Gearr

Warning: Use at your own risk! Gearr is currently in alpha stage and has not undergone extensive testing.

Gearr is an open-source video transcoding software designed to efficiently convert videos to the x265 format. It leverages the powerful ffmpeg library under the hood to ensure high-quality transcoding. Gearr also supports distributed transcoding across multiple agents, facilitating faster processing of video files.

Features

• x265 Transcoding: Convert videos to the x265 format for improved compression and quality.
• Distributed Transcoding: Utilize multiple agents to distribute transcoding tasks, speeding up the process.
• Queue system: Connect to a RabbitMQ server to manage transcoding job queues.
• PGS Subtitle Conversion: Extract PGS subtitles from video files and convert them to STR format for wider compatibility.

Container Images

• Server: ghcr.io/pando85/gearr:latest-server
• Worker: ghcr.io/pando85/gearr:latest-worker
• PGS Worker: ghcr.io/pando85/gearr:latest-worker-pgs

Configuration

Environment Variables

The application supports configuration through environment variables. Below is a table of supported environment variables and their default values:

Server

Variable	Description	Default Value
BROKER_HOST	Broker host address	localhost
BROKER_PORT	Broker port	5672
BROKER_USER	Broker username	broker
BROKER_PASSWORD	Broker password	broker
BROKER_TASKENCODEQUEUE	Broker tasks queue name for encoding	tasks
BROKER_TASKPGSQUEUE	Broker tasks queue name for PGS to SRT conversion	tasks_pgstosrt
BROKER_EVENTQUEUE	Broker tasks events queue name	task_events
DATABASE_DRIVER	Database driver	postgres
DATABASE_HOST	Database host address	localhost
DATABASE_PORT	Database port	5432
DATABASE_USER	Database username	postgres
DATABASE_PASSWORD	Database password	postgres
DATABASE_DATABASE	Database name	gearr
DATABASE_SSLMODE	Database SSL mode	disable
LOG_LEVEL	Log level (debug, info, warning, error, fatal)	info
SCHEDULER_DOMAIN	Base domain for worker downloads and uploads	http://localhost:8080
SCHEDULER_SCHEDULETIME	Scheduling loop execution interval	5m
SCHEDULER_JOBTIMEOUT	Requeue jobs running for more than specified duration	24h
SCHEDULER_DOWNLOADPATH	Download path for workers	/data/current
SCHEDULER_UPLOADPATH	Upload path for workers	/data/processed
SCHEDULER_MINFILESIZE	Minimum file size for worker processing	100000000
WEB_PORT	Web server port	8080
WEB_TOKEN	Web server token	admin

Worker

Variable	Description	Default Value
BROKER_HOST	Broker host address	localhost
BROKER_PORT	Broker port	5672
BROKER_USER	Broker username	broker
BROKER_PASSWORD	Broker password	broker
BROKER_TASKENCODEQUEUE	Broker tasks queue name for encoding	tasks
BROKER_TASKPGSQUEUE	Broker tasks queue name for PGS to SRT conversion	tasks_pgstosrt
BROKER_EVENTQUEUE	Broker tasks events queue name	task_events
LOG_LEVEL	Set the log level (options: "debug", "info", "warning", "error")	info
WORKER_TEMPORALPATH	Path used for temporal data	system temporary directory
WORKER_NAME	Worker name used for statistics	hostname
WORKER_THREADS	Number of worker threads	number of CPU cores
WORKER_ACCEPTEDJOBS	Type of jobs the worker will accept	["encode"]
WORKER_MAXPREFETCHJOBS	Maximum number of jobs to prefetch	1
WORKER_ENCODEJOBS	Number of parallel worker jobs for encoding	1
WORKER_PGJOBS	Number of parallel worker jobs for PGS to SRT conversion	0
WORKER_DOTNETPATH	Path to the dotnet executable	"/usr/bin/dotnet"
WORKER_PGSTOSRTDLLPATH	Path to the PGSToSrt.dll library	"/app/PgsToSrt.dll"
WORKER_TESSERACTDATAPATH	Path to the tesseract data	"/tessdata"
WORKER_STARTAFTER	Accept jobs only after the specified time (format: HH:mm)	-
WORKER_STOPAFTER	Stop accepting new jobs after the specified time (format: HH:mm)	-
SCHEDULER_DOMAIN	Base domain for worker downloads and uploads	http://localhost:8080
SCHEDULER_SCHEDULETIME	Scheduling loop execution interval	5m
SCHEDULER_JOBTIMEOUT	Requeue jobs running for more than specified duration	24h
SCHEDULER_DOWNLOADPATH	Download path for workers	/data/current
SCHEDULER_UPLOADPATH	Upload path for workers	/data/processed
SCHEDULER_MINFILESIZE	Minimum file size for worker processing	100000000
WEB_PORT	Web server port	8080
WEB_TOKEN	Web server token	admin

Configuration File

The application also supports configuration through a YAML file. The default configuration file format is YAML. If you want to use a different file format, please specify it in the CONFIG_FILE environment variable.

Example YAML configuration file:

Server

logLevel: info

broker:
  host: localhost
  port: 5672
  user: broker
  password: broker
  taskEncodeQueue: tasks
  taskPGSQueue: tasks_pgstosrt
  eventQueue: task_events

database:
  Driver: postgres
  Host: localhost
  port: 5432
  User: postgres
  Password: postgres
  Database: gearr
  SSLMode: disable

scheduler:
  domain: http://localhost:8080
  scheduleTime: 5m
  jobTimeout: 24h
  downloadPath: /data/current
  uploadPath: /data/processed
  minFileSize: 100000000

web:
  port: 8080
  token: admin

Worker

broker:
  host: localhost
  port: 5672
  user: broker
  password: broker
  taskEncodeQueue: tasks
  taskPGSQueue: tasks_pgstosrt
  eventQueue: task_events

logLevel: info

worker:
  temporalPath: /path/to/temp/data
  name: my-worker
  threads: 4
  acceptedJobs:
    - encode
  maxPrefetchJobs: 2
  encodeJobs: 2
  pgJobs: 1
  dotnetPath: /usr/local/bin/dotnet
  pgsToSrtDLLPath: /custom/path/PgsToSrt.dll
  tesseractDataPath: /custom/tessdata
  startAfter: "08:00"
  stopAfter: "17:00"

Client Execution

Worker

DIR=/data/images/encode

mkdir -p $DIR
docker run -it -d --restart unless-stopped --cpuset-cpus 16-32 \
    --name gearr-worker --hostname $(hostname) \
    -v $DIR:/tmp/ ghcr.io/pando85/gearr:latest-worker \
    --broker.host gearr.example.com

Note: Adjust the --cpuset-cpus and other parameters according to your system specifications.

PGS Worker

DIR=/data/images/pgs

mkdir -p $DIR
docker run -it -d --restart unless-stopped \
    --name gearr-worker-pgs --hostname $(hostname) \
    -v $DIR:/tmp/ ghcr.io/pando85/gearr:latest-worker-pgs \
    --broker.host gearr.example.com

Warning: The PGS agent must be started in advance if PGS is detected. It should run before detection to create the RabbitMQ queue.

Integrations

Add movies from Radarr

go run ./integrations/radarr/add/main.go --api-key XXXXXX --url https://radarr.example.com --movies 5 --gearr-url 'https://gearr.example.com' --gearr-token XXXXXX

Feel free to customize the parameters based on your Radarr and Gearr setup.

Update movies in Radarr

In your radarr server:

MOVIDES_DIR=/movies
find ${MOVIES_DIR} -name '*_encoded.mkv'

Then execute:

go run ./integrations/radarr/update/main.go --api-key XXXXXX --url https://radarr.example.com "${FIND_OUTPUT}"

Then you can go to Radarr: Edit Movies -> Select All -> Rename Files

Add episodes from Sonarr

go run ./integrations/sonarr/add/main.go --api-key XXXXXX --url https://sonarr.example.com --series 5  --episodes 10 --gearr-url 'https://gearr.example.com' --gearr-token XXXXXX

Feel free to customize the parameters based on your Radarr and Gearr setup.

Roadmap

I'm currently not developing it more but if I want to code something I will:

• Remove RabbitMQ in flavour of a gRPC API to communicate agents just with the server.
• Replace repository by an ORM with Postgres and SQlite support.