From cbe4fe09dc516a394c55965e7e9ae487cff0b113 Mon Sep 17 00:00:00 2001 From: MaxwellGoblin Date: Sat, 17 Jun 2017 01:46:31 +0800 Subject: [PATCH] rewrite qq_download_by_vid --- src/you_get/extractors/qq.py | 95 ++++++++++++------------------------ 1 file changed, 31 insertions(+), 64 deletions(-) diff --git a/src/you_get/extractors/qq.py b/src/you_get/extractors/qq.py index 4a9d5f89..e7d5f146 100644 --- a/src/you_get/extractors/qq.py +++ b/src/you_get/extractors/qq.py @@ -3,75 +3,42 @@ __all__ = ['qq_download'] from ..common import * +from ..util.log import * from .qie import download as qieDownload from urllib.parse import urlparse,parse_qs def qq_download_by_vid(vid, title, output_dir='.', merge=True, info_only=False): - info_api = 'http://vv.video.qq.com/getinfo?otype=json&appver=3%2E2%2E19%2E333&platform=11&defnpayver=1&vid=' + vid - info = get_html(info_api) + info_api = 'http://vv.video.qq.com/getinfo?otype=json&appver=3.2.19.333&platform=11&defnpayver=1&vid={}'.format(vid) + info = get_content(info_api) video_json = json.loads(match1(info, r'QZOutputJson=(.*)')[:-1]) - parts_vid = video_json['vl']['vi'][0]['vid'] - parts_ti = video_json['vl']['vi'][0]['ti'] - parts_prefix = video_json['vl']['vi'][0]['ul']['ui'][0]['url'] - parts_formats = video_json['fl']['fi'] - if parts_prefix.endswith('/'): - parts_prefix = parts_prefix[:-1] - # find best quality - # only looking for fhd(1080p) and shd(720p) here. - # 480p usually come with a single file, will be downloaded as fallback. - best_quality = '' - for part_format in parts_formats: - if part_format['name'] == 'fhd': - best_quality = 'fhd' - break - if part_format['name'] == 'shd': - best_quality = 'shd' + if video_json['exem'] != 0: + log.wtf(video_json['msg']) + fn_pre = video_json['vl']['vi'][0]['lnk'] + title = video_json['vl']['vi'][0]['ti'] + host = video_json['vl']['vi'][0]['ul']['ui'][0]['url'] + streams = video_json['fl']['fi'] + seg_cnt = len(video_json['vl']['vi'][0]['cl']['ci']) - for part_format in parts_formats: - if (not best_quality == '') and (not part_format['name'] == best_quality): - continue - part_format_id = part_format['id'] - part_format_sl = part_format['sl'] - if part_format_sl == 0: - part_urls= [] - total_size = 0 - try: - # For fhd(1080p), every part is about 100M and 6 minutes - # try 100 parts here limited download longest single video of 10 hours. - for part in range(1,100): - filename = vid + '.p' + str(part_format_id % 10000) + '.' + str(part) + '.mp4' - key_api = "http://vv.video.qq.com/getkey?otype=json&platform=11&format=%s&vid=%s&filename=%s" % (part_format_id, parts_vid, filename) - #print(filename) - #print(key_api) - part_info = get_html(key_api) - key_json = json.loads(match1(part_info, r'QZOutputJson=(.*)')[:-1]) - #print(key_json) - vkey = key_json['key'] - url = '%s/%s?vkey=%s' % (parts_prefix, filename, vkey) - part_urls.append(url) - _, ext, size = url_info(url, faker=True) - total_size += size - except: - pass - print_info(site_info, parts_ti, ext, total_size) - if not info_only: - download_urls(part_urls, parts_ti, ext, total_size, output_dir=output_dir, merge=merge) - else: - fvkey = video_json['vl']['vi'][0]['fvkey'] - mp4 = video_json['vl']['vi'][0]['cl'].get('ci', None) - if mp4: - old_id = mp4[0]['keyid'].split('.')[1] - new_id = 'p' + str(int(old_id) % 10000) - mp4 = mp4[0]['keyid'].replace(old_id, new_id) + '.mp4' - else: - mp4 = video_json['vl']['vi'][0]['fn'] - url = '%s/%s?vkey=%s' % ( parts_prefix, mp4, fvkey ) - _, ext, size = url_info(url, faker=True) + best_quality = streams[-1]['name'] + part_format_id = streams[-1]['id'] - print_info(site_info, title, ext, size) - if not info_only: - download_urls([url], title, ext, size, output_dir=output_dir, merge=merge) + part_urls= [] + total_size = 0 + for part in range(1, seg_cnt+1): + filename = fn_pre + '.p' + str(part_format_id % 10000) + '.' + str(part) + '.mp4' + key_api = "http://vv.video.qq.com/getkey?otype=json&platform=11&format={}&vid={}&filename={}&appver=3.2.19.333".format(part_format_id, vid, filename) + part_info = get_content(key_api) + key_json = json.loads(match1(part_info, r'QZOutputJson=(.*)')[:-1]) + vkey = key_json['key'] + url = '{}{}?vkey={}'.format(host, filename, vkey) + part_urls.append(url) + _, ext, size = url_info(url) + total_size += size + + print_info(site_info, title, ext, total_size) + if not info_only: + download_urls(part_urls, title, ext, total_size, output_dir=output_dir, merge=merge) def kg_qq_download_by_shareid(shareid, output_dir='.', info_only=False, caption=False): BASE_URL = 'http://cgi.kg.qq.com/fcgi-bin/kg_ugc_getdetail' @@ -124,7 +91,7 @@ def qq_download(url, output_dir='.', merge=True, info_only=False, **kwargs): return if 'mp.weixin.qq.com/s?' in url: - content = get_html(url) + content = get_content(url) vids = matchall(content, [r'\bvid=(\w+)']) for vid in vids: qq_download_by_vid(vid, vid, output_dir, merge, info_only) @@ -143,7 +110,7 @@ def qq_download(url, output_dir='.', merge=True, info_only=False, **kwargs): url = new_url if 'kuaibao.qq.com' in url or re.match(r'http://daxue.qq.com/content/content/id/\d+', url): - content = get_html(url) + content = get_content(url) vid = match1(content, r'vid\s*=\s*"\s*([^"]+)"') title = match1(content, r'title">([^"]+)

') title = title.strip() if title else vid @@ -152,7 +119,7 @@ def qq_download(url, output_dir='.', merge=True, info_only=False, **kwargs): # for embedded URLs; don't know what the title is title = vid else: - content = get_html(url) + content = get_content(url) vid = parse_qs(urlparse(url).query).get('vid') #for links specified vid like http://v.qq.com/cover/p/ps6mnfqyrfo7es3.html?vid=q0181hpdvo5 vid = vid[0] if vid else match1(content, r'vid"*\s*:\s*"\s*([^"]+)"') #general fallback title = match1(content,r''%vid)