vvd-extractor

Extractor

This module, as the final module, mainly takes charge of two things:

1. Extracting and wrapping the audio source from the downloaded media source from the previous module (vvd-downloader).
2. Filling the metadata and embedding the thumbnail of the extracted audio file with the song information from the -songInfo JSON file

The output of this module is only a directory of audio files with metadata and thumbnail embedded.

Before moving on

If you haven't read the common document yet, please read it first: Common document

Additional Prerequisites

Besides the common prerequisites, you also need:

1. Python3 (This module relies on some python codes to perform metadata manipulation)
2. pip install mutagen (Run this command to install the mutagen library)
3. FFmpeg (Highly recommended the yt-dlp's distribution)
4. MediaInfo CLI

All configurations

All configurations can be found in application.yml file.

Below is a copy of the content of the application.yml file for your reference. However, make sure to check for any updates to the file yourself:

vocadb-video-downloader-new/vvd-extractor/src/main/resources/application.yml

Lines 1 to 62 in 292351f

	io: # all fields are required
	# this must be pointing to the output directory of the previous module, the task producer module. In another word, the directory specified in the 'io.output-directory' field of the previous module
	# can be an absolute path or a relative path from the application current running directory
	input-directory:
	# the output directory of this module, can be an absolute path or a relative path from the application current running directory
	output-directory:
	# the error directory of this module, used for reporting errors for debugging, can be an absolute path or a relative path from the application current running directory
	error-directory:
	config: # configuration
	# all fields are required
	environment:
	# specify the launch CMD of each dependency as a list of string.
	# All commands will be executed using the root directory of the project as the current directory
	python-launch-cmd:
	ffmpeg-launch-cmd:
	mediainfo-launch-cmd:
	# all fields are required
	batch:
	# number of files to process in parallel, default is 0 which represents # of cores
	batch-size: 0
	# all fields are required
	process:
	# This two fields control the timeout of each process in this module
	# Each process in this module should be run in fairly quick time, even a one-hour video should be run in less than 2 minutes
	# if you think you need more time, change the setting here
	timeout: 5
	unit: minutes
	# all fields are required
	retry:
	retry-on-extraction: 2 # number of times to retry on audio extraction error
	retry-on-tagging: 2 # number of times to retry on audio file tagging error
	# optional fields
	current-time:
	# The module will modify the last modified time of the audio file to match the order of input tasks.
	# Specifically, a task with a smaller order number in the `-task` JSON file will have an earlier last modified time.
	# But the module needs a base timestamp to calculate the last modified time of each audio file
	# By default, this field is empty means using the time when this app is launched.
	# If you want to start from a specific time, change the setting here
	# The format is yyyy-MM-dd HH:mm:ss, with the local time zone in this running device
	start-from: ""
	# spring setting, DON't touch. unless you know what you are doing
	spring:
	jackson:
	deserialization:
	fail-on-unknown-properties: true
	serialization:
	indent-output: true
	# default-property-inclusion: non_null # disabling for now, we want 3rd app read our json easily
	config:
	import: optional:file:./my-extractor-config.yml,optional:file:./my-extractor-config.yaml,optional:file:./my-extractor-config.properties
	logging:
	level:
	mikufan: info