Skip to content

Latest commit

 

History

History
234 lines (168 loc) · 7.71 KB

README.md

File metadata and controls

234 lines (168 loc) · 7.71 KB
Raw

Arkitektonika

Arkitektonika is a REST repository for NBT data. It accepts uploads of valid NBT data and stores them in a local folder while accounting for its metadata in a local sqlite database. Optionally, uploaded files can be expired based on the configurable age by running the prune script. Files can always be deleted via their deletion key.

Example Instances:

Address Expiry
https://api.schematic.cloud/ 30 days

To Run

With Docker

docker pull intellectualsites/arkitektonika

Available on https://hub.docker.com/r/intellectualsites/arkitektonika

From scratch

git clone https://github.com/IntellectualSites/Arkitektonika.git &&
cd Arkitektonika &&
yarn install

With Typescript transpiling (recommended)

yarn start:prod

Without Typescript transpiling

yarn start

Build image locally

Clone the entire repository and run the following commands:

docker build -t intellectualsites/arkitektonika:<TAG> .

Example docker compose:

version: '3.8'

services:
  arkitektonika:
    container_name: Arkitektonika
    image: intellectualsites/arkitektonika:dev
    restart: unless-stopped
    volumes:
      - /home/ark/data:/data # Mount the data folder (containing config file, database and schematic storage)
    environment:
      - LOG_LEVEL=DEBUG   # if debug logs should be printed to the console

/data is mounted to the host at /home/ark/data as that folder contains persistent data.

Prune data

Execute the start command with the prune flag to execute the prune routine:

yarn start:prod --prune

Set up Expiration

Create a cron job that runs at whatever frequency you desire. As an example, this will run the pruning script every 12 hours:

0 */12 * * * cd /srv/arkitektonika && /usr/bin/yarn start:prod --prune

Or with a docker-compose configuration:

0 */12 * * * cd /srv/arkitektonika && docker-compose run --rm arkitektonika node app/launch.js --prune

Configuration

{
  "port": 3000,
  "prune": 1800000,
  "maxIterations": 20,
  "maxSchematicSize": 1000000,
  "limiter": {
    "windowMs": 60000,
    "delayAfter": 30,
    "delayMs": 500
  }
}
Config Key Description
port on which port should the application bind
prune defines how old records must be to be deleted by the prune script (in ms)
maxIterations maximum amount of iterations to obtain a unique download and deletion token
maxSchematicSize maximum size of schematic files to be accepted (in bytes)
limiter.windowMs the frame of the limiter (after what duration should the limit gets reset)
limiter.delayAfter After how many requests during windowMs should delayMs be applied
limiter.delayMs How many ms should the request take longer. Formula: currentRequestDelay = (currentRequestAmount - delayAfter) * delayMs

File structure:

data
├── config.json
├── database.db
└── schemata
    ├── fe65d7edc37149c47171962dc26a039b
    └── a98f299c5cf294e6555617e83226bcdd

config.json holds the user configuration data
database.db holds the required data for each schematic
schemata holds all schematic file data

Routes

All routes will be available at the exposed port (e.g. localhost:3000).

Upload a file

POST INSTANCE_URL/upload: send your file as multipart/form-data; example:

curl --location --request POST 'http://localhost:3000/upload' \
--form 'schematic=@/path/to/plot.schem'

response:

code meaning
200 file was of valid NBT format and was accepted
400 file was not of valid NBT format
413 file payload was too large and rejected
500 file could not be found on disk after being uploaded (upload failed)

success body:

{
  "download_key": "db6186c8795740379d26fc61ecba1a24",
  "delete_key": "11561161dffe4a1298992ce063be5ff9"
}

The download key allows you to download the file, and the delete key lets you delete it. Share the download_key, but not the delete_key.

Check download headers

HEAD INSTANCE_URL/download/:download_key: check what headers you'd get if you sent a POST request for a file with the given download_key; example:

curl --location --head 'http://localhost:3000/download/db6186c8795740379d26fc61ecba1a24'

The response for this is in the form of status codes only.

Status-Code Meaning
200 File was found, prospective download would succeed
404 File was not found in the database
410 File metadata is in accounting table, but file is not on disk or already expired
500 An internal server error occurred due to corrupted metadata (missing data in database)

Download a file

GET INSTANCE_URL/download/:download_key: download a file with the given download_key; example:

curl --location --request GET 'http://localhost:3000/download/db6186c8795740379d26fc61ecba1a24'

response: see Check download headers above.

On success, the file is sent as an attachment for download to the browser / requester.

Check deletion headers

HEAD INSTANCE_URL/delete/:delete_key: check what headers you'd get if you sent a DELETE request for a file with the given delete_key; example:

curl --location --head 'http://localhost:3000/delete/11561161dffe4a1298992ce063be5ff9'

The response for this is in the form of status codes only.

Status-Code Meaning
200 File was found, prospective deletion would succeed
404 File was not found in the database
410 File metadata is in accounting table, but file is not on disk or already expired
500 An internal server error occurred due to corrupted metadata (missing data in database)

Delete a file

DELETE PUBLIC_URL/delete/:delete_key: delete a file with the given delete_key; example:

curl --location --request DELETE 'http://localhost:3000/delete/11561161dffe4a1298992ce063be5ff9'

response: see Check deletion headers above.

On success, the file is deleted and the record is marked as expired in the database.