code.delx.au - webdl/blob - common.py

   1 import hashlib
   2 import io
   3 import json
   4 import logging
   5 import lxml.etree
   6 import lxml.html
   7 import os
   8 import re
   9 import requests
  10 import requests_cache
  11 import shutil
  12 import signal
  13 import subprocess
  14 import sys
  15 import time
  16 import urllib.parse
  17
  18 USER_AGENT = "Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/112.0"
  19
  20 try:
  21     import autosocks
  22     autosocks.try_autosocks()
  23 except ImportError:
  24     pass
  25
  26
  27 logging.basicConfig(
  28     format = "%(levelname)s %(message)s",
  29     level = logging.INFO if os.environ.get("DEBUG", None) is None else logging.DEBUG,
  30     stream = sys.stdout,
  31 )
  32
  33 CACHE_FILE = os.path.join(
  34     os.environ.get("XDG_CACHE_HOME", os.path.expanduser("~/.cache")),
  35     "webdl",
  36     "requests_cache"
  37 )
  38 if not os.path.isdir(os.path.dirname(CACHE_FILE)):
  39     os.makedirs(os.path.dirname(CACHE_FILE))
  40
  41 requests_cache.install_cache(CACHE_FILE, backend='sqlite', expire_after=3600)
  42
  43
  44 class Node(object):
  45     def __init__(self, title, parent=None):
  46         self.title = title
  47         if parent:
  48             parent.children.append(self)
  49         self.parent = parent
  50         self.children = []
  51         self.can_download = False
  52
  53     def get_children(self):
  54         if not self.children:
  55             self.fill_children()
  56             self.children = natural_sort(self.children, key=lambda node: node.title)
  57         return self.children
  58
  59     def fill_children(self):
  60         pass
  61
  62     def download(self):
  63         raise NotImplemented
  64
  65
  66 def load_root_node():
  67     root_node = Node("Root")
  68
  69     import iview
  70     iview.fill_nodes(root_node)
  71
  72     import sbs
  73     sbs.fill_nodes(root_node)
  74
  75     import ten
  76     ten.fill_nodes(root_node)
  77
  78     return root_node
  79
  80 valid_chars = frozenset("-_.()!@#%^ abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789")
  81 def sanify_filename(filename):
  82     filename = "".join(c for c in filename if c in valid_chars)
  83     assert len(filename) > 0
  84     return filename
  85
  86 def ensure_scheme(url):
  87     parts = urllib.parse.urlparse(url)
  88     if parts.scheme:
  89         return url
  90     parts = list(parts)
  91     parts[0] = "http"
  92     return urllib.parse.urlunparse(parts)
  93
  94 http_session = requests.Session()
  95 http_session.headers["User-Agent"] = USER_AGENT
  96
  97 def grab_text(url):
  98     logging.debug("grab_text(%r)", url)
  99     request = http_session.prepare_request(requests.Request("GET", url))
 100     response = http_session.send(request)
 101     return response.text
 102
 103 def grab_html(url):
 104     logging.debug("grab_html(%r)", url)
 105     request = http_session.prepare_request(requests.Request("GET", url))
 106     response = http_session.send(request)
 107     doc = lxml.html.parse(io.BytesIO(response.content), lxml.html.HTMLParser(encoding="utf-8", recover=True))
 108     response.close()
 109     return doc
 110
 111 def grab_xml(url):
 112     logging.debug("grab_xml(%r)", url)
 113     request = http_session.prepare_request(requests.Request("GET", url))
 114     response = http_session.send(request)
 115     doc = lxml.etree.parse(io.BytesIO(response.content), lxml.etree.XMLParser(encoding="utf-8", recover=True))
 116     response.close()
 117     return doc
 118
 119 def grab_json(url):
 120     logging.debug("grab_json(%r)", url)
 121     request = http_session.prepare_request(requests.Request("GET", url))
 122     response = http_session.send(request)
 123     return response.json()
 124
 125 def exec_subprocess(cmd):
 126     logging.debug("Executing: %s", cmd)
 127     try:
 128         p = subprocess.Popen(cmd)
 129         ret = p.wait()
 130         if ret != 0:
 131             logging.error("%s exited with error code: %s", cmd[0], ret)
 132             return False
 133         else:
 134             return True
 135     except OSError as e:
 136         logging.error("Failed to run: %s -- %s", cmd[0], e)
 137     except KeyboardInterrupt:
 138         logging.info("Cancelled: %s", cmd)
 139         try:
 140             p.terminate()
 141             p.wait()
 142         except KeyboardInterrupt:
 143             p.send_signal(signal.SIGKILL)
 144             p.wait()
 145     return False
 146
 147
 148 def check_command_exists(cmd):
 149     try:
 150         subprocess.check_output(cmd, stderr=subprocess.STDOUT)
 151         return True
 152     except Exception:
 153         return False
 154
 155 def get_duration(filename):
 156     cmd = [
 157         "ffprobe",
 158         filename,
 159         "-show_entries", "format=duration",
 160         "-v", "quiet",
 161     ]
 162     output = subprocess.check_output(cmd).decode("utf-8")
 163     for line in output.split("\n"):
 164         m = re.search(R"([0-9]+)", line)
 165         if not m:
 166             continue
 167         duration = m.group(1)
 168         if duration.isdigit():
 169             return int(duration)
 170
 171
 172     logging.debug("Falling back to full decode to find duration: %s % filename")
 173
 174     cmd = [
 175         "ffmpeg",
 176         "-i", filename,
 177         "-vn",
 178         "-f", "null", "-",
 179     ]
 180     output = subprocess.check_output(cmd, stderr=subprocess.STDOUT).decode("utf-8")
 181     duration = None
 182     for line in re.split(R"[\r\n]", output):
 183         m = re.search(R"time=([0-9:]*)\.", line)
 184         if not m:
 185             continue
 186         [h, m, s] = m.group(1).split(":")
 187         # ffmpeg prints the duration as it reads the file, we want the last one
 188         duration = int(h) * 3600 + int(m) * 60 + int(s)
 189
 190     if duration:
 191         return duration
 192     else:
 193         raise Exception("Unable to determine video duration of " + filename)
 194
 195 def check_video_durations(flv_filename, mp4_filename):
 196     flv_duration = get_duration(flv_filename)
 197     mp4_duration = get_duration(mp4_filename)
 198
 199     if abs(flv_duration - mp4_duration) > 1:
 200         logging.error(
 201             "The duration of %s is suspicious, did the remux fail? Expected %s == %s",
 202             mp4_filename, flv_duration, mp4_duration
 203         )
 204         return False
 205
 206     return True
 207
 208 def remux(infile, outfile):
 209     logging.info("Converting %s to mp4", infile)
 210
 211     cmd = [
 212         "ffmpeg",
 213         "-i", infile,
 214         "-bsf:a", "aac_adtstoasc",
 215         "-acodec", "copy",
 216         "-vcodec", "copy",
 217         "-y",
 218         outfile,
 219     ]
 220     if not exec_subprocess(cmd):
 221         return False
 222
 223     if not check_video_durations(infile, outfile):
 224         return False
 225
 226     os.unlink(infile)
 227     return True
 228
 229 def convert_to_mp4(filename):
 230     with open(filename, "rb") as f:
 231         fourcc = f.read(4)
 232     basename, ext = os.path.splitext(filename)
 233
 234     if ext == ".mp4" and fourcc == b"FLV\x01":
 235         os.rename(filename, basename + ".flv")
 236         ext = ".flv"
 237         filename = basename + ext
 238
 239     if ext in (".flv", ".ts"):
 240         filename_mp4 = basename + ".mp4"
 241         return remux(filename, filename_mp4)
 242
 243     return ext == ".mp4"
 244
 245
 246 def download_hds(filename, video_url, pvswf=None):
 247     filename = sanify_filename(filename)
 248     logging.info("Downloading: %s", filename)
 249
 250     video_url = "hds://" + video_url
 251     if pvswf:
 252         param = "%s pvswf=%s" % (video_url, pvswf)
 253     else:
 254         param = video_url
 255
 256     cmd = [
 257         "streamlink",
 258         "--force",
 259         "--output", filename,
 260         param,
 261         "best",
 262     ]
 263     if exec_subprocess(cmd):
 264         return convert_to_mp4(filename)
 265     else:
 266         return False
 267
 268 def download_hls(filename, video_url):
 269     filename = sanify_filename(filename)
 270     video_url = "hlsvariant://" + video_url
 271     logging.info("Downloading: %s", filename)
 272
 273     cmd = [
 274         "streamlink",
 275         "--http-header", "User-Agent=" + USER_AGENT,
 276         "--force",
 277         "--output", filename,
 278         video_url,
 279         "best",
 280     ]
 281     if exec_subprocess(cmd):
 282         return convert_to_mp4(filename)
 283     else:
 284         return False
 285
 286 def download_mpd(filename, video_url):
 287     filename = sanify_filename(filename)
 288     video_url = "dash://" + video_url
 289     logging.info("Downloading: %s", filename)
 290
 291     cmd = [
 292         "streamlink",
 293         "--force",
 294         "--output", filename,
 295         video_url,
 296         "best",
 297     ]
 298     if exec_subprocess(cmd):
 299         return convert_to_mp4(filename)
 300     else:
 301         return False
 302
 303 def download_http(filename, video_url):
 304     filename = sanify_filename(filename)
 305     logging.info("Downloading: %s", filename)
 306
 307     cmd = [
 308         "curl",
 309         "--fail", "--retry", "3",
 310         "-o", filename,
 311         video_url,
 312     ]
 313     if exec_subprocess(cmd):
 314         return convert_to_mp4(filename)
 315     else:
 316         return False
 317
 318 def natural_sort(l, key=None):
 319     ignore_list = ["a", "the"]
 320     def key_func(k):
 321         if key is not None:
 322             k = key(k)
 323         k = k.lower()
 324         newk = []
 325         for c in re.split("([0-9]+)", k):
 326             c = c.strip()
 327             if c.isdigit():
 328                 newk.append(c.zfill(5))
 329             else:
 330                 for subc in c.split():
 331                     if subc not in ignore_list:
 332                         newk.append(subc)
 333         return newk
 334
 335     return sorted(l, key=key_func)
 336
 337 def append_to_qs(url, params):
 338     r = list(urllib.parse.urlsplit(url))
 339     qs = urllib.parse.parse_qs(r[3])
 340     for k, v in params.items():
 341         if v is not None:
 342             qs[k] = v
 343         elif k in qs:
 344             del qs[k]
 345     r[3] = urllib.parse.urlencode(sorted(qs.items()), True)
 346     url = urllib.parse.urlunsplit(r)
 347     return url
 348