def get_subtitles(self, video_name, sub_num=5):
print(prefix + ' Searching ZIMUZU...', end='\r')
keywords, info_dict = Downloader.get_keywords(video_name)
keyword = ' '.join(keywords)
sub_dict = order_dict()
s = requests.session()
while True:
# 当前关键字查询
r = s.get(ZimuzuDownloader.search_url.format(keyword),
headers=Downloader.header,
timeout=10)
bs_obj = BeautifulSoup(r.text, 'html.parser')
tab_text = bs_obj.find('div', {'class': 'article-tab'}).text
tab_text = tab_text.encode('utf8') if py == 2 else tab_text
if '字幕(0)' not in tab_text:
for one_box in bs_obj.find_all('div',
{'class': 'search-item'}):
sub_name = ZimuzuDownloader.choice_prefix + \
one_box.find('strong', {'class': 'list_title'}).text
sub_name = sub_name.encode('utf8') if py == 2 else sub_name
if info_dict['type'] == 'movie' and '美剧字幕' in sub_name:
continue
a = one_box.find('a')
text = a.text.encode('utf8') if py == 2 else a.text
sub_url = ZimuzuDownloader.site_url + a.attrs['href']
type_score = 0
type_score += ('英文' in text) * 1
type_score += ('繁体' in text) * 2
type_score += ('简体' in text) * 4
type_score += ('中英' in text) * 8
sub_dict[sub_name] = {
'lan': type_score,
'link': sub_url,
'session': None
}
if len(sub_dict) >= sub_num:
del keywords[:] # 字幕条数达到上限，清空keywords
break
if len(keywords) > 1: # 字幕数未满，更换关键词继续查询
keyword = keyword.replace(keywords[-1], '')
keywords.pop(-1)
continue
break
# 第一个候选字幕没有双语
if (len(sub_dict.items()) > 0
and list(sub_dict.items())[0][1]['lan'] < 8):
sub_dict = order_dict(
sorted(sub_dict.items(),
key=lambda e: e[1]['lan'],
reverse=True))
return sub_dict
def get_subtitles(self, video_name, sub_num=5):
print(PREFIX + " Searching ZIMUZU...", end="\r")
keywords, info_dict = Downloader.get_keywords(video_name)
keyword = " ".join(keywords)
sub_dict = order_dict()
s = requests.session()
while True:
# 当前关键字查询
r = s.get(
ZimuzuDownloader.search_url.format(keyword),
headers=Downloader.header,
timeout=10,
)
bs_obj = BeautifulSoup(r.text, "html.parser")
tab_text = bs_obj.find("div", {"class": "article-tab"}).text
if "字幕(0)" not in tab_text:
for one_box in bs_obj.find_all("div", {"class": "search-item"}):
sub_name = (
ZimuzuDownloader.choice_prefix
+ one_box.find("strong", {"class": "list_title"}).text
)
if info_dict["type"] == "movie" and "美剧字幕" in sub_name:
continue
a = one_box.find("a")
text = a.text
sub_url = ZimuzuDownloader.site_url + a.attrs["href"]
type_score = 0
type_score += ("英文" in text) * 1
type_score += ("繁体" in text) * 2
type_score += ("简体" in text) * 4
type_score += ("中英" in text) * 8
sub_dict[sub_name] = {
"lan": type_score,
"link": sub_url,
"session": None,
}
if len(sub_dict) >= sub_num:
del keywords[:] # 字幕条数达到上限，清空keywords
break
if len(keywords) > 1: # 字幕数未满，更换关键词继续查询
keyword = keyword.replace(keywords[-1], "")
keywords.pop(-1)
continue
break
# 第一个候选字幕没有双语
if len(sub_dict.items()) > 0 and list(sub_dict.items())[0][1]["lan"] < 8:
sub_dict = order_dict(
sorted(sub_dict.items(), key=lambda e: e[1]["lan"], reverse=True)
)
return sub_dict
def test_get_keywords(self):
"""
Test video info extracting
"""
names = (
'Show.S01E01.ShowName.1080p.AMZN.WEB-DL.DDP5.1.H.264-GRP.mkv',
'Hanzawa.Naoki.Ep10.Final.Chi_Jap.BDrip.1280X720-ZhuixinFan.mp4',
'Homeland.S02E12.PROPER.720p.HDTV.x264-EVOLVE.mkv',
'La.La.Land.2016.1080p.BluRay.x264.Atmos.TrueHD.7.1-HDChina.mkv')
results = (['Show%20s01', 'e01', 'Web', 'GRP', 'amzn', '1080p'], [
'Hanzawa%20Naoki', 'e10', 'Bluray', 'ZhuixinFan', '720p'
], ['Homeland%20s02', 'e12', 'HDTV', 'EVOLVE',
'720p'], ['La%20La%20Land', '2016', 'Bluray', 'HDChina', '1080p'])
for n, r in zip(names, results):
self.assertEqual(Downloader.get_keywords(n)[0], r)
def test_get_keywords(self):
"""
Test video info extracting
"""
names = (
"Show.S01E01.ShowName.1080p.AMZN.WEB-DL.DDP5.1.H.264-GRP.mkv",
"Hanzawa.Naoki.Ep10.Final.Chi_Jap.BDrip.1280X720-ZhuixinFan.mp4",
"Homeland.S02E12.PROPER.720p.HDTV.x264-EVOLVE.mkv",
"La.La.Land.2016.1080p.BluRay.x264.Atmos.TrueHD.7.1-HDChina.mkv",
)
results = (
["Show", "s01", "e01", "Web", "GRP", "amzn", "1080p"],
["Hanzawa%20Naoki", "e10", "Bluray", "ZhuixinFan", "720p"],
["Homeland", "s02", "e12", "HDTV", "EVOLVE", "720p"],
["La%20La%20Land", "2016", "Bluray", "HDChina", "1080p"],
)
for n, r in zip(names, results):
self.assertEqual(Downloader.get_keywords(n)[0], r)
def get_subtitles(self, video_name, sub_num=5):
print("Searching SUBHD...", end="\r")
keywords, info_dict = Downloader.get_keywords(video_name)
keyword = " ".join(keywords)
sub_dict = order_dict()
s = requests.session()
s.headers.update(Downloader.header)
while True:
# 当前关键字查询
r = s.get(
SubHDDownloader.search_url + keyword,
timeout=10,
)
bs_obj = BeautifulSoup(r.text, "html.parser")
try:
small_text = bs_obj.find("small").text
except AttributeError:
char_error = "The URI you submitted has disallowed characters"
if char_error in bs_obj.text:
print("[SUBHD ERROR] " + char_error + ": " + keyword)
return sub_dict
# 搜索验证按钮
time.sleep(2)
continue
if "总共 0 条" not in small_text:
results = bs_obj.find_all(
"div", class_="mb-4 bg-white rounded shadow-sm")
for one_box in results:
if info_dict["type"] == "movie" and not one_box.find(
"div",
class_="px-1 rounded-sm bg-danger text-white"):
continue
a = one_box.find("div", class_="f12 pt-1").find("a")
sub_url = SubHDDownloader.site_url + a.attrs["href"]
sub_name = SubHDDownloader.choice_prefix + a.text
text = one_box.text
if "/a" in a.attrs["href"]:
type_score = 0
type_score += ("英文" in text) * 1
type_score += ("繁体" in text) * 2
type_score += ("简体" in text) * 4
type_score += ("双语" in text) * 8
sub_dict[sub_name] = {
"lan": type_score,
"link": sub_url,
"session": None,
}
if len(sub_dict) >= sub_num:
del keywords[:] # 字幕条数达到上限，清空keywords
break
if len(keywords) > 1: # 字幕数未满，更换关键词继续查询
keyword = keyword.replace(keywords[-1], "")
keywords.pop(-1)
continue
break
if len(sub_dict.items()) > 0 and list(
sub_dict.items())[0][1]["lan"] < 8:
# 第一个候选字幕没有双语
sub_dict = order_dict(
sorted(sub_dict.items(),
key=lambda e: e[1]["lan"],
reverse=True))
return sub_dict
def get_subtitles(self, video_name, sub_num=5):
print(prefix + ' Searching SUBHD...', end='\r')
keywords, info_dict = Downloader.get_keywords(video_name)
keyword = ' '.join(keywords)
sub_dict = order_dict()
s = requests.session()
while True:
# 当前关键字查询
r = s.get(SubHDDownloader.search_url + keyword,
headers=Downloader.header,
timeout=10)
bs_obj = BeautifulSoup(r.text, 'html.parser')
try:
small_text = bs_obj.find('small').text
except AttributeError as e:
char_error = 'The URI you submitted has disallowed characters'
if char_error in bs_obj.text:
print(prefix + ' [SUBHD ERROR] ' + char_error + ': ' +
keyword)
return sub_dict
# 搜索验证按钮
time.sleep(2)
continue
if "总共 0 条" not in small_text:
results = bs_obj.find_all(
"div", class_="mb-4 bg-white rounded shadow-sm")
for one_box in results:
if info_dict['type'] == 'movie' \
and not one_box.find('div', class_="px-1 rounded-sm bg-danger text-white"):
continue
a = one_box.find('div', class_="f12 pt-1").find('a')
sub_url = SubHDDownloader.site_url + a.attrs['href']
sub_name = SubHDDownloader.choice_prefix + a.text
text = one_box.text
if '/a' in a.attrs['href']:
type_score = 0
type_score += ('英文' in text) * 1
type_score += ('繁体' in text) * 2
type_score += ('简体' in text) * 4
type_score += ('双语' in text) * 8
sub_dict[sub_name] = {
'lan': type_score,
'link': sub_url,
'session': None
}
if len(sub_dict) >= sub_num:
del keywords[:] # 字幕条数达到上限，清空keywords
break
if len(keywords) > 1: # 字幕数未满，更换关键词继续查询
keyword = keyword.replace(keywords[-1], '')
keywords.pop(-1)
continue
break
if (len(sub_dict.items()) > 0
and list(sub_dict.items())[0][1]['lan'] < 8):
# 第一个候选字幕没有双语
sub_dict = order_dict(
sorted(sub_dict.items(),
key=lambda e: e[1]['lan'],
reverse=True))
return sub_dict
def get_subtitles(self, video_name, sub_num=5):
print(prefix + ' Searching SUBHD...', end='\r')
keywords, info_dict = Downloader.get_keywords(video_name)
keyword = ' '.join(keywords)
sub_dict = order_dict()
s = requests.session()
while True:
# 当前关键字查询
r = s.get(SubHDDownloader.search_url + keyword,
headers=Downloader.header,
timeout=10)
bs_obj = BeautifulSoup(r.text, 'html.parser')
try:
if py == 2:
small_text = bs_obj.find('small').text.encode('utf8')
else:
small_text = bs_obj.find('small').text
except AttributeError as e:
char_error = 'The URI you submitted has disallowed characters'
if char_error in bs_obj.text:
print(prefix + ' [SUBHD ERROR] ' + char_error + ': ' +
keyword)
return sub_dict
# 搜索验证按钮
time.sleep(2)
continue
if '总共 0 条' not in small_text:
for one_box in bs_obj.find_all('div', {'class': 'box'}):
if info_dict['type'] == 'movie' \
and not one_box.find('div', {'class': 'movielist'}):
continue
a = one_box.find('div', {'class': 'd_title'}).find('a')
sub_url = SubHDDownloader.site_url + a.attrs['href']
sub_name = SubHDDownloader.choice_prefix + a.text.encode('utf8') if py == 2 \
else SubHDDownloader.choice_prefix + a.text
if py == 2:
text = one_box.text.encode('utf8')
else:
text = one_box.text
if '/ar' in a.attrs['href']:
type_score = 0
type_score += ('英文' in text) * 1
type_score += ('繁体' in text) * 2
type_score += ('简体' in text) * 4
type_score += ('双语' in text) * 8
# no_dot_text=text.replace('.',' ').lower()
# for qkeyword in keywords:
# if no_dot_text.find(qkeyword.strip().lower()) != -1:
# type_score += 3
sub_dict[sub_name] = {
'lan': type_score,
'link': sub_url,
'session': None
}
if len(sub_dict) >= sub_num:
del keywords[:] # 字幕条数达到上限，清空keywords
break
if len(keywords) > 1: # 字幕数未满，更换关键词继续查询
keyword = keyword.replace(keywords[-1], '')
keywords.pop(-1)
continue
break
if (len(sub_dict.items()) > 0
and list(sub_dict.items())[0][1]['lan'] < 8):
# 第一个候选字幕没有双语
sub_dict = order_dict(
sorted(sub_dict.items(),
key=lambda e: e[1]['lan'],
reverse=True))
return sub_dict
def get_subtitles(self, video_name, sub_num=10):
print(prefix + ' Searching ZIMUKU...', end='\r')
keywords, info_dict = Downloader.get_keywords(video_name)
keyword = ' '.join(keywords)
info = guessit(keyword)
keywords.pop(0)
keywords.insert(0, info['title'])
if info.get('season'):
season = str(info['season']).zfill(2)
keywords.insert(1, 's' + season)
sub_dict = order_dict()
s = requests.session()
s.headers.update(Downloader.header)
while True:
# 当前关键字搜索
r = s.get(ZimukuDownloader.search_url + keyword, timeout=10)
html = r.text
if '搜索不到相关字幕' not in html:
bs_obj = BeautifulSoup(r.text, 'html.parser')
if bs_obj.find('div', {'class': 'item'}):
# 综合搜索页面
for item in bs_obj.find_all('div', {'class': 'item'}):
title_boxes = item.find(
'div', {'class': 'title'}).find_all('p')
title_box = title_boxes[0]
sub_title_box = title_boxes[1]
item_title = title_box.text
item_sub_title = sub_title_box.text
item_info = guessit(item_title)
if info.get('year') and item_info.get('year'):
if info['year'] != item_info['year']:
# 年份不匹配，跳过
continue
item_titles = [
item_info.get('title', '').lower(),
item_info.get('alternative_title', '').lower()
] + item_sub_title.lower().strip().split(',')
title_included = sum([
1 for _ in item_sub_title
if info['title'].lower() not in _
])
if title_included == 0:
# guessit抽取标题不匹配，跳过
item_title_split = \
[one.split() for one in item_titles]
info_title_split = info['title'].lower().split()
sum1 = sum([1 for _ in info_title_split
if _ in item_title_split[0]])
sum2 = sum([1 for _ in info_title_split
if _ in item_title_split[1]])
if not (sum1 / len(info_title_split) >= 0.5
or sum2 / len(info_title_split) >= 0.5):
# 标题不匹配，跳过
continue
for a in item.find_all('td', {'class': 'first'})[:3]:
a = a.a
a_link = ZimukuDownloader.site_url + \
a.attrs['href']
a_title = a.text
a_title = ZimukuDownloader.choice_prefix + a_title
sub_dict[a_title] = {'type': 'default',
'link': a_link}
elif bs_obj.find('div', {'class': 'persub'}):
# 射手字幕页面
for persub in bs_obj.find_all('div', {'class': 'persub'}):
a_title = persub.h1.text
a_link = ZimukuDownloader.site_url + \
persub.h1.a.attrs['href']
a_title = ZimukuDownloader.choice_prefix + a_title
sub_dict[a_title] = {'type': 'shooter', 'link': a_link}
else:
raise ValueError('Zimuku搜索结果出现未知结构页面')
if len(sub_dict) >= sub_num:
del keywords[:]
break
if len(keywords) > 1:
keyword = keyword.replace(keywords[-1], '').strip()
keywords.pop(-1)
continue
break
for sub_name, sub_info in sub_dict.items():
if sub_info['type'] == 'default':
# 综合搜索字幕页面
r = s.get(sub_info['link'], timeout=60)
bs_obj = BeautifulSoup(r.text, 'html.parser')
lang_box = bs_obj.find('ul', {'class': 'subinfo'}).find('li')
type_score = 0
for lang in lang_box.find_all('img'):
if 'uk' in lang.attrs['src']:
type_score += 1
elif 'hongkong' in lang.attrs['src']:
type_score += 2
elif 'china' in lang.attrs['src']:
type_score += 4
elif 'jollyroger' in lang.attrs['src']:
type_score += 8
sub_info['lan'] = type_score
download_link = bs_obj.find('a', {'id': 'down1'}).attrs['href']
download_link = urljoin(
ZimukuDownloader.site_url, download_link)
r = s.get(download_link, timeout=60)
bs_obj = BeautifulSoup(r.text, 'html.parser')
download_link = bs_obj.find('a', {'rel': 'nofollow'})
download_link = download_link.attrs['href']
download_link = urljoin(
ZimukuDownloader.site_url, download_link)
sub_info['link'] = download_link
else:
# 射手字幕页面
r = s.get(sub_info['link'], timeout=60)
bs_obj = BeautifulSoup(r.text, 'html.parser')
lang_box = bs_obj.find('ul', {'class': 'subinfo'}).find('li')
type_score = 0
text = lang_box.text
if '英' in text:
type_score += 1
elif '繁' in text:
type_score += 2
elif '简' in text:
type_score += 4
elif '双语' in text:
type_score += 8
sub_info['lan'] = type_score
download_link = bs_obj.find('a', {'id': 'down1'}).attrs['href']
sub_info['link'] = download_link
backup_session = requests.session()
backup_session.headers.update(s.headers)
backup_session.headers['Referer'] = sub_info['link']
backup_session.cookies.update(s.cookies)
sub_info['session'] = backup_session
if (len(sub_dict.items()) > 0
and list(sub_dict.items())[0][1]['lan'] < 8):
# 第一个候选字幕没有双语
sub_dict = order_dict(
sorted(sub_dict.items(),
key=lambda e: e[1]['lan'], reverse=True)
)
keys = list(sub_dict.keys())[:sub_num]
return {key: sub_dict[key] for key in keys}
def get_subtitles(self, video_name, sub_num=10):
print(PREFIX + " Searching ZIMUKU...", end="\r")
keywords, info_dict = Downloader.get_keywords(video_name)
keyword = " ".join(keywords)
info = guessit(keyword)
keywords.pop(0)
keywords.insert(0, info["title"])
if info.get("season"):
season = str(info["season"]).zfill(2)
keywords.insert(1, "s" + season)
sub_dict = order_dict()
s = requests.session()
s.headers.update(Downloader.header)
while True:
# 当前关键字搜索
r = s.get(ZimukuDownloader.search_url + keyword, timeout=10)
html = r.text
if "搜索不到相关字幕" not in html:
bs_obj = BeautifulSoup(r.text, "html.parser")
if bs_obj.find("div", {"class": "item"}):
# 综合搜索页面
for item in bs_obj.find_all("div", {"class": "item"}):
title_boxes = item.find("div", {"class": "title"}).find_all("p")
title_box = title_boxes[0]
sub_title_box = title_boxes[1]
item_title = title_box.text
item_sub_title = sub_title_box.text
item_info = guessit(item_title)
if info.get("year") and item_info.get("year"):
if info["year"] != item_info["year"]:
# 年份不匹配，跳过
continue
item_titles = [
item_info.get("title", "").lower(),
item_info.get("alternative_title", "").lower(),
] + item_sub_title.lower().strip().split(",")
title_included = sum(
[
1
for _ in item_sub_title
if info["title"].lower() not in _
]
)
if title_included == 0:
# guessit抽取标题不匹配，跳过
item_title_split = [one.split() for one in item_titles]
info_title_split = info["title"].lower().split()
sum1 = sum(
[
1
for _ in info_title_split
if _ in item_title_split[0]
]
)
sum2 = sum(
[
1
for _ in info_title_split
if _ in item_title_split[1]
]
)
if not (
sum1 / len(info_title_split) >= 0.5
or sum2 / len(info_title_split) >= 0.5
):
# 标题不匹配，跳过
continue
for a in item.find_all("td", {"class": "first"})[:3]:
a = a.a
a_link = ZimukuDownloader.site_url + a.attrs["href"]
a_title = a.text
a_title = ZimukuDownloader.choice_prefix + a_title
sub_dict[a_title] = {"type": "default", "link": a_link}
elif bs_obj.find("div", {"class": "persub"}):
# 射手字幕页面
for persub in bs_obj.find_all("div", {"class": "persub"}):
a_title = persub.h1.text
a_link = ZimukuDownloader.site_url + persub.h1.a.attrs["href"]
a_title = ZimukuDownloader.choice_prefix + a_title
sub_dict[a_title] = {"type": "shooter", "link": a_link}
else:
raise ValueError("Zimuku搜索结果出现未知结构页面")
if len(sub_dict) >= sub_num:
del keywords[:]
break
if len(keywords) > 1:
keyword = keyword.replace(keywords[-1], "").strip()
keywords.pop(-1)
continue
break
for sub_name, sub_info in sub_dict.items():
if sub_info["type"] == "default":
# 综合搜索字幕页面
r = s.get(sub_info["link"], timeout=60)
bs_obj = BeautifulSoup(r.text, "html.parser")
lang_box = bs_obj.find("ul", {"class": "subinfo"}).find("li")
type_score = 0
for lang in lang_box.find_all("img"):
if "uk" in lang.attrs["src"]:
type_score += 1
elif "hongkong" in lang.attrs["src"]:
type_score += 2
elif "china" in lang.attrs["src"]:
type_score += 4
elif "jollyroger" in lang.attrs["src"]:
type_score += 8
sub_info["lan"] = type_score
download_link = bs_obj.find("a", {"id": "down1"}).attrs["href"]
download_link = urljoin(ZimukuDownloader.site_url, download_link)
r = s.get(download_link, timeout=60)
bs_obj = BeautifulSoup(r.text, "html.parser")
download_link = bs_obj.find("a", {"rel": "nofollow"})
download_link = download_link.attrs["href"]
download_link = urljoin(ZimukuDownloader.site_url, download_link)
sub_info["link"] = download_link
else:
# 射手字幕页面
r = s.get(sub_info["link"], timeout=60)
bs_obj = BeautifulSoup(r.text, "html.parser")
lang_box = bs_obj.find("ul", {"class": "subinfo"}).find("li")
type_score = 0
text = lang_box.text
if "英" in text:
type_score += 1
elif "繁" in text:
type_score += 2
elif "简" in text:
type_score += 4
elif "双语" in text:
type_score += 8
sub_info["lan"] = type_score
download_link = bs_obj.find("a", {"id": "down1"}).attrs["href"]
sub_info["link"] = download_link
backup_session = requests.session()
backup_session.headers.update(s.headers)
backup_session.headers["Referer"] = sub_info["link"]
backup_session.cookies.update(s.cookies)
sub_info["session"] = backup_session
if len(sub_dict.items()) > 0 and list(sub_dict.items())[0][1]["lan"] < 8:
# 第一个候选字幕没有双语
sub_dict = order_dict(
sorted(sub_dict.items(), key=lambda e: e[1]["lan"], reverse=True)
)
keys = list(sub_dict.keys())[:sub_num]
return {key: sub_dict[key] for key in keys}