Docker image¶

Docker images are built by GitHub Actions from the LIMA sources and published on Docker Hub. This is the easiest way to get the latest version of LIMA.

Image	Content
`aymara/lima-ubuntu22.04`	LIMA on Ubuntu 22.04
`ghcr.io/aymara/lima:master`	LIMA on Debian 12 (GitHub Container Registry), latest `master`

The Docker Hub image is tagged so that you can pin the exact code it contains:

Tag	Meaning
`latest`	Latest build of the `master` branch
`YYYYMMDD`	Build of `master` from that day
`sha-<commit>`	Build of that commit
`<x.y.z>`	Release `vx.y.z`

Running LIMA in a container¶

docker pull aymara/lima-ubuntu22.04:latest
docker run -it --rm -v "$PWD":/data aymara/lima-ubuntu22.04:latest bash

The image does not include language models. Inside the container, install the ones you need and analyze your files:

lima_models.py -l eng-UD_English-EWT
analyzeText -l eng-UD_English-EWT -p deepud /data/my-text.txt

To keep the models between runs, mount a volume on the models directory, e.g. -v lima-models:/root/.local/share/lima.

Running the LIMA server¶

The image also contains limaserver, an HTTP server exposing the analyzer (port 8080 by default). Give it the languages and pipelines to load (defaults are read from lima-server.xml; see limaserver --help):

docker run -p 8080:8080 -v lima-models:/root/.local/share/lima \
    aymara/lima-ubuntu22.04:latest \
    limaserver --language eng-UD_English-EWT --pipelines deepud

Then send it some text:

curl "http://localhost:8080/?lang=eng-UD_English-EWT&pipeline=deepud" \
    --data-binary @file.txt

Custom configuration¶

To change LIMA's configuration inside the container, put your own configuration files in a folder of the host, mount it in the container and make the LIMA_CONF variable point to it, before the system configuration (see Configuring LIMA):

docker run -it --rm -v "$PWD/my-conf":/my-conf \
    -e LIMA_CONF=/my-conf:/usr/share/config/lima \
    aymara/lima-ubuntu22.04:latest bash