Codegen

You can generate a few different types of code when using rindexer codegen.

GraphQL

You can generate .graphql prebuilt queries to get up and running in seconds. These will be generated in a queries folder in the root of where the rindexer yaml is.

rindexer codegen graphql

By default it will point to http://localhost:3001 graphql endpoint, you can change this by passing the endpoint flag:

rindexer codegen graphql --endpoint=YOUR_GRAPHQL_API_URL

TypeScript

graphql-codegen is the best tool on the market to generate TypeScript typings for your GraphQL queries, mutations, and subscriptions.

learn about the codegen.ts config here

the graphql API url is the schema in the config, you can set this to your graphql endpoint like so:

import { CodegenConfig } from '@graphql-codegen/cli'
 
const config: CodegenConfig = {
  // this is YOUR_GRAPHQL_API_URL
  schema: 'http://localhost:3001/graphql', 
  ...
}
 
export default config

then how you hook up the config with your tool of choice, below are some links to documentation:

React Apollo - https://the-guild.dev/graphql/codegen/plugins/typescript/typescript-react-apollo#with-react-hooks
React Query - https://the-guild.dev/graphql/codegen/plugins/typescript/typescript-react-query
Node app - https://the-guild.dev/graphql/codegen/plugins/typescript/typescript-urql

.NET, Dart, Java, Flow

codegen for other languages can be found here

Typings

When creating a new rust project with rindexer it will create you a typings folder, this has pretty advanced typings for all your contracts, events and network information. This is generated from the ABIs you provide in the YAML configuration file. This folder is not meant to be manually edited and should always be generated using codegen.

You can regenerate the typings folder by running the following command:

rindexer tries to be as smart as possible when it comes to updating the typings based on the rindexer.yaml, it will resolves as much as it can without needing a regenerate but like any codegen tool if you change certain aspects it does need to be regenerated.

if you change any of these properties in the rindexer.yaml file it will need to be regenerated:

indexer name
anything in the network section including adding and removing networks
enabling or disabling a new storage provider
changing the contract name
changing from address contract indexing to filter indexing or vice versa
changing the contract ABI
anything in the global section

Also if you do regenerate your indexer files may need to be updated to match the new typings, you can manually migrate them or generate them again using indexer codegen command

rindexer codegen typings

Indexers

When creating a new rust project with rindexer it will create you a indexers folder, this is where you will write your custom logic for the indexer. This is where you will do all your indexing logic, you can do anything you want in here, you can do http requests, on chain lookups, custom logic, custom DBs, anything you can think of. rindexer gives you the foundations and also baked in extendability. Rust enforces a strong type system, all logs will be streamed to you just focus on the logic you want.

By default if you turn storage postgres on in the YAML configuration file it will also create you postgres tables, also write SQL for you to use and expose you a postgres client. This is a great starting point for you to build on.

The tables creation can be skipped by using the disable_create_tables in the YAML configuration file.

If you also enable the CSV storage it will also generate code in the handler to write to that CSV files.

You can regenerate the indexers folder by running the following command, please note this will overwrite any custom logic you have written if you run it on an existing project.

rindexer codegen indexers