Xunlei subtitle downloader

subtitle_downloader.py

#!/usr/bin/python
# -*- coding: utf-8 -*-
# __author__ = 'ihciah'
# cid_hash_file function from https://github.com/binux/lixian.xunlei/blob/master/libs/tools.py

import hashlib
import inotify.adapters
import os
import requests
import re


class SubtitleDownloader:
    video_types = [u'mkv', u'mp4', u'avi', u'rm', u'rmvb', u'wmv', u'webm', u'mpeg', u'mpe', u'flv', u'3gp']
    subtitle_types = [u'ass', u'srt', u'sub', u'sst', u'son', u'ssa', u'smi', u'tts', u'psb', u'pjs', u'stl', u'vsf']
    download_count = 3
    watch_dir = b'/hdd/downloads'

    @staticmethod
    def download_srt(subtitle_url, video_base_path, video_name, num):
        dot = subtitle_url.rfind(u'.')
        if dot < 0:
            return
        subtitle_type = subtitle_url[dot + 1:].lower()
        if subtitle_type not in SubtitleDownloader.subtitle_types:
            return
        response = requests.get(subtitle_url)
        if response.status_code == 200:
            with open(os.path.join(video_base_path, video_name + u'.%d.' % num + subtitle_type), 'wb') as f:
                f.write(response.content)

    @staticmethod
    def cid_hash_file(path):
        h = hashlib.sha1()
        size = os.path.getsize(path)
        with open(path, 'rb') as stream:
            if size < 0xF000:
                h.update(stream.read())
            else:
                h.update(stream.read(0x5000))
                stream.seek(size//3)
                h.update(stream.read(0x5000))
                stream.seek(size-0x5000)
                h.update(stream.read(0x5000))
        return h.hexdigest().upper()

    @staticmethod
    def fetch_subtitle_list(cid):
        patten = re.compile(b'surl="(.*?)"')
        url_base = u'http://subtitle.kankan.xunlei.com:8000/submatch/%s/%s/%s.lua'
        r = requests.get(url_base % (cid[:2], cid[-2:], cid)).content
        srt_urls = patten.findall(r)[:SubtitleDownloader.download_count]
        return list(map(lambda url: url.decode(u'utf-8'), srt_urls))

    @staticmethod
    def download_subtitle(video):
        if not video or not os.path.isfile(video):
            print(u"Video file does not exist.")
            return -1
        video_base_path, video_filename = os.path.split(os.path.abspath(video))
        dot = video_filename.rfind(u'.')
        if dot < 0:
            print(u"Not a video file.")
            return -1
        video_name = video_filename[:dot]
        video_type = video_filename[dot+1:]
        if video_type.lower() not in SubtitleDownloader.video_types:
            print(u"Not a video file.")
            return -2
        cid = SubtitleDownloader.cid_hash_file(video)
        subtitle_list = SubtitleDownloader.fetch_subtitle_list(cid)
        for num, subtitle in enumerate(subtitle_list):
            SubtitleDownloader.download_srt(subtitle, video_base_path, video_name, num)

    @staticmethod
    def inotify_loop():
        i = inotify.adapters.InotifyTree(SubtitleDownloader.watch_dir)
        try:
            for event in i.event_gen():
                if event is not None:
                    try:
                        (header, type_names, watch_path, filename) = event
                        filename = filename.decode('utf-8')
                        watch_path = watch_path.decode('utf-8')
                        if 'IN_DELETE' in type_names and filename.endswith(u'.aria2'):
                            video_filename = filename[:filename.rfind(u'.')]
                            SubtitleDownloader.download_subtitle(os.path.join(watch_path, video_filename))
                    except:
                        pass
        finally:
            i.remove_watch(SubtitleDownloader.watch_dir)

if __name__ == '__main__':
    SubtitleDownloader.inotify_loop()