npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

postgraphile-plugin-custom-filter

v1.0.1

Published

Customized filter on Connections in postgraphile

Downloads

1,987

Readme

postgraphile-plugin-custom-filter

plugin for postgraphile that adds customized filter

Inspired by the postgraphile connection filter plugin, this plugin is to add a "CustomFilter" object in each of the Connection Arguments. Just by appending this plugin, the CustomFilter would be empty. It requires the user to add some code for the custom filters.

The plugin requires the user to code up the logic of custom filter yourself. When using this plugin, an option parameter must be specified. Currently the option parameter only needs a filters attribute, which consist of an object that describes all the filters to all the models. The structure of filter is

filter = {
  modelName:{
    fieldName:{
      fieldType:'String'|'Boolean'|'Int'|'Float',
      modifier:()=>{}
    }
  }
}

This setup ensures that for each Model there's one object, and in that object, we add multiple fieldNames, for each fieldName, it should only have one object. This avoid adding duplicated fieldNames for different purposes.

const plugin = require('postgraphile-plugin-custom-filter');

const options = {
  filters:{
    User:{
      foo:{
        fieldType:'String',
        modifier:(queryBuilder, value, build)=>{console.log(value)}
      }
    }
  }
}

buildPlugin = (build)=>{
  plugin(build,options);
}

const express = require("express");
const { postgraphile } = require("postgraphile");

const app = express();

app.use(postgraphile(process.env.DATABASE_URL || "postgres://localhost/","public",{
  appendPlugins:[
    // put the plugin in here
    buildPlugin
  ]
}));

app.listen(process.env.PORT || 3000);

This will create a 'foo' field in CustomFilter, then after the postGraphile constructed the query, the hooks will call your modifier function with 4 parameters, the queryBuilder, (details in QueryBuilder.js), the value of the 'foo' parameter, and two help object, 'build', and 'context'. These two objects are described more in postGraphile plugin page. (see more in plugins and server plugin).

These two pages helped a lot when writing your own plugins, but it still lacks a lot of details. To be more familiar with the hook system, a lot of digging through source was required. So I created this plugin so that it gives some freedom for others to add query based on their needs. Let's see an example!

Example

Suppose you have a UserConnection, this plugin will create a arg called CustomFilter in the UserConnect(xxx) parameter allUsers(first:10,CustomFilter:{...}){ } Suppose the User is associated with a role with a role_id field. And Role have a column called role_name. You want to get all users that has role_name='admin'. This is hard to do because postgraphile does not support filter based on nested object attributes. If you know the role_id, say

| id | name | | --------- |:-----:| | 1 | admin |

You can fake it by

# the condition arg is already provided out of box with postGraphile
allUsers(first:10,conditions:{role_id:1}){
}

But what if it's more complicated?

Add extra where

Assume the postgraphile generates a correct query that does all other filter, order. What we want is to add an extra where statement.

We can use this customized filter, let's add a field roleName.

const filter = {
  User:{
    byRoleName:{
      fieldType:'String',
      modifier: (queryBuilder, value, build)=>{
      // I'll get into it later
      }
    }
  }
}

This will allow you to query the users with a 'byRoleName' parameter, like this:

allUsers(CustomFilter:{
  byRoleName:"admin"
}){
  nodes{
  ...
  }
}

Now we need to modify the queryBuilder when the request send {byRoleName:"admin"} The way to do it is through that modifier function. The queryBuilder is the main object that postGraphile uses to create the query. There's not much documentation to it, but if you create breakpoints, and explore the object, you could see a lot of functions it provides. It also holds the currently created query.

Now we want to modify the query. Intuitively, we want

select * from users
left join roles on users.role_id = roles.id
where roles.name = 'Admin';

But the queryBuilder don't let you do join, Instead we'll have to use an workaround.

select * from users
where users.id in (select id from users left join roles on users.role_id = roles.id where roles.name = 'Admin')

This subquery looks a lot more expensive, and if you are from the old days SQL, when join is much quicker than subquery, you probably gonna hate it. But postgres query optimizer does an amazing job that it eventually optimizes the query and produce the result without much time difference. (as long as you index these keys).

Now we know what to put in, but there is one more important thing. The queryBuilder does not allow just random string parameters, which means

modifier = (queryBuilder, value, build)=>{
  if(value){
    queryBuilder.where("users.id in (select id from users left join roles on users.role_id = roles.id where roles.name = "+value+")");
  }
}

is not gonna work. Also the queryBuilder actually uses alias for table names, so "users.id" won't work neither. To get the table name we'll need "queryBuilder.getTableAlias()" function, and each field is a pgSql.identifier, each value is a pgSQL.value. (read more in pgSql). So this where is convert to

modifier = (queryBuilder, value, build)=>{
  const pgSql = build.pgSql;
  if(value){
    queryBuilder.where(pgSql.query`${queryBuilder.getTableAlias()}.${pgSql.identifier("id")} in (${pgSql.query`select id from users left join roles on users.role_id = roles.id where roles.name = $pgSql.value(value)}`})`);
  }
}

Tricks

Even though this plugin is called custom filter, it really opens up a lot of possibilities because you are exposed to the queryBuilder, the key of the queries. You could also sort it with

queryBuilder.sortBy(pgSql.query`select name from roles where roles.id = ${queryBuilder.getTableAlias()}.${pgSql.identifier("role_id")}`)

This will add a sorter to sort it by the role name. Of course you'll remember to not specify other orderBy. A full example of sort by nested fields

const filter = {
  User:{
    sortByRoleName:{
      fieldType:'String',
      modifier: (queryBuilder, value, build)=>{
        if(value){
          queryBuilder.sortBy(pgSql.query`
          select name from roles where roles.id = ${queryBuilder.getTableAlias()}. ${pgSql.identifier("role_id")}
          `,true) // true/false means asc/desc        
        }
      }
    }
  }
}

Then if you do a query

allUsers(orderBy:[],customFilter:{
  sortByRoleName:true
}){
  node{
    name
  }
}

Then you'll be able to see the list of users that's sorted by the role name. However, if you do

allUsers(orderBy:PRIMARY_KEY_ASC,customFilter:{
  sortByRoleName:true
}){
  node{
    name
  }
}

Then you won't be able to see the user list in correct order.

(Notice the orderBy:[], it is important to NOT specify any other orderBy). The "orderBy" function allows you to specify true/false as direction. If you do not specify order is unique, (queryBuilder.orderIsUnique()). Then it will append the PRIMARY_KEY as a back up order after the initial order, which ensure the order is unique.

License MIT Contact hansolo[email protected]