Skip to content

Latest commit

 

History

History
200 lines (155 loc) · 6.74 KB

README.md

File metadata and controls

200 lines (155 loc) · 6.74 KB

Query string to mongodb paginated aggregation

Coverage lines Coverage functions Coverage branches Coverage statements

License code style: prettier Standard - JavaScript Style Guide

✍️ Introduction

The "query to mongo aggregate" (q2ma in short) is a tool to execute a mongodb paginated query (using find or aggregate) based on URI query parameters using query-to-mongo.

If you need to execute a aggregation query and provide your own aggregation stages (other than pagination related stages) you can simply pass a pipeline option. Otherwise the paginated query is executed using find.

⛹️ Examples

Simple Paginated Query

To run a simple paginated query on a collection (which is internally executed using find):

const { q2ma } = require(q2ma);
const myModel = require("./model");
const queryString = "name=john&age>21&fields=name,age&sort=name,-age&offset=0&limit=10";

await q2ma(myModel, {queryString});

/* 
{
  Result: [
    {
      "_id": "23fr42tv426gv"
      "name": "john 1",
      "age": 25
    },
    {
      "_id": "ryb456ubn56un"
      "name": "john 2",
      "age": 24
    }
  ],
  Total: 10
}
*/

Using query-to-mongo we produce the following from that queryString:

const criteria = {
  name: 'john',
  age: { $gt: 21 }
}
const fields = {
  name: true, age: true
}
const options = {
  sort: { name: 1, age: -1 },
  offset: 10,
  limit: 10
}

Then q2ma will execute the following queries in parallel:

myModel.find(criteria, projects, options)
myModel.find(criteria).count()

Paginated Aggregation Query

To apply pagination on an aggregation query

const { q2ma } = require(q2ma);
const myModel = require("./model");
const queryString = "age>21&fields=_id,names&sort=_id&offset=0&limit=10";
const pipelines = [
  { $group: {
    _id: "age",
    names: { $push: "$name" }
  } }
];

await q2ma(myModel, {queryString, pipelines});

/* 
{
  Result: [
    {
      "age": 24,
      "names": ["john 1"]
    },
    {
      "age": 25,
      "names": ["john 2", "susie"]
    }
  ],
  Total: 10
}
*/

Then q2ma produces and executes the following aggregation query using that queryString and pipelines:

myModel.aggregate([
  { $match: { age: { $gt: 21 } } },
  { $group: {
    _id: "age",
    names: { $push: "$name" }
  } },
  { $facet: {
      total: [{ $group: { _id: "total", sum: { $sum: 1 } } }],
      pagedResult: [{ $sort: { name: 1, age: -1, _id: -1 } }, { $skip: 0 }, { $limit: 10 }, { $project: { _id: 1, names: 1 } }],
  } }
]);

Note that your filters will go into a match stage before your own pipeline stages and the sorting and paging related stages goes last. To change this behavior you could pass {matchPosition: 'END'} in the options.

🚀 Installation

$ npm i q2ma
# or
$ yarn add q2ma

📖 Documentation

q2ma(collection, {options})
Parameter Format Description Required
collection Object mongo driver collection reference or mongoose model name
options Object Options is an object like follow: { filter, project, options, pipelines, queryString, dateFields, dateFormat, matchPosition }

options

If you have mongodb pipelines aggregation you can use following combination:

{pipelines, queryString, dateFields, dateFormat, matchPosition}

Parameter Format Description Example Default Value
pipelines Array --- [ { $unwind: 'profile.cards' } ] ---
queryString String --- name=john&age>21&fields=name,age&sort=name,-age&offset=10&limit=10 ---
dateFields String Array --- --- ["createdAt", "modifiedAt", "updatedAt", "removedAt", "deletedAt", "verifiedAt", "confirmedAt", "timestamp"]
dateFormat String enum `NUMBER DATE` ---
matchPosition String where do you want to add your custom pipelines before queryString match or after it. enum `START END` ---

If your query is simple and then need some kind of filter and projection like find or findOne you can use following combination:

{filter, project, options, queryString, dateFields, dateFormat}

Parameter Format Description Example Default Value
filter Object like input parameter to find/findOne {name: "Ed", 'profile.card': xx-xxx-xxx} ---
project Object like input parameter to find/findOne {profile: 1, name: 1, transaction: 1} ---
options Object like input parameter to find/findOne {sort: {'profile.phone': -1}, skit: 10} ---
queryString String like url string name=john&age>21&fields=name,age&sort=name,-age&offset=10&limit=10 ---
dateFields String Array --- ['timeAt'] ["createdAt", "modifiedAt", "updatedAt", "removedAt", "deletedAt", "verifiedAt", "confirmedAt", "timestamp"]
dateFormat String enum `NUMBER DATE` ---

NOTE: Sort default is based on _id.

🤝 Contributing

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

Note: Please make sure to update tests as appropriate.

👋 Contact

If you have any further questions, please don’t hesitate, you can reach me by the following:

📝 License

This project is Apache licensed.