summaryrefslogtreecommitdiffhomepage
diff options
context:
space:
mode:
authormorpheus65535 <[email protected]>2023-04-21 08:24:47 -0400
committermorpheus65535 <[email protected]>2023-04-21 08:24:47 -0400
commit039fa5e1ad278b58546a51d246622f15ce35fe8e (patch)
tree828d079c9cd82e37cce009628fa428a0e8ffcbdb
parentb9dcadc06fa2a6f5dd95f05955c55da558c925e3 (diff)
parente2ba532cee04b6fbf1e536d18666dc4d017f23ee (diff)
downloadbazarr-1.2.1-beta.18.tar.gz
bazarr-1.2.1-beta.18.zip
Merge remote-tracking branch 'origin/development' into developmentv1.2.1-beta.18
-rw-r--r--bazarr/app/config.py2
-rw-r--r--bazarr/init.py2
-rw-r--r--bazarr/subtitles/upgrade.py3
-rw-r--r--frontend/src/pages/Settings/Providers/list.ts6
-rw-r--r--libs/subliminal_patch/providers/zimuku.py101
-rw-r--r--requirements.txt3
6 files changed, 74 insertions, 43 deletions
diff --git a/bazarr/app/config.py b/bazarr/app/config.py
index acf5e9104..306c46539 100644
--- a/bazarr/app/config.py
+++ b/bazarr/app/config.py
@@ -280,7 +280,7 @@ defaults = {
'database': '',
'username': '',
'password': '',
- }
+ },
}
settings = SimpleConfigParser(defaults=defaults, interpolation=None)
diff --git a/bazarr/init.py b/bazarr/init.py
index 3c4f5a621..f0e06416f 100644
--- a/bazarr/init.py
+++ b/bazarr/init.py
@@ -76,7 +76,7 @@ def is_virtualenv():
# deploy requirements.txt
if not args.no_update:
try:
- import lxml, numpy, webrtcvad, setuptools # noqa E401
+ import lxml, numpy, webrtcvad, setuptools, PIL # noqa E401
except ImportError:
try:
import pip # noqa W0611
diff --git a/bazarr/subtitles/upgrade.py b/bazarr/subtitles/upgrade.py
index e292f1621..746d4ccde 100644
--- a/bazarr/subtitles/upgrade.py
+++ b/bazarr/subtitles/upgrade.py
@@ -138,6 +138,9 @@ def parse_upgradable_list(upgradable_list, perfect_score, media_type):
for item in upgradable_list:
logging.debug(f"Trying to validate eligibility to upgrade for this subtitles: "
f"{item['subtitles_path']}")
+ if not os.path.exists(path_replace_method(item['subtitles_path'])):
+ logging.debug("Subtitles file doesn't exists anymore, we skip this one.")
+ continue
if (item['video_path'], item['language']) in \
[(x['video_path'], x['language']) for x in items_to_upgrade]:
logging.debug("Newer video path and subtitles language combination already in list of subtitles to "
diff --git a/frontend/src/pages/Settings/Providers/list.ts b/frontend/src/pages/Settings/Providers/list.ts
index 780258764..c266f5203 100644
--- a/frontend/src/pages/Settings/Providers/list.ts
+++ b/frontend/src/pages/Settings/Providers/list.ts
@@ -448,5 +448,9 @@ export const ProviderList: Readonly<ProviderInfo[]> = [
description: "Bulgarian Subtitles Provider",
},
{ key: "yifysubtitles", name: "YIFY Subtitles" },
- { key: "zimuku", description: "Chinese Subtitles Provider" },
+ {
+ key: "zimuku",
+ name: "Zimuku",
+ description: "Chinese Subtitles Provider. Anti-captcha required",
+ },
];
diff --git a/libs/subliminal_patch/providers/zimuku.py b/libs/subliminal_patch/providers/zimuku.py
index 27251e657..1940f0f1c 100644
--- a/libs/subliminal_patch/providers/zimuku.py
+++ b/libs/subliminal_patch/providers/zimuku.py
@@ -1,11 +1,13 @@
# -*- coding: utf-8 -*-
from __future__ import absolute_import
+import base64
import io
import logging
import os
import zipfile
import re
import copy
+from PIL import Image
try:
from urlparse import urljoin
@@ -20,6 +22,7 @@ from requests import Session
from six import text_type
from random import randint
+from python_anticaptcha import AnticaptchaClient, ImageToTextTask
from subliminal.providers import ParserBeautifulSoup
from subliminal_patch.providers import Provider
from subliminal.subtitle import (
@@ -39,6 +42,7 @@ language_converters.register('zimuku = subliminal_patch.converters.zimuku:zimuku
supported_languages = list(language_converters['zimuku'].to_zimuku.keys())
+
class ZimukuSubtitle(Subtitle):
"""Zimuku Subtitle."""
@@ -80,6 +84,13 @@ class ZimukuSubtitle(Subtitle):
return matches
+def string_to_hex(s):
+ val = ""
+ for i in s:
+ val += hex(ord(i))[2:]
+ return val
+
+
class ZimukuProvider(Provider):
"""Zimuku Provider."""
@@ -87,40 +98,58 @@ class ZimukuProvider(Provider):
video_types = (Episode, Movie)
logger.info(str(supported_languages))
- server_url = "http://zimuku.org"
- search_url = "/search?q={}&security_verify_data={}"
- download_url = "http://zimuku.org/"
+ server_url = "https://so.zimuku.org"
+ search_url = "/search?q={}"
subtitle_class = ZimukuSubtitle
def __init__(self):
self.session = None
- def stringToHex(self, s):
- val = ""
- for i in s:
- val += hex(ord(i))[2:]
- return val
- vertoken = ""
+ verify_token = ""
+ code = ""
location_re = re.compile(
- r'self\.location = "(.*)" \+ stringToHex\(screendate\)')
+ r'self\.location = "(.*)" \+ stringToHex\(text\)')
+ verification_image_re = re.compile(r'<img.*?src="data:image/bmp;base64,(.*?)".*?>')
def yunsuo_bypass(self, url, *args, **kwargs):
+ def parse_verification_image(image_content: str):
+
+ def bmp_to_image(base64_str, img_type='png'):
+ img_data = base64.b64decode(base64_str)
+ img = Image.open(io.BytesIO(img_data))
+ img = img.convert("RGB")
+ img_fp = io.BytesIO()
+ img.save(img_fp, img_type)
+ img_fp.seek(0)
+ return img_fp
+
+ fp = bmp_to_image(image_content)
+ task = ImageToTextTask(fp)
+ client = AnticaptchaClient(os.environ.get('ANTICAPTCHA_ACCOUNT_KEY'))
+ job = client.createTask(task)
+ job.join()
+ return job.get_captcha_text()
+
i = -1
while True:
i += 1
r = self.session.get(url, *args, **kwargs)
- if(r.status_code == 404):
+ if r.status_code == 404:
+ # mock js script logic
tr = self.location_re.findall(r.text)
- self.session.cookies.set("srcurl", self.stringToHex(r.url))
- if(tr):
+ verification_image = self.verification_image_re.findall(r.text)
+ self.code = parse_verification_image(verification_image[0])
+ self.session.cookies.set("srcurl", string_to_hex(r.url))
+ if tr:
verify_resp = self.session.get(
- self.server_url+tr[0]+self.stringToHex("1920,1080"), allow_redirects=False)
- if(verify_resp.status_code == 302 and self.session.cookies.get("security_session_verify") != None):
+ self.server_url + tr[0] + string_to_hex(self.code), allow_redirects=False)
+ if verify_resp.status_code == 302 \
+ and self.session.cookies.get("security_session_verify") is not None:
pass
continue
if len(self.location_re.findall(r.text)) == 0:
- self.vertoken = self.stringToHex("1920,1080")
+ self.verify_token = string_to_hex(self.code)
return r
def initialize(self):
@@ -147,14 +176,14 @@ class ZimukuProvider(Provider):
language = Language("eng")
for img in sub.find("td", class_="tac lang").find_all("img"):
if (
- "china" in img.attrs["src"]
- and "hongkong" in img.attrs["src"]
+ "china" in img.attrs["src"]
+ and "hongkong" in img.attrs["src"]
):
language = Language("zho").add(Language('zho', 'TW', None))
- logger.debug("language:"+str(language))
+ logger.debug("language:" + str(language))
elif (
- "china" in img.attrs["src"]
- or "jollyroger" in img.attrs["src"]
+ "china" in img.attrs["src"]
+ or "jollyroger" in img.attrs["src"]
):
language = Language("zho")
elif "hongkong" in img.attrs["src"]:
@@ -171,8 +200,6 @@ class ZimukuProvider(Provider):
return subs
def query(self, keyword, season=None, episode=None, year=None):
- if self.vertoken == "":
- self.yunsuo_bypass(self.server_url + '/')
params = keyword
if season:
params += ".S{season:02d}".format(season=season)
@@ -181,8 +208,8 @@ class ZimukuProvider(Provider):
logger.debug("Searching subtitles %r", params)
subtitles = []
- search_link = self.server_url + text_type(self.search_url).format(params, self.vertoken)
-
+ search_link = self.server_url + text_type(self.search_url).format(params)
+
r = self.yunsuo_bypass(search_link, timeout=30)
r.raise_for_status()
@@ -198,7 +225,7 @@ class ZimukuProvider(Provider):
while parts:
parts.reverse()
redirect_url = urljoin(self.server_url, "".join(parts))
- r = self.query_resp(redirect_url, timeout=30)
+ r = self.session.get(redirect_url, timeout=30)
html = r.content.decode("utf-8", "ignore")
parts = re.findall(pattern, html)
logger.debug("search url located: " + redirect_url)
@@ -267,26 +294,22 @@ class ZimukuProvider(Provider):
return subtitles
def download_subtitle(self, subtitle):
- def _get_archive_dowload_link(yunsuopass, sub_page_link):
- r = yunsuopass(sub_page_link)
+ def _get_archive_download_link(yunsuopass, sub_page_link):
+ res = yunsuopass(sub_page_link)
bs_obj = ParserBeautifulSoup(
- r.content.decode("utf-8", "ignore"), ["html.parser"]
+ res.content.decode("utf-8", "ignore"), ["html.parser"]
)
down_page_link = bs_obj.find("a", {"id": "down1"}).attrs["href"]
down_page_link = urljoin(sub_page_link, down_page_link)
- r = yunsuopass(down_page_link)
+ res = yunsuopass(down_page_link)
bs_obj = ParserBeautifulSoup(
- r.content.decode("utf-8", "ignore"), ["html.parser"]
+ res.content.decode("utf-8", "ignore"), ["html.parser"]
)
- download_link = bs_obj.find("a", {"rel": "nofollow"})
- download_link = download_link.attrs["href"]
- download_link = urljoin(sub_page_link, download_link)
- return download_link
+ return urljoin(sub_page_link, bs_obj.find("a", {"rel": "nofollow"}).attrs["href"])
# download the subtitle
logger.info("Downloading subtitle %r", subtitle)
- self.session = subtitle.session
- download_link = _get_archive_dowload_link(self.yunsuo_bypass, subtitle.page_link)
+ download_link = _get_archive_download_link(self.yunsuo_bypass, subtitle.page_link)
r = self.yunsuo_bypass(download_link, headers={'Referer': subtitle.page_link}, timeout=30)
r.raise_for_status()
try:
@@ -404,7 +427,7 @@ def _extract_name(name):
result = [start, end]
start = end
end += 1
- new_name = name[result[0] : result[1]]
+ new_name = name[result[0]: result[1]]
new_name = new_name.strip() + suffix
return new_name
@@ -413,7 +436,7 @@ def num_to_cn(number):
""" convert numbers(1-99) to Chinese """
assert number.isdigit() and 1 <= int(number) <= 99
- trans_map = {n: c for n, c in zip(("123456789"), ("一二三四五六七八九"))}
+ trans_map = {n: c for n, c in zip("123456789", "一二三四五六七八九")}
if len(number) == 1:
return trans_map[number]
diff --git a/requirements.txt b/requirements.txt
index c061d866e..5489479b5 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,4 +1,5 @@
setuptools
lxml>=4.3.0
numpy>=1.12.0
-webrtcvad-wheels>=2.0.10 \ No newline at end of file
+webrtcvad-wheels>=2.0.10
+Pillow>=9.0.0 \ No newline at end of file