full-text-search-api

A RESTful full text search service using Postgres.

Create records keyed by project that are indexed and searchable using Postgres full text search.

Usage

Fields

Field	Description
data	Records are searched on the data field. Can be any data type (number, string, array, object)
meta	Provided for any relevant information needed to associate with a discovered record. Can be any data type (number, string, array, object)
project	Project string, used to organize searchable data into buckets. This is also the partition key for Postgres.

cURL Commands

# create some data for a project called 'test'
curl -X POST -d '{"data": {"name": "Bart", "some_other_count": 30}, "meta":{"source_id": 10222}}' localhost:5001/test
curl -X POST -d '{"data": {"name": "Jimbo", "age": 30}, "meta":{"source_id": 10222}}' localhost:5001/test
curl -X POST -d '{"data": {"one": "another query for debug"}, "meta":{}}' localhost:5001/test
curl -X POST -d '{"data": {"one": "test query"}, "meta":{}}' localhost:5001/test

# full text search with `_search` query parameter
# multiple results containing `query`
$ curl -s "localhost:5001/test?_search=query" | jq "."
{
  "items": [
    {
      "id": "83947653-454d-4e29-85a4-0c4614fb3ed3",
      "project": "test",
      "data": {
        "one": "test query"
      },
      "meta": {}
    },
    {
      "id": "0b8961ea-7b76-44eb-ab49-3f2a23bf7f3f",
      "project": "test",
      "data": {
        "one": "another query for debug"
      },
      "meta": {}
    },
    {
      "id": "32c3675f-b922-4733-9041-9c7939b3894d",
      "project": "test",
      "data": {
        "one": "another query for debug"
      },
      "meta": {
        "source_id": 10001
      }
    }
  ]
}

# multiple results containing 30
$ curl -s "localhost:5001/test?_search=30" | jq "."
{
  "items": [
    {
      "id": "45de46fd-af9e-4f54-84ed-241a9f803b2d",
      "project": "test",
      "data": {
        "age": 30,
        "name": "Jimbo"
      },
      "meta": {
        "source_id": 10222
      }
    },
    {
      "id": "082bd15b-0e52-489b-9834-b31a07903b60",
      "project": "test",
      "data": {
        "name": "Bart",
        "some_other_count": 30
      },
      "meta": {
        "source_id": 10222
      }
    }
  ]
}

# single result for '"age": 30'
$ curl -s "localhost:5001/test?_search=%22age%22%3A%2030" | jq "."
{
  "items": [
    {
      "id": "45de46fd-af9e-4f54-84ed-241a9f803b2d",
      "project": "test",
      "data": {
        "age": 30,
        "name": "Jimbo"
      },
      "meta": {
        "source_id": 10222
      }
    }
  ]
}

Run Locally

This project runs two docker containers; api and postgres.

# build without cache
rebuild:
	docker-compose build --no-cache

# build image
build:
	docker-compose build

# spin up containers (postgres and api)
up:
	docker-compose up -d

# bring down running containers
down:
	docker-compose down

# stop running containers
stop:
	docker-compose stop

# remove containers and images
remove:
	docker-compose rm -f

# cleanup images and volumes
clean:
	docker-compose down --rmi all -v --remove-orphans

License

MIT Copyright © Nick Sjostrom

Testing again