diff options
Diffstat (limited to 'youtube_dlc')
| -rw-r--r-- | youtube_dlc/YoutubeDL.py | 61 | ||||
| -rw-r--r-- | youtube_dlc/__init__.py | 4 | ||||
| -rw-r--r-- | youtube_dlc/compat.py | 26 | ||||
| -rw-r--r-- | youtube_dlc/options.py | 21 | ||||
| -rw-r--r-- | youtube_dlc/utils.py | 81 |
5 files changed, 190 insertions, 3 deletions
diff --git a/youtube_dlc/YoutubeDL.py b/youtube_dlc/YoutubeDL.py index ee6d74910..97e4f451f 100644 --- a/youtube_dlc/YoutubeDL.py +++ b/youtube_dlc/YoutubeDL.py @@ -51,6 +51,9 @@ from .utils import ( DEFAULT_OUTTMPL, determine_ext, determine_protocol, + DOT_DESKTOP_LINK_TEMPLATE, + DOT_URL_LINK_TEMPLATE, + DOT_WEBLOC_LINK_TEMPLATE, DownloadError, encode_compat_str, encodeFilename, @@ -61,6 +64,7 @@ from .utils import ( formatSeconds, GeoRestrictedError, int_or_none, + iri_to_uri, ISO3166Utils, locked_file, make_HTTPS_handler, @@ -84,6 +88,7 @@ from .utils import ( std_headers, str_or_none, subtitles_filename, + to_high_limit_path, UnavailableVideoError, url_basename, version_tuple, @@ -187,6 +192,11 @@ class YoutubeDL(object): writeannotations: Write the video annotations to a .annotations.xml file writethumbnail: Write the thumbnail image to a file write_all_thumbnails: Write all thumbnail formats to files + writelink: Write an internet shortcut file, depending on the + current platform (.url/.webloc/.desktop) + writeurllink: Write a Windows internet shortcut file (.url) + writewebloclink: Write a macOS internet shortcut file (.webloc) + writedesktoplink: Write a Linux internet shortcut file (.desktop) writesubtitles: Write the video subtitles to a file writeautomaticsub: Write the automatically generated subtitles to a file allsubtitles: Downloads all the subtitles of the video @@ -1984,6 +1994,57 @@ class YoutubeDL(object): self._write_thumbnails(info_dict, filename) + # Write internet shortcut files + url_link = webloc_link = desktop_link = False + if self.params.get('writelink', False): + if sys.platform == "darwin": # macOS. + webloc_link = True + elif sys.platform.startswith("linux"): + desktop_link = True + else: # if sys.platform in ['win32', 'cygwin']: + url_link = True + if self.params.get('writeurllink', False): + url_link = True + if self.params.get('writewebloclink', False): + webloc_link = True + if self.params.get('writedesktoplink', False): + desktop_link = True + + if url_link or webloc_link or desktop_link: + if 'webpage_url' not in info_dict: + self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information') + return + ascii_url = iri_to_uri(info_dict['webpage_url']) + + def _write_link_file(extension, template, newline, embed_filename): + linkfn = replace_extension(filename, extension, info_dict.get('ext')) + if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(linkfn)): + self.to_screen('[info] Internet shortcut is already present') + else: + try: + self.to_screen('[info] Writing internet shortcut to: ' + linkfn) + with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile: + template_vars = {'url': ascii_url} + if embed_filename: + template_vars['filename'] = linkfn[:-(len(extension) + 1)] + linkfile.write(template % template_vars) + except (OSError, IOError): + self.report_error('Cannot write internet shortcut ' + linkfn) + return False + return True + + if url_link: + if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False): + return + if webloc_link: + if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False): + return + if desktop_link: + if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True): + return + + # Download + must_record_download_archive = False if not self.params.get('skip_download', False): try: if info_dict.get('requested_formats') is not None: diff --git a/youtube_dlc/__init__.py b/youtube_dlc/__init__.py index df07016e1..d183016b6 100644 --- a/youtube_dlc/__init__.py +++ b/youtube_dlc/__init__.py @@ -389,6 +389,10 @@ def _real_main(argv=None): 'writeinfojson': opts.writeinfojson, 'writethumbnail': opts.writethumbnail, 'write_all_thumbnails': opts.write_all_thumbnails, + 'writelink': opts.writelink, + 'writeurllink': opts.writeurllink, + 'writewebloclink': opts.writewebloclink, + 'writedesktoplink': opts.writedesktoplink, 'writesubtitles': opts.writesubtitles, 'writeautomaticsub': opts.writeautomaticsub, 'allsubtitles': opts.allsubtitles, diff --git a/youtube_dlc/compat.py b/youtube_dlc/compat.py index ac889ddd7..4a69b098f 100644 --- a/youtube_dlc/compat.py +++ b/youtube_dlc/compat.py @@ -38,14 +38,19 @@ except ImportError: # Python 2 import urllib as compat_urllib_parse try: + import urllib.parse as compat_urlparse +except ImportError: # Python 2 + import urlparse as compat_urlparse + +try: from urllib.parse import urlparse as compat_urllib_parse_urlparse except ImportError: # Python 2 from urlparse import urlparse as compat_urllib_parse_urlparse try: - import urllib.parse as compat_urlparse + from urllib.parse import urlunparse as compat_urllib_parse_urlunparse except ImportError: # Python 2 - import urlparse as compat_urlparse + from urlparse import urlunparse as compat_urllib_parse_urlunparse try: import urllib.response as compat_urllib_response @@ -2366,6 +2371,20 @@ except NameError: compat_str = str try: + from urllib.parse import quote as compat_urllib_parse_quote + from urllib.parse import quote_plus as compat_urllib_parse_quote_plus +except ImportError: # Python 2 + def compat_urllib_parse_quote(string, safe='/'): + return compat_urllib_parse.quote( + string.encode('utf-8'), + str(safe)) + + def compat_urllib_parse_quote_plus(string, safe=''): + return compat_urllib_parse.quote_plus( + string.encode('utf-8'), + str(safe)) + +try: from urllib.parse import unquote_to_bytes as compat_urllib_parse_unquote_to_bytes from urllib.parse import unquote as compat_urllib_parse_unquote from urllib.parse import unquote_plus as compat_urllib_parse_unquote_plus @@ -3033,11 +3052,14 @@ __all__ = [ 'compat_tokenize_tokenize', 'compat_urllib_error', 'compat_urllib_parse', + 'compat_urllib_parse_quote', + 'compat_urllib_parse_quote_plus', 'compat_urllib_parse_unquote', 'compat_urllib_parse_unquote_plus', 'compat_urllib_parse_unquote_to_bytes', 'compat_urllib_parse_urlencode', 'compat_urllib_parse_urlparse', + 'compat_urllib_parse_urlunparse', 'compat_urllib_request', 'compat_urllib_request_DataHandler', 'compat_urllib_response', diff --git a/youtube_dlc/options.py b/youtube_dlc/options.py index 44eba3e9c..bd85abd3a 100644 --- a/youtube_dlc/options.py +++ b/youtube_dlc/options.py @@ -830,7 +830,25 @@ def parseOpts(overrideArguments=None): action='store_true', dest='list_thumbnails', default=False, help='Simulate and list all available thumbnail formats') - postproc = optparse.OptionGroup(parser, 'Post-processing Options') + link = optparse.OptionGroup(parser, 'Internet Shortcut Options') + link.add_option( + '--write-link', + action='store_true', dest='writelink', default=False, + help='Write an internet shortcut file, depending on the current platform (.url/.webloc/.desktop). The URL may be cached by the OS.') + link.add_option( + '--write-url-link', + action='store_true', dest='writeurllink', default=False, + help='Write a Windows internet shortcut file (.url). Note that the OS caches the URL based on the file path.') + link.add_option( + '--write-webloc-link', + action='store_true', dest='writewebloclink', default=False, + help='Write a macOS internet shortcut file (.webloc)') + link.add_option( + '--write-desktop-link', + action='store_true', dest='writedesktoplink', default=False, + help='Write a Linux internet shortcut file (.desktop)') + + postproc = optparse.OptionGroup(parser, 'Post-Processing Options') postproc.add_option( '-x', '--extract-audio', action='store_true', dest='extractaudio', default=False, @@ -932,6 +950,7 @@ def parseOpts(overrideArguments=None): parser.add_option_group(downloader) parser.add_option_group(filesystem) parser.add_option_group(thumbnail) + parser.add_option_group(link) parser.add_option_group(verbosity) parser.add_option_group(workarounds) parser.add_option_group(video_format) diff --git a/youtube_dlc/utils.py b/youtube_dlc/utils.py index 68b4ca944..d814eb2ac 100644 --- a/youtube_dlc/utils.py +++ b/youtube_dlc/utils.py @@ -60,6 +60,9 @@ from .compat import ( compat_urllib_parse, compat_urllib_parse_urlencode, compat_urllib_parse_urlparse, + compat_urllib_parse_urlunparse, + compat_urllib_parse_quote, + compat_urllib_parse_quote_plus, compat_urllib_parse_unquote_plus, compat_urllib_request, compat_urlparse, @@ -5714,3 +5717,81 @@ def random_birthday(year_field, month_field, day_field): month_field: str(random_date.month), day_field: str(random_date.day), } + +# Templates for internet shortcut files, which are plain text files. +DOT_URL_LINK_TEMPLATE = ''' +[InternetShortcut] +URL=%(url)s +'''.lstrip() + +DOT_WEBLOC_LINK_TEMPLATE = ''' +<?xml version="1.0" encoding="UTF-8"?> +<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> +<plist version="1.0"> +<dict> +\t<key>URL</key> +\t<string>%(url)s</string> +</dict> +</plist> +'''.lstrip() + +DOT_DESKTOP_LINK_TEMPLATE = ''' +[Desktop Entry] +Encoding=UTF-8 +Name=%(filename)s +Type=Link +URL=%(url)s +Icon=text-html +'''.lstrip() + + +def iri_to_uri(iri): + """ + Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only). + + The function doesn't add an additional layer of escaping; e.g., it doesn't escape `%3C` as `%253C`. Instead, it percent-escapes characters with an underlying UTF-8 encoding *besides* those already escaped, leaving the URI intact. + """ + + iri_parts = compat_urllib_parse_urlparse(iri) + + if '[' in iri_parts.netloc: + raise ValueError('IPv6 URIs are not, yet, supported.') + # Querying `.netloc`, when there's only one bracket, also raises a ValueError. + + # The `safe` argument values, that the following code uses, contain the characters that should not be percent-encoded. Everything else but letters, digits and '_.-' will be percent-encoded with an underlying UTF-8 encoding. Everything already percent-encoded will be left as is. + + net_location = '' + if iri_parts.username: + net_location += compat_urllib_parse_quote(iri_parts.username, safe=r"!$%&'()*+,~") + if iri_parts.password is not None: + net_location += ':' + compat_urllib_parse_quote(iri_parts.password, safe=r"!$%&'()*+,~") + net_location += '@' + + net_location += iri_parts.hostname.encode('idna').decode('utf-8') # Punycode for Unicode hostnames. + # The 'idna' encoding produces ASCII text. + if iri_parts.port is not None and iri_parts.port != 80: + net_location += ':' + str(iri_parts.port) + + return compat_urllib_parse_urlunparse( + (iri_parts.scheme, + net_location, + + compat_urllib_parse_quote_plus(iri_parts.path, safe=r"!$%&'()*+,/:;=@|~"), + + # Unsure about the `safe` argument, since this is a legacy way of handling parameters. + compat_urllib_parse_quote_plus(iri_parts.params, safe=r"!$%&'()*+,/:;=@|~"), + + # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component. + compat_urllib_parse_quote_plus(iri_parts.query, safe=r"!$%&'()*+,/:;=?@{|}~"), + + compat_urllib_parse_quote_plus(iri_parts.fragment, safe=r"!#$%&'()*+,/:;=?@{|}~"))) + + # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes. + + +def to_high_limit_path(path): + if sys.platform in ['win32', 'cygwin']: + # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited. + return r'\\?\ '.rstrip() + os.path.abspath(path) + + return path |