X-Git-Url: https://code.delx.au/webdl/blobdiff_plain/0a8b0318cb27682375b0f6526d0075b5ddfeed34..e28e65545c03e6452287c6e0cb6427eff49bd358:/common.py diff --git a/common.py b/common.py index ead72c6..f0f827b 100644 --- a/common.py +++ b/common.py @@ -1,19 +1,18 @@ -import python2_compat - import hashlib -import http.cookiejar +import io import json import logging import lxml.etree import lxml.html import os import re +import requests +import requests_cache import shutil import signal import subprocess import time import urllib.parse -import urllib.request try: @@ -28,12 +27,15 @@ logging.basicConfig( level = logging.INFO if os.environ.get("DEBUG", None) is None else logging.DEBUG, ) -CACHE_DIR = os.path.join( +CACHE_FILE = os.path.join( os.environ.get("XDG_CACHE_HOME", os.path.expanduser("~/.cache")), - "webdl" + "webdl", + "requests_cache" ) +if not os.path.isdir(os.path.dirname(CACHE_FILE)): + os.makedirs(os.path.dirname(CACHE_FILE)) -USER_AGENT = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:21.0) Gecko/20100101 Firefox/21.0" +requests_cache.install_cache(CACHE_FILE, backend='sqlite', expire_after=3600) class Node(object): @@ -66,8 +68,8 @@ def load_root_node(): import sbs sbs.fill_nodes(root_node) - import brightcove - brightcove.fill_nodes(root_node) + import ten + ten.fill_nodes(root_node) return root_node @@ -85,83 +87,36 @@ def ensure_scheme(url): parts[0] = "http" return urllib.parse.urlunparse(parts) -cookiejar = http.cookiejar.CookieJar() -urlopener = urllib.request.build_opener(urllib.request.HTTPCookieProcessor(cookiejar)) -def _urlopen(url, referrer=None): - url = ensure_scheme(url) - req = urllib.request.Request(url) - req.add_header("User-Agent", USER_AGENT) - if referrer: - req.add_header("Referer", referrer) - return urlopener.open(req) - -def urlopen(url, max_age): - logging.debug("urlopen(%r, %r)", url, max_age) - - if not os.path.isdir(CACHE_DIR): - os.makedirs(CACHE_DIR) - - if max_age <= 0: - return _urlopen(url) - - filename = hashlib.md5(url.encode("utf-8")).hexdigest() - filename = os.path.join(CACHE_DIR, filename) - if os.path.exists(filename): - file_age = int(time.time()) - os.path.getmtime(filename) - if file_age < max_age: - logging.debug("loading from cache: %s", filename) - return open(filename, "rb") - - logging.debug("downloading: %s -> %s", url, filename) - src = _urlopen(url) - dst = open(filename, "wb") - try: - shutil.copyfileobj(src, dst) - except Exception as e: - try: - os.unlink(filename) - except OSError: - pass - raise e - src.close() - dst.close() - - return open(filename, "rb") - -def grab_text(url, max_age): - f = urlopen(url, max_age) - text = f.read().decode("utf-8") - f.close() - return text - -def grab_html(url, max_age): - f = urlopen(url, max_age) - doc = lxml.html.parse(f, lxml.html.HTMLParser(encoding="utf-8", recover=True)) - f.close() +http_session = requests.Session() +http_session.headers["User-Agent"] = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:21.0) Gecko/20100101 Firefox/21.0" + +def grab_text(url): + logging.debug("grab_text(%r)", url) + request = http_session.prepare_request(requests.Request("GET", url)) + response = http_session.send(request) + return response.text + +def grab_html(url): + logging.debug("grab_html(%r)", url) + request = http_session.prepare_request(requests.Request("GET", url)) + response = http_session.send(request, stream=True) + doc = lxml.html.parse(io.BytesIO(response.content), lxml.html.HTMLParser(encoding="utf-8", recover=True)) + response.close() return doc -def grab_xml(url, max_age): - f = urlopen(url, max_age) - doc = lxml.etree.parse(f, lxml.etree.XMLParser(encoding="utf-8", recover=True)) - f.close() +def grab_xml(url): + logging.debug("grab_xml(%r)", url) + request = http_session.prepare_request(requests.Request("GET", url)) + response = http_session.send(request, stream=True) + doc = lxml.etree.parse(io.BytesIO(response.content), lxml.etree.XMLParser(encoding="utf-8", recover=True)) + response.close() return doc -def grab_json(url, max_age, skip_assignment=False, skip_function=False): - f = urlopen(url, max_age) - text = f.read().decode("utf-8") - - if skip_assignment: - pos = text.find("=") - text = text[pos+1:] - - elif skip_function: - pos = text.find("(") - rpos = text.rfind(")") - text = text[pos+1:rpos] - - doc = json.loads(text) - f.close() - return doc +def grab_json(url): + logging.debug("grab_json(%r)", url) + request = http_session.prepare_request(requests.Request("GET", url)) + response = http_session.send(request) + return response.json() def exec_subprocess(cmd): logging.debug("Executing: %s", cmd) @@ -207,6 +162,13 @@ def find_ffprobe(): raise Exception("You must install ffmpeg or libav-tools") +def find_streamlink(): + for streamlink in ["streamlink", "livestreamer"]: + if check_command_exists([streamlink, "--help"]): + return streamlink + + raise Exception("You must install streamlink or livestreamer") + def get_duration(filename): ffprobe = find_ffprobe() @@ -218,12 +180,37 @@ def get_duration(filename): ] output = subprocess.check_output(cmd).decode("utf-8") for line in output.split("\n"): - if line.startswith("duration="): - return float(line.split("=")[1]) # ffprobe - if re.match(R'^[0-9.]*$', line): - return float(line) # avprobe + m = re.search(R"([0-9]+)", line) + if not m: + continue + duration = m.group(1) + if duration.isdigit(): + return int(duration) + - raise Exception("Unable to determine video duration of " + filename) + logging.debug("Falling back to full decode to find duration: %s % filename") + + ffmpeg = find_ffmpeg() + cmd = [ + ffmpeg, + "-i", filename, + "-vn", + "-f", "null", "-", + ] + output = subprocess.check_output(cmd, stderr=subprocess.STDOUT).decode("utf-8") + duration = None + for line in re.split(R"[\r\n]", output): + m = re.search(R"time=([0-9:]*)\.", line) + if not m: + continue + [h, m, s] = m.group(1).split(":") + # ffmpeg prints the duration as it reads the file, we want the last one + duration = int(h) * 3600 + int(m) * 60 + int(s) + + if duration: + return duration + else: + raise Exception("Unable to determine video duration of " + filename) def check_video_durations(flv_filename, mp4_filename): flv_duration = get_duration(flv_filename) @@ -248,6 +235,7 @@ def remux(infile, outfile): "-bsf:a", "aac_adtstoasc", "-acodec", "copy", "-vcodec", "copy", + "-y", outfile, ] if not exec_subprocess(cmd): @@ -277,6 +265,8 @@ def convert_to_mp4(filename): def download_hds(filename, video_url, pvswf=None): + streamlink = find_streamlink() + filename = sanify_filename(filename) logging.info("Downloading: %s", filename) @@ -287,7 +277,8 @@ def download_hds(filename, video_url, pvswf=None): param = video_url cmd = [ - "livestreamer", + streamlink, + "-f", "-o", filename, param, "best", @@ -298,12 +289,34 @@ def download_hds(filename, video_url, pvswf=None): return False def download_hls(filename, video_url): + streamlink = find_streamlink() + filename = sanify_filename(filename) video_url = "hlsvariant://" + video_url logging.info("Downloading: %s", filename) cmd = [ - "livestreamer", + streamlink, + "-f", + "-o", filename, + video_url, + "best", + ] + if exec_subprocess(cmd): + return convert_to_mp4(filename) + else: + return False + +def download_mpd(filename, video_url): + streamlink = find_streamlink() + + filename = sanify_filename(filename) + video_url = "dash://" + video_url + logging.info("Downloading: %s", filename) + + cmd = [ + streamlink, + "-f", "-o", filename, video_url, "best",