, info_dict))
+        for tmpl in self.params['forceprint'].get(key, []):
+            self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
+
+        for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
+            filename = self.evaluate_outtmpl(file_tmpl, info_dict)
+            tmpl = format_tmpl(tmpl)
+            self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
+            with io.open(filename, 'a', encoding='utf-8') as f:
+                f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
 
     def __forced_printings(self, info_dict, filename, incomplete):
         def print_mandatory(field, actual_field=None):
@@ -2719,10 +2737,11 @@ def print_optional(field):
         elif 'url' in info_dict:
             info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
 
-        if self.params['forceprint'].get('video') or self.params.get('forcejson'):
+        if (self.params.get('forcejson')
+                or self.params['forceprint'].get('video')
+                or self.params['print_to_file'].get('video')):
             self.post_extract(info_dict)
-        for tmpl in self.params['forceprint'].get('video', []):
-            self._forceprint(tmpl, info_dict)
+        self._forceprint('video', info_dict)
 
         print_mandatory('title')
         print_mandatory('id')
@@ -3290,8 +3309,7 @@ def run_pp(self, pp, infodict):
         return infodict
 
     def run_all_pps(self, key, info, *, additional_pps=None):
-        for tmpl in self.params['forceprint'].get(key, []):
-            self._forceprint(tmpl, info)
+        self._forceprint(key, info)
         for pp in (additional_pps or []) + self._pps[key]:
             info = self.run_pp(pp, info)
         return info

yt_dlp/__init__.py

@@ -356,6 +356,10 @@ def validate_outtmpl(tmpl, msg):
     for type_, tmpl_list in opts.forceprint.items():
         for tmpl in tmpl_list:
             validate_outtmpl(tmpl, f'{type_} print template')
+    for type_, tmpl_list in opts.print_to_file.items():
+        for tmpl, file in tmpl_list:
+            validate_outtmpl(tmpl, f'{type_} print-to-file template')
+            validate_outtmpl(file, f'{type_} print-to-file filename')
     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
     for k, tmpl in opts.progress_template.items():
         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
@@ -663,6 +667,7 @@ def report_deprecation(val, old, new=None):
         'forcefilename': opts.getfilename,
         'forceformat': opts.getformat,
         'forceprint': opts.forceprint,
+        'print_to_file': opts.print_to_file,
         'forcejson': opts.dumpjson or opts.print_json,
         'dump_single_json': opts.dump_single_json,
         'force_write_download_archive': opts.force_write_download_archive,

yt_dlp/options.py

@@ -173,11 +173,16 @@ def _dict_from_options_callback(
             process_key=str.lower, append=False):
 
         out_dict = dict(getattr(parser.values, option.dest))
+        multiple_args = not isinstance(value, str)
         if multiple_keys:
             allowed_keys = r'(%s)(,(%s))*' % (allowed_keys, allowed_keys)
-        mobj = re.match(r'(?i)(?P<keys>%s)%s(?P<val>.*)$' % (allowed_keys, delimiter), value)
+        mobj = re.match(
+            r'(?i)(?P<keys>%s)%s(?P<val>.*)$' % (allowed_keys, delimiter),
+            value[0] if multiple_args else value)
         if mobj is not None:
             keys, val = mobj.group('keys').split(','), mobj.group('val')
+            if multiple_args:
+                val = [val, *value[1:]]
         elif default_key is not None:
             keys, val = [default_key], value
         else:
@@ -923,6 +928,18 @@ def _dict_from_options_callback(
             'Field name or output template to print to screen, optionally prefixed with when to print it, separated by a ":". '
             'Supported values of "WHEN" are the same as that of --use-postprocessor, and "video" (default). '
             'Implies --quiet and --simulate (unless --no-simulate is used). This option can be used multiple times'))
+    verbosity.add_option(
+        '--print-to-file',
+        metavar='[WHEN:]TEMPLATE FILE', dest='print_to_file', default={}, type='str', nargs=2,
+        action='callback', callback=_dict_from_options_callback,
+        callback_kwargs={
+            'allowed_keys': 'video|' + '|'.join(map(re.escape, POSTPROCESS_WHEN)),
+            'default_key': 'video',
+            'multiple_keys': False,
+            'append': True,
+        }, help=(
+            'Append given template to the file. The values of WHEN and TEMPLATE are same as that of --print. '
+            'FILE uses the same syntax as the output template. This option can be used multiple times'))
     verbosity.add_option(
         '-g', '--get-url',
         action='store_true', dest='geturl', default=False,