diff options
Diffstat (limited to 'youtube_dl/downloader')
| -rw-r--r-- | youtube_dl/downloader/f4m.py | 25 | ||||
| -rw-r--r-- | youtube_dl/downloader/fragment.py | 22 | ||||
| -rw-r--r-- | youtube_dl/downloader/hls.py | 32 | ||||
| -rw-r--r-- | youtube_dl/downloader/http.py | 3 |
4 files changed, 62 insertions, 20 deletions
diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index c8fde9a..fdb80f4 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -243,8 +243,17 @@ def remove_encrypted_media(media): media)) -def _add_ns(prop): - return '{http://ns.adobe.com/f4m/1.0}%s' % prop +def _add_ns(prop, ver=1): + return '{http://ns.adobe.com/f4m/%d.0}%s' % (ver, prop) + + +def get_base_url(manifest): + base_url = xpath_text( + manifest, [_add_ns('baseURL'), _add_ns('baseURL', 2)], + 'base URL', default=None) + if base_url: + base_url = base_url.strip() + return base_url class F4mFD(FragmentFD): @@ -330,13 +339,13 @@ class F4mFD(FragmentFD): rate, media = list(filter( lambda f: int(f[0]) == requested_bitrate, formats))[0] - base_url = compat_urlparse.urljoin(man_url, media.attrib['url']) + # Prefer baseURL for relative URLs as per 11.2 of F4M 3.0 spec. + man_base_url = get_base_url(doc) or man_url + + base_url = compat_urlparse.urljoin(man_base_url, media.attrib['url']) bootstrap_node = doc.find(_add_ns('bootstrapInfo')) - # From Adobe F4M 3.0 spec: - # The <baseURL> element SHALL be the base URL for all relative - # (HTTP-based) URLs in the manifest. If <baseURL> is not present, said - # URLs should be relative to the location of the containing document. - boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, man_url) + boot_info, bootstrap_url = self._parse_bootstrap_node( + bootstrap_node, man_base_url) live = boot_info['live'] metadata_node = media.find(_add_ns('metadata')) if metadata_node is not None: diff --git a/youtube_dl/downloader/fragment.py b/youtube_dl/downloader/fragment.py index bccc8ec..ea5e3a4 100644 --- a/youtube_dl/downloader/fragment.py +++ b/youtube_dl/downloader/fragment.py @@ -107,19 +107,26 @@ class FragmentFD(FileDownloader): def _append_fragment(self, ctx, frag_content): try: ctx['dest_stream'].write(frag_content) + ctx['dest_stream'].flush() finally: if self.__do_ytdl_file(ctx): self._write_ytdl_file(ctx) if not self.params.get('keep_fragments', False): - os.remove(ctx['fragment_filename_sanitized']) + os.remove(encodeFilename(ctx['fragment_filename_sanitized'])) del ctx['fragment_filename_sanitized'] def _prepare_frag_download(self, ctx): if 'live' not in ctx: ctx['live'] = False + if not ctx['live']: + total_frags_str = '%d' % ctx['total_frags'] + ad_frags = ctx.get('ad_frags', 0) + if ad_frags: + total_frags_str += ' (not including %d ad)' % ad_frags + else: + total_frags_str = 'unknown (live)' self.to_screen( - '[%s] Total fragments: %s' - % (self.FD_NAME, ctx['total_frags'] if not ctx['live'] else 'unknown (live)')) + '[%s] Total fragments: %s' % (self.FD_NAME, total_frags_str)) self.report_destination(ctx['filename']) dl = HttpQuietDownloader( self.ydl, @@ -151,10 +158,15 @@ class FragmentFD(FileDownloader): if self.__do_ytdl_file(ctx): if os.path.isfile(encodeFilename(self.ytdl_filename(ctx['filename']))): self._read_ytdl_file(ctx) + if ctx['fragment_index'] > 0 and resume_len == 0: + self.report_warning( + 'Inconsistent state of incomplete fragment download. ' + 'Restarting from the beginning...') + ctx['fragment_index'] = resume_len = 0 + self._write_ytdl_file(ctx) else: self._write_ytdl_file(ctx) - if ctx['fragment_index'] > 0: - assert resume_len > 0 + assert ctx['fragment_index'] == 0 dest_stream, tmpfilename = sanitize_open(tmpfilename, open_mode) diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index 46308cf..4dc3ab4 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -75,15 +75,30 @@ class HlsFD(FragmentFD): fd.add_progress_hook(ph) return fd.real_download(filename, info_dict) - total_frags = 0 + def anvato_ad(s): + return s.startswith('#ANVATO-SEGMENT-INFO') and 'type=ad' in s + + media_frags = 0 + ad_frags = 0 + ad_frag_next = False for line in s.splitlines(): line = line.strip() - if line and not line.startswith('#'): - total_frags += 1 + if not line: + continue + if line.startswith('#'): + if anvato_ad(line): + ad_frags += 1 + ad_frag_next = True + continue + if ad_frag_next: + ad_frag_next = False + continue + media_frags += 1 ctx = { 'filename': filename, - 'total_frags': total_frags, + 'total_frags': media_frags, + 'ad_frags': ad_frags, } self._prepare_and_start_frag_download(ctx) @@ -101,10 +116,14 @@ class HlsFD(FragmentFD): decrypt_info = {'METHOD': 'NONE'} byte_range = {} frag_index = 0 + ad_frag_next = False for line in s.splitlines(): line = line.strip() if line: if not line.startswith('#'): + if ad_frag_next: + ad_frag_next = False + continue frag_index += 1 if frag_index <= ctx['fragment_index']: continue @@ -144,7 +163,8 @@ class HlsFD(FragmentFD): return False if decrypt_info['METHOD'] == 'AES-128': iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence) - decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(decrypt_info['URI']).read() + decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen( + self._prepare_url(info_dict, decrypt_info['URI'])).read() frag_content = AES.new( decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content) self._append_fragment(ctx, frag_content) @@ -175,6 +195,8 @@ class HlsFD(FragmentFD): 'start': sub_range_start, 'end': sub_range_start + int(splitted_byte_range[0]), } + elif anvato_ad(line): + ad_frag_next = True self._finish_frag_download(ctx) diff --git a/youtube_dl/downloader/http.py b/youtube_dl/downloader/http.py index 8a6638c..3ff26ff 100644 --- a/youtube_dl/downloader/http.py +++ b/youtube_dl/downloader/http.py @@ -284,8 +284,7 @@ class HttpFD(FileDownloader): while count <= retries: try: establish_connection() - download() - return True + return download() except RetryDownload as e: count += 1 if count <= retries: |