Function to generate `.info.json` from `ffmpeg` details.

create_info_json

import os
import ffmpeg
import datetime
import time
import json

def create_info_json(video, file_dest, reason="private"):
    print("Attempting to generate .info.json details from video metadata.")
    info_ext = "info.json"
    info_file = f"{os.path.splitext(os.path.basename(video))[0]}.{info_ext}"
    try:
      vid_ff = ffmpeg.probe(f"{video}")
    except ffmpeg.Error as e:
      print(e.stderr.decode(), file=sys.stderr)
      sys.exit(1)
    primary_stream = {}
    for stream in vid_ff["streams"]:
      if stream["codec_type"] not in primary_stream:
        primary_stream[stream["codec_type"]] = int(stream["index"])
      if primary_stream[stream["codec_type"]] > int(stream["index"]):
        primary_stream[stream["codec_type"]] = int(stream["index"])
    info_json = {}
    info_json["id"] = vid_ff["format"]["tags"]["comment"].split("?v=")[1]
    info_json["title"] = vid_ff["format"]["tags"]["title"]
    info_json["formats"] = []
    vid_format = {}
    vid_format["format_id"] = vid_ff["format"]["tags"]["major_brand"]
    vid_format["format_note"] = f"{vid_ff['format']['tags']['major_brand']}."
    vid_format["format_note"] += f"{vid_ff['format']['tags']['minor_version']}"
    vid_format["ext"] = os.path.splitext(video)[-1]
    vid_format["protocol"] = vid_ff["format"]["tags"]["compatible_brands"]
    vid_format["acodec"] = vid_ff["streams"][primary_stream["audio"]][
        "codec_name"
    ]
    vid_format["vcodec"] = vid_ff["streams"][primary_stream["video"]][
        "codec_name"
    ]
    vid_format["url"] = vid_ff["format"]["tags"]["comment"]
    vid_format["width"] = vid_ff["streams"][primary_stream["video"]][
        "coded_width"
    ]
    vid_format["height"] = vid_ff["streams"][primary_stream["video"]][
        "coded_height"
    ]
    vid_format["fps"] = 0
    # Fragments are not needed for private or non-existent videos
    vid_format["fragments"] = []
    vid_format["audio_ext"] = vid_ff["streams"][primary_stream["audio"]][
        "codec_long_name"
    ]
    vid_format["video_ext"] = vid_ff["streams"][primary_stream["video"]][
        "codec_long_name"
    ]
    vid_format["format"] = vid_ff["format"]["tags"]["compatible_brands"]
    vid_format["resolution"] = f"{vid_format['width']}x{vid_format['height']}"
    ratio1, ratio2 = vid_ff["streams"][0]["display_aspect_ratio"].split(":", 1)
    vid_format["aspect_ratio"] = round(float(ratio1) / float(ratio2), 2)
    vid_format["http_headers"] = {
        "User-Agent": "Chrome/0.0.0.1",
        "Accept": "text/html,application/xhtml+xml,application/xml",
        "Accept-Language": "en-us,en;q=0.5",
        "Sec-Fetch-Mode": "navigate",
    }
    info_json["formats"].append(vid_format)
    info_json["thumbnails"] = []
    info_json["thumbnail"] = ""
    info_json["description"] = vid_ff["format"]["tags"]["description"]
    info_json["uploader"] = os.getenv("SHOW_NAME")
    ch_id = ""
    try:
        with open(f"{file_dest}/channel.id") as file:
            for line in file.readlines():
                if len(line) > 0:
                    ch_id = line.strip()
    except BaseException as E:
        print(
            f"Unable to open channel.id file at the destination: {file_dest}",
            "\nNot including a channel ID.",
        )
        print(E)
    info_json["uploader_id"] = ch_id
    info_json[
        "uploader_url"
    ] = f"http://www.youtube.com/channel/{info_json['uploader_id']}"
    info_json["channel_id"] = ch_id
    info_json[
        "channel_url"
    ] = f"http://www.youtube.com/channel/{info_json['uploader_id']}"
    info_json["duration"] = int(round(float(vid_ff["format"]["duration"]), 0))
    info_json["view_count"] = 0
    info_json["age_limit"] = 0
    info_json["webpage_url"] = vid_ff["format"]["tags"]["comment"]
    info_json["categories"] = []
    info_json["categories"].append(reason)
    info_json["tags"] = []
    info_json["categories"].append(reason)
    info_json["playable_in_embed"] = True
    info_json["live_status"] = "not_live"
    info_json["automatic_captions"] = {}
    info_json["subtitles"] = {}
    info_json["comment_count"] = 0
    info_json["like_count"] = 0
    info_json["channel"] = os.getenv("SHOW_NAME")
    info_json["channel_follow_count"] = 0
    info_json["upload_date"] = vid_ff["format"]["tags"]["date"]
    info_json["availability"] = reason
    info_json["webpage_url_basename"] = "watch"
    info_json["webpage_url_domain"] = "youtube.com"
    info_json["extractor"] = "youtube"
    info_json["extractor_key"] = "Youtube"
    info_json["display_id"] = info_json["id"]
    info_json["fulltitle"] = vid_ff["format"]["tags"]["title"]
    info_json["duration_string"] = str(
        datetime.timedelta(seconds=info_json["duration"])
    )
    info_json["is_live"] = False
    info_json["was_live"] = False
    info_json["format_id"] = "0"
    info_json["ext"] = "mp4"
    info_json["protocol"] = "https"
    info_json["format_note"] = vid_format["format_note"]
    info_json["filesize_approx"] = vid_ff["format"]["size"]
    info_json["tbr"] = 0  # TBR Values are not necessary
    info_json["width"] = vid_format["width"]
    info_json["height"] = vid_format["height"]
    info_json["resolution"] = f"{info_json['width']}x{info_json['height']}"
    info_json["format"] = f"{info_json['format_id']} -"
    info_json["format"] += f" {info_json['format_note']}"
    info_json["format"] += f" ({info_json['resolution']})"
    info_json["fps"] = 24  # Assume default
    info_json["dynamic_range"] = "SDR"  # Assume default
    info_json["vcodec"] = vid_format["vcodec"]
    info_json["vbr"] = 0  # VBR Values are not necessary
    info_json["aspect_ratio"] = vid_format["aspect_ratio"]
    info_json["acodec"] = vid_format["acodec"]
    info_json["abr"] = 0  # ABR Values are not necessary
    info_json["asr"] = 0  # ASR Values are not necessary
    info_json["audio_channels"] = vid_ff["streams"][primary_stream["audio"]][
        "channels"
    ]
    info_json["epoch"] = int(time.time())
    info_json["_type"] = "video"
    info_json["_version"] = {}
    info_json["_version"]["version"] = "2023.01.01-dummy"
    info_json["_version"]["release_git_head"] = "0000000000000000000000000000000000000000"
    info_json["_version"]["repository"] = "yt-dlp/yt-dlp"
    print("Dictionary for `.info.json` file is filled. Writing to file.")
    with open(f"{file_dest}/{info_file}", "w") as file:
        json.dump(info_json, file)
    print(f"File {file_dest}/{info_file} successfully written.")
    return 0
    

#Run the function
create_info_json(filename,dir)

Here is an example of running it for each item in a chosen directory. This can be done at any level.

import re

rx = '[0-9]{8}_[a-zA-Z0-9_-]{11}_*.*'
dir = '/youtube'
for root, _, files in os.walk(dir):
  for filename in files:
    current_count += 1
    all_files.append(os.path.join(root,filename))
    match = re.search(rx, filename, re.IGNORECASE)
    if match:
      create_info_json(filename,root)

To get the file sizes inside of it, and combining it with the above:

import re

def conv_size(size):
        if size < 1024:
            return f"{size} bytes"
        elif size < pow(1024,2):
            return f"{round(size/1024, 2)} KB"
        elif size < pow(1024,3):
            return f"{round(size/(pow(1024,2)), 2)} MB"
        elif size < pow(1024,4):
            return f"{round(size/(pow(1024,3)), 2)} GB"
        else
            return f"{round(size/pow(1024,4)), 2} TB"

rx = '[0-9]{8}_[a-zA-Z0-9_-]{11}_*.*'
dir = '/youtube'
f_sizes = {}
current_count = 0
for root, folders, files in os.walk(dir):
  for filename in files:
    current_count += 1
    all_files.append(os.path.join(root,filename))
    match = re.search(rx, filename, re.IGNORECASE)
    if match:
      create_info_json(filename,root)
      print(f"{filename}: {conv_size(os.path.getsize(os.path.join(root, filename)))"})
      if not f_sizes.get(root):
        f_sizes[root] = 0
      f_sizes[root] += os.path.getsize(os.path.join(root,filename))
print("Folder sizes for moving items:")
for key in f_sizes.keys():
  print(f"{key}: {conv_size(f_sizes[key])}")