NestJS Search

Search package for NestJS with MongoDb and AWS DocumentDB

PS, It is a copy of Nestjs-Keyset-Paginator, was added new features(i.e. between was added)

Installation

Use the package manager npm to install NestJS Search.

npm i nestjs-search

Usage

• In example.controller.ts use PaginationDto to Validate params and pass it to service.

import { PaginationDto, projectionDto } from 'nestjs-search'

@Controller('example')
export class ExampleController {
    constructor(private readonly exampleService: ExampleService) {}

    @Get()
    findAll(@Body() params: PaginationDto) {
        return this.exampleService.findAll(
            params.skip,
            params.limit,
            params?.start_key,
            params?.sort?.field,
            params?.sort?.order,
            params?.filter,
            params?.projection
        )
    }
}

• Then in example.service.ts pass those params to "paginate()" along with you model (Mongoose Model).

import paginate, { filterDto, projectionDto } from 'nestjs-search'

@Injectable()
export class ExampleService {
    constructor(
        @Inject(EXAMPLE_MODEL)
        private readonly exampleModel: Model<ExampleDocument>
    ) {}

    async findAll(
        skip = 0,
        limit = 10,
        start_key?,
        sort_field?: string,
        sort_order?: number,
        filter?: filterDto[],
        projection?: projectionDto[]
    ) {
        return paginate(this.exampleModel, skip, limit, start_key, sort_field, sort_order, filter, projection)
    }
}

• Paginate function will return with promise of:

{ docs: docs, next_key }

Example param

Example:-

{
    "filter": [
        {
            "name": "score",
            "value": 400,
            "operator": "lt"
        },
        {
            "name": "isPassed",
            "value": true,
            "operator": "eq"
        },
        {
            "name": ["outer_field_name", "inner_field_name"],
            "value": "user one",
            "operator": "eq"
        },
        {
            "name": "time",
            "arr_value": [40, 60],
            "operator": "in"
        },
        {
            "name": "left_count",
            "arr_value": [0, 1],
            "operator": "nin"
        }
    ],
    "sort": {
        "field": "score",
        "order": 1
    },
    "projection": [
        {
            "name": "password",
            "mode": 0
        }
    ],
    "limit": 4
}

Please note: for the same same field, when you use "lt" and "gt" filters at the same time, the filters overriding each other, instead please use "between" as follows.

Example: between

{
    "filter": [
        {
            "name": "eventStartDateTime",
            "arr_value": ["2021-11-16", "2021-11-17"],
            "operator": "between"
        }
    ]
}

or any other ISODate

{
    "filter": [
        {
            "name": "eventStartDateTime",
            "arr_value": ["2021-11-16T11:30:01.001+00:00", "2021-11-17T11:30:01.001+00:00"],
            "operator": "between"
        }
    ]
}

Example: fulltext search Please note: Aws DocumentDB doesn't support fulltext search now(suggests Elastic Search instead). Therefore, you cannot use this search on DocDB. Instead, you can use like search option like we shown on the next example.

{
    "filter": [
        {
            "name": "text",
            "value": "zero2hero metaverse event",
            "operator": "search"
        }
    ]
}

it works regarding to MongoDb Text Indexes, the document link: https://www.mongodb.com/docs/manual/core/index-text/

Example: Like search on multiple fields Please note: Aws DocumentDB doesn't support fulltext search now(suggests Elastic Search instead). Therefore, we added enhanced like search on multiple field. Just keep in your mind, more data will consume more CPU and memory on here. You may think to move to MongoDB to save some cost instead.

{
    "filter": [
        {
            "name": "groupName",
            "value": "Test",
            "operator": "like",
            "mode": "bnm"
        },
        {
            "name": "groupDetails.groupDetails",
            "value": "test",
            "operator": "like",
            "mode": "bnm"
        }
    ]
}

it works regarding to MongoDb Text Indexes, the document link: https://www.mongodb.com/docs/manual/core/index-text/

Example: in

{
    "filter": [
        {
            "name": "eventCategory",
            "arr_value": ["sport", "training"],
            "operator": "in"
        }
    ]
}

• As response, you will also get "next_key".

Example:

{
    "next_key": [
        {
            "key": "_id",
            "value": "61a4c444f9534392c70afaf6"
        },
        {
            "key": "score",
            "value": 100
        }
    ]
}

• To get next page use this "next_key" object as "start_key" in next request.

Example:

{
    "filter": [
        {
            "name": "score",
            "value": 400,
            "operator": "lt"
        },
        {
            "name": "isPassed",
            "value": true,
            "operator": "eq"
        }
    ],
    "sort": {
        "field": "score",
        "order": 1
    },
    "limit": 4,
    "start_key": [
        {
            "key": "_id",
            "value": "61a4c444f9534392c70afaf6"
        },
        {
            "key": "score",
            "value": 100
        }
    ]
}

• If you provide "start_key" this will skip previous Documents.

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate if applicable.

License

MIT