scan-vid/scan-vid.py

#!/usr/local/bin/python

# выбирает информацию по видео файлу.

import os
import subprocess
import json
import zlib
import datetime

# правильный перевод!
def humanbytes(B):
    """Return the given bytes as a human friendly KB, MB, GB, or TB string."""
    B = float(B)
    KB = float(1024)
    MB = float(KB ** 2) # 1,048,576
    GB = float(KB ** 3) # 1,073,741,824
    TB = float(KB ** 4) # 1,099,511,627,776

    if B < KB:
        return '{0} {1}'.format(B,'Bytes' if 0 == B > 1 else 'Byte')
    elif KB <= B < MB:
        return '{0:.2f} KB'.format(B / KB)
    elif MB <= B < GB:
        return '{0:.2f} MB'.format(B / MB)
    elif GB <= B < TB:
        return '{0:.2f} GB'.format(B / GB)
    elif TB <= B:
        return '{0:.2f} TB'.format(B / TB)

# file_info
def file_info(fn):
	meta = {}
	with open(fn, 'rb') as fp:
		filename, file_extension = os.path.splitext(fn)
		created = os.path.getmtime(fn)
		dt_m = datetime.datetime.fromtimestamp(created)
		size = os.path.getsize(fn)
		data = fp.read()

		meta["name"] = fn
		meta["path"] = os.path.splitdrive(fn)[1]
		meta["ext"] = file_extension.lower()
		meta["date"] = dt_m
		meta["type"] = "meta"
		meta["crc"] = zlib.crc32(data)
		meta["size"] = size
	return meta	

def humanbitrate(s):
	items = s.split("/")
	a = float(items[0])
	b = float(items[1])
	bitrate = "%8.1f" % (a / b)
	return  bitrate.strip()

# крманда ffmpeg для получения постера
def poster(fn):
	seconds = 5
	path = fn
	image = postername(fn)
	ffmpeg = "ffmpeg -ss %s -i \"%s\" -f image2 -vframes 1 -y \"%s\"" % (seconds, path, image)
	return ffmpeg

def postername(file_path):
	file_name, file_extension = os.path.splitext(file_path)
	return file_name + ".jpg"

def check_file(file):
	meta = file_info(file)
	#print(meta)

	cmd = "ffprobe -v quiet -print_format json -show_format -show_streams \"%s\"" % file
	output = subprocess.run(cmd, capture_output=True, text=True, shell=True)

	# обычный вывод
	#output = subprocess.run(cmd, text=True, shell=True)
	#print(output)

	data = json.loads(output.stdout)

	info = {}
	info["audio"] = len(data["streams"]) > 1
	total_seconds = datetime.timedelta(seconds=float(data["format"]["duration"]))
	info["duration"] = total_seconds
	#print("seconds", total_seconds, file)
	info["hduration"] = "00:00:00"
	try:
		info["hduration"] = datetime.datetime.strptime(str(total_seconds), "%H:%M:%S.%f").strftime("%H:%M:%S")
	except:
		print("не удалось получить:", total_seconds)
	info["width"] = data["streams"][0]["width"]
	info["height"] = data["streams"][0]["height"]
	info["bit_rate"] = humanbitrate(data["streams"][0]["avg_frame_rate"])
	info["codec"] = data["streams"][0]["codec_name"]
	info["filename"] = meta["name"]
	info["ext"] = meta["ext"]
	info["crc"] = meta["crc"]
	info["created"] = meta["date"]
	info["hcreated"] = meta["date"].strftime("%d.%m.%Y")
	info["size"] = data["format"]["size"]
	info["hsize"] = humanbytes(data["format"]["size"])
	info["poster"] = poster(file)
	return info


def valid_file(file):
	valid_ext = ["mp4", "mpeg", "webm"]
	ext = file.split(".")[-1]
	for e in valid_ext:
		if ext == e:
			return True
	return False

# =================== начало программы ====================

dir = "d:\\vids\\ero"

files = os.listdir(dir)
for file in files:
	if valid_file(file):
		path = os.path.join(dir, file)
		print(path)
		try:
			meta = check_file(path)
			print(meta)
		except:
			print("=== Не удалось обработать файл:", path)
	else:
		print("пропускаем файл:", file)