fetchYoutubeSubtitle.py

import json
import math
import time
from typing import Optional
import xml.etree.ElementTree as ElementTree
from html import unescape
import yt_dlp

# yt-dlp subtitle types: json3,srv1,srv2,srv3,ttml,vtt, xml(youtube url with out extargs)

# "subtitles": {
#     "live_chat": [
#       {
#         "url": "https://www.youtube.com/watch?v=ANtM2bHRz04&bpctr=9999999999&has_verified=1",
#         "ext": "json",
#         "video_id": "ANtM2bHRz04",
#         "protocol": "youtube_live_chat_replay"
#       }
#     ]
#   }
def getUrlFromSubtitles(item, lang='en', subType="vtt"):
    langs = item.keys()
    if len(langs) == 0:
        return None

    l = lang if lang in langs else ('en' if 'en' in langs else list(langs)[0] )
    if l is  None:
        return

    for subtitle in item[l]:
        # print("getUrlFromSubtitles subtitle: %s" % subtitle)
        if l != "live_chat" and subType =="xml":
            # print("subtitle source url: {}".format(subtitle.get("url")))
            return subtitle.get("url").replace("&fmt="+subtitle.get("ext"),"")
        if subtitle.get("ext") == subType:
            return subtitle.get("url")
    return None

async def fetchSubtitle(url: str, lang: Optional[str] = 'en', subType: Optional[str] = "vtt") -> dict:
    return await fetchSubtitlebyType(url, lang, subType)

async def fetchSubtitlebyType(url: str, lang: Optional[str] = 'en', subType="vtt") -> dict:
    ydl_opts = {
        "noplaylist": True,
        "writesubtitles": False,
        "allsubtitles": True,
        "subtitleslangs": [lang] if lang else [],
        "skip_download": True,
        "socket_timeout": 20
    }

    title = "unknow"
    duration = ""
    try:
        with yt_dlp.YoutubeDL(ydl_opts) as ydl:
            info_dict = ydl.extract_info(url, download=False)
            title = info_dict.get("title", "unknow")
            seconds = info_dict.get("duration")
            duration = str(seconds) if seconds else ""
            isSrt = False
            if info_dict.get("extractor") == "youtube" and subType == "srt":
                subType = "xml"
                isSrt = True

            for subtitle_item in ["subtitles", "automatic_captions"]: # "requested_subtitles" item is dict
                if info_dict.get(subtitle_item):
                    subtitle_url = getUrlFromSubtitles(info_dict.get(subtitle_item), lang, subType)
                    if subtitle_url:
                        # print("subtitle_url: {}".format(subtitle_url))
                        with ydl.urlopen(subtitle_url) as response:
                            subtitle = xml_caption_to_srt(response.read().decode()) if isSrt else response.read().decode()
                            print("url{}, title:{}, duration:{} len(subtitle): {}".format(url, title, duration, len(subtitle)))
                            return {"title": title, "duration": duration,"subtitle": subtitle}
    except Exception as e:
        return {"error": str(e)}
    return {"title": title, "duration": duration, "error": "No subtitles"}

def float_to_srt_time_format(d: float) -> str:
    """Convert decimal durations into proper srt format.
    :rtype: str
    :returns:
        SubRip Subtitle (str) formatted time duration.
    float_to_srt_time_format(3.89) -> '00:00:03,890'
    """
    fraction, whole = math.modf(d)
    time_fmt = time.strftime("%H:%M:%S,", time.gmtime(whole))
    ms = f"{fraction:.3f}".replace("0.", "")
    return time_fmt + ms

def xml_caption_to_srt( xml_captions: str) -> str:
    """Convert xml caption tracks to "SubRip Subtitle (srt)".
    :param str xml_captions:
        XML formatted caption tracks.
    """
    segments = []
    root = ElementTree.fromstring(xml_captions)
    for i, child in enumerate(list(root)):
        text = child.text or ""
        caption = unescape(text.replace("\n", " ").replace("  ", " "),)
        try:
            duration = float(child.attrib["dur"])
        except KeyError:
            duration = 0.0
        start = float(child.attrib["start"])
        end = start + duration
        sequence_number = i + 1  # convert from 0-indexed to 1.
        line = "{seq}\n{start} --> {end}\n{text}\n".format(
            seq=sequence_number,
            start=float_to_srt_time_format(start),
            end=float_to_srt_time_format(end),
            text=caption,
        )
        segments.append(line)
    return "\n".join(segments).strip()

async def fetchSubtitleUrls(url: str) -> json:
    ydl_opts = {
        "noplaylist": True,
        "writesubtitles": False,
        "allsubtitles": True,
        "skip_download": True,
    }
    title = "unknow"
    duration = ""
    try:
        with yt_dlp.YoutubeDL(ydl_opts) as ydl:
            info_dict = ydl.extract_info(url, download=False)
            title = info_dict.get("title", "unknow")
            seconds = info_dict.get("duration")
            duration = str(seconds) if seconds else ""
            if info_dict.get("subtitles"):
                langs = info_dict.get("subtitles").keys()
                if not (len(langs) == 1 and "live_chat" in langs):
                    return {"title": title, "duration": duration, "subtitles": info_dict.get("subtitles")}
                if info_dict.get("automatic_captions"):
                    return {"title": title, "duration": duration, "subtitles": info_dict.get("automatic_captions")}
    except Exception as e:
        return {"error": str(e)}
    return {"title": title, "duration": duration, "error": "No subtitles"}