Query Methods

TL;DR

The most basic way to expose your canister's functionality publicly is through a query method. Here's an example of a simple query method named getString: 

import { Canister, query, text } from 'azle/experimental';

export default Canister({
    getString: query([], text, () => {
        return 'This is a query method!';
    })
});

Query methods are defined inside of a call to Canister using the query function.

The first parameter to query is an array of CandidType objects that will be used to decode the Candid bytes of the arguments sent from the client when calling your query method.

The second parameter to query is a CandidType object used to encode the return value of your function to Candid bytes to then be sent back to the client.

The third parameter to query is the function that receives the decoded arguments, performs some computation, and then returns a value to be encoded. The TypeScript signature of this function (parameter and return types) will be inferred from the CandidType arguments in the first and second parameters to query.

getString can be called from the outside world through the IC's HTTP API. You'll usually invoke this API from the dfx command line, dfx web UI, or an agent.

From the dfx command line you can call it like this: 

dfx canister call my_canister getString

Query methods are read-only. They do not persist any state changes. Take a look at the following example: 

import { Canister, query, text, Void } from 'azle/experimental';

let db: {
    [key: string]: string;
} = {};

export default Canister({
    set: query([text, text], Void, (key, value) => {
        db[key] = value;
    })
});

Calling set will perform the operation of setting the key property on the db object to value, but after the call finishes that change will be discarded.

This is because query methods are executed on a single node machine and do not go through consensus. This results in lower latencies, perhaps on the order of 100 milliseconds.

There is a limit to how much computation can be done in a single call to a query method. The current query call limit is 5 billion Wasm instructions. Here's an example of a query method that runs the risk of reaching the limit: 

import { Canister, nat32, query, text } from 'azle/experimental';

export default Canister({
    pyramid: query([nat32], text, (levels) => {
        return new Array(levels).fill(0).reduce((acc, _, index) => {
            const asterisks = new Array(index + 1).fill('*').join('');
            return `${acc}${asterisks}\n`;
        }, '');
    })
});

From the dfx command line you can call pyramid like this: 

dfx canister call my_canister pyramid '(1_000)'

With an argument of 1_000, pyramid will fail with an error ...exceeded the instruction limit for single message execution.

Keep in mind that each query method invocation has up to 4 GiB of heap available.

In terms of query scalability, an individual canister likely has an upper bound of ~36k queries per second.