imaging-plaza-search

A microservice for fetching relevant softwares to a certain search term + filters input. This service will retrieve software from GraphDB and then perform a fuzzy search using Fuzon.

How to use it?

It's necessary to send a post request to the API endpoint (URL/v1/search) with a body following this example. If you want to try you can do it directly in URL/docs by clicking in the Try it out button.

{
    "search": "deep",
    "filters": [
        {
            "key": "programmingLanguage",
            "schema_key": "schema:programmingLanguage",
            "value": [
                "Python"
            ]
        },
        {
            "key": "featureList",
            "schema_key": "schema:featureList",
            "value": [
                "Object detection"
            ]
        }
    ]
}

The reply should looks like:

[
  {
    "label": "DeepLabCut",
    "uri": "<https://github.com/DeepLabCut/DeepLabCut>"
  },
  {
    "label": "stardist",
    "uri": "<https://github.com/stardist/stardist>"
  },
  {
    "label": "spotiflow",
    "uri": "<https://github.com/weigertlab/spotiflow>"
  },
  {
    "label": "DEFCoN",
    "uri": "<https://github.com/LEB-EPFL/DEFCoN>"
  },
  {
    "label": "detection-attributes-fields",
    "uri": "<https://github.com/vita-epfl/detection-attributes-fields>"
  },
  {
    "label": "butterflydetector",
    "uri": "<https://github.com/vita-epfl/butterflydetector>"
  }
]

How to deploy it?

First define the env variables needed to connect to graphDB. You can copy and rename .env.dist into .env.

Deploy with docker using Just

If you have just available you can build and run the image as follows:

just image build

just image run

Then connect to localhost:7123. The port is hardcoded in tools/just/image.just

Deploy directly with docker

If you prefer you can run this directly with docker

docker build -t imaging-plaza-search -f tools/image/Dockerfile .

And then run:

docker run --rm --name imaging-plaza-search -p 7123:15400 --env-file .env imaging-plaza-search

How to develop?

To develop and see changes in real time you can mount the source folder into docker like this:

docker run --rm --name imaging-plaza-search -p 7123:15400 -v ./src/:/app/src --env-file .env imaging-plaza-search

Credits

This MS has been developed by the Swiss Data Science Center.