1
0
mirror of https://github.com/spaam/svtplay-dl.git synced 2024-11-24 04:05:39 +01:00
svtplay-dl/lib/svtplay_dl/postprocess/__init__.py

232 lines
9.2 KiB
Python
Raw Normal View History

2019-08-25 00:40:39 +02:00
import logging
import os
import pathlib
import platform
import re
import sys
2019-08-25 00:40:39 +02:00
from random import sample
from shutil import which
2019-08-25 00:40:39 +02:00
from requests import codes
from requests import post
from requests import Timeout
2023-04-19 03:27:59 +02:00
from svtplay_dl.utils.http import FIREFOX_UA
2018-05-13 01:44:21 +02:00
from svtplay_dl.utils.output import formatname
from svtplay_dl.utils.proc import run_program
from svtplay_dl.utils.stream import subtitle_filter
2019-08-25 00:33:51 +02:00
class postprocess:
2018-05-13 01:44:00 +02:00
def __init__(self, stream, config, subfixes=None):
self.stream = stream
2018-05-13 01:44:00 +02:00
self.config = config
self.subfixes = [x.subfix for x in subtitle_filter(subfixes)]
self.detect = None
for i in ["ffmpeg", "avconv"]:
self.detect = which(i)
if self.detect:
break
if self.detect is None and platform.system() == "Windows":
path = pathlib.Path(sys.executable).parent / "ffmpeg.exe"
if os.path.isfile(path):
self.detect = path
def merge(self, merge_subtitle):
if self.detect is None:
2018-11-18 12:47:19 +01:00
logging.error("Cant detect ffmpeg or avconv. Cant mux files without it.")
2016-03-26 21:38:31 +01:00
return
if self.stream.finished is False:
return
orig_filename = formatname(self.stream.output, self.config)
ext = orig_filename.suffix
2021-06-25 09:52:25 +02:00
new_name = orig_filename.with_suffix(f".{self.config.get('output_format')}")
if ext == ".ts":
if self.stream.audio:
if not str(orig_filename).endswith(".audio.ts"):
audio_filename = orig_filename.with_suffix(".audio.ts")
else:
audio_filename = orig_filename
else:
audio_filename = orig_filename.with_suffix(".m4a")
cmd = [self.detect]
if (self.config.get("only_video") or not self.config.get("only_audio")) or (not self.stream.audio and self.config.get("only_audio")):
cmd += ["-i", str(orig_filename)]
if self.stream.audio:
cmd += ["-i", str(audio_filename)]
2021-12-18 21:37:09 +01:00
_, _, stderr = run_program(cmd, False) # return 1 is good here.
streams = _streams(stderr)
videotrack, audiotrack = _checktracks(streams)
if merge_subtitle:
2021-12-18 21:36:16 +01:00
logging.info("Merge audio, video and subtitle into %s", new_name.name)
else:
logging.info(f"Merge audio and video into {str(new_name.name).replace('.audio', '')}")
tempfile = orig_filename.with_suffix(".temp")
arguments = []
if self.config.get("only_audio"):
arguments += ["-vn"]
if self.config.get("only_video"):
arguments += ["-an"]
arguments += ["-c:v", "copy", "-c:a", "copy", "-f", "matroska" if self.config.get("output_format") == "mkv" else "mp4"]
if ext == ".ts":
if audiotrack and "aac" in _getcodec(streams, audiotrack):
arguments += ["-bsf:a", "aac_adtstoasc"]
if videotrack and "dvh1" in _getcodec(streams, videotrack):
if self.config.get("output_format") == "mkv":
logging.warning("HDR and mkv is not supported.")
arguments += ["-strict", "unofficial"]
cmd = [self.detect]
if self.config.get("only_video") or (not self.config.get("only_audio") or (not self.stream.audio and self.config.get("only_audio"))):
cmd += ["-i", str(orig_filename)]
if self.stream.audio:
cmd += ["-i", str(audio_filename)]
if videotrack:
2021-02-28 22:05:15 +01:00
arguments += ["-map", f"{videotrack}"]
if audiotrack:
2021-02-28 22:05:15 +01:00
arguments += ["-map", f"{audiotrack}"]
if merge_subtitle:
langs = _sublanguage(self.stream, self.config, self.subfixes)
tracks = [x for x in [videotrack, audiotrack] if x]
subs_nr = 0
sub_start = 0
# find what sub track to start with. when a/v is in one file it start with 1
# if seperate it will start with 2
for i in tracks:
if int(i[0]) >= sub_start:
sub_start += 1
for stream_num, language in enumerate(langs, start=sub_start):
2019-08-25 00:27:31 +02:00
arguments += [
"-map",
f"{str(stream_num)}:0",
"-c:s:" + str(subs_nr),
2021-06-25 09:52:25 +02:00
"mov_text" if self.config.get("output_format") == "mp4" else "copy",
"-metadata:s:s:" + str(subs_nr),
2019-08-25 00:27:31 +02:00
"language=" + language,
]
subs_nr += 1
if self.subfixes and self.config.get("get_all_subtitles"):
for subfix in self.subfixes:
subfile = orig_filename.parent / (orig_filename.stem + "." + subfix + ".srt")
cmd += ["-i", str(subfile)]
else:
subfile = orig_filename.with_suffix(".srt")
cmd += ["-i", str(subfile)]
2017-02-15 23:15:50 +01:00
arguments += ["-y", str(tempfile)]
2016-03-26 21:38:31 +01:00
cmd += arguments
returncode, stdout, stderr = run_program(cmd)
if returncode != 0:
2016-03-26 21:38:31 +01:00
return
if self.config.get("keep_original") is True:
logging.info("Merging done, keeping original files.")
os.rename(tempfile, orig_filename)
return
2018-11-18 12:47:19 +01:00
logging.info("Merging done, removing old files.")
if self.config.get("only_video") or (not self.config.get("only_audio") or (not self.stream.audio and self.config.get("only_audio"))):
os.remove(orig_filename)
if (self.stream.audio and self.config.get("only_audio")) or (self.stream.audio and not self.config.get("only_video")):
os.remove(audio_filename)
# This if statement is for use cases where both -S and -M are specified to not only merge the subtitle but also store it separately.
if merge_subtitle and not self.config.get("subtitle"):
if self.subfixes and len(self.subfixes) >= 2 and self.config.get("get_all_subtitles"):
for subfix in self.subfixes:
subfile = orig_filename.parent / (orig_filename.stem + "." + subfix + ".srt")
os.remove(subfile)
2018-01-30 20:11:37 +01:00
else:
os.remove(subfile)
os.rename(tempfile, str(new_name).replace(".audio", ""))
def _streams(output):
2023-10-09 00:07:12 +02:00
return re.findall(r"Stream \#(\d:\d)([\[\(][^:\[]+[\]\)])?([\(\)\w]+)?: (Video|Audio): (.*)", output)
def _getcodec(streams, number):
for stream in streams:
if stream[0] == number:
return stream[4]
return None
def _checktracks(streams):
videotrack = None
audiotrack = None
for stream in streams:
if stream[3] == "Video":
videotrack = stream[0]
if stream[3] == "Audio":
if stream[4] == "mp3, 0 channels":
continue
audiotrack = stream[0]
return videotrack, audiotrack
def _sublanguage(stream, config, subfixes):
# parse() function partly borrowed from a guy on github. /thanks!
# https://github.com/riobard/srt.py/blob/master/srt.py
def parse(self):
def parse_block(block):
lines = block.strip("-").split("\n")
txt = "\r\n".join(lines[2:])
return txt
if platform.system() == "Windows":
fd = open(self, encoding="utf8")
else:
fd = open(self)
return list(map(parse_block, fd.read().strip().replace("\r", "").split("\n\n")))
def query(self):
_ = parse(self)
random_sentences = " ".join(sample(_, len(_) if len(_) < 8 else 8)).replace("\r\n", "")
2023-04-19 03:27:59 +02:00
url = "https://svtplay-dl.se/langdetect/"
2023-11-26 22:07:23 +01:00
bits = "64" if sys.maxsize > 2**32 else "32"
headers = {"User-Agent": f"{FIREFOX_UA} {platform.machine()} {platform.platform()} {bits}"}
try:
2023-04-19 03:27:59 +02:00
r = post(url, json={"query": random_sentences}, headers=headers, timeout=30)
if r.status_code == codes.ok:
try:
response = r.json()
return response["language"]
except TypeError:
return "und"
else:
logging.error("Server error appeared. Setting language as undetermined.")
return "und"
except Timeout:
logging.error("30 seconds server timeout reached. Setting language as undetermined.")
return "und"
langs = []
exceptions = {"lulesamiska": "smj", "meankieli": "fit", "jiddisch": "yid"}
logging.info("Determining the language of the subtitle(s).")
if config.get("get_all_subtitles"):
for subfix in subfixes:
if [exceptions[key] for key in exceptions.keys() if re.match(key, subfix.strip("-"))]:
if "oversattning" in subfix.strip("-"):
subfix = subfix.strip("-").split(".")[0]
else:
subfix = subfix.strip("-")
langs += [exceptions[subfix]]
continue
sfile = formatname(stream.output, config)
subfile = sfile.parent / (sfile.stem + "." + subfix + ".srt")
langs += [query(subfile)]
else:
subfile = formatname(stream.output, config).with_suffix(".srt")
langs += [query(subfile)]
if len(langs) >= 2:
2021-12-18 21:37:09 +01:00
logging.info("Language codes: %s", ", ".join(langs))
else:
2021-12-18 21:37:09 +01:00
logging.info("Language code: %s", langs[0])
return langs