jcuffe

param(
    [string]$inputFilePath,
    [string]$configFilePath
)

if (-not $inputFilePath) {
    Write-Host "Please provide a valid input file path"
    exit
}

jcuffe

import { __testVisible as self } from './index.mjs';

function a() {
  return 1;
}

export function b() {
  return self.a();
}

jcuffe

It's never been easier to elevate your latest project from your laptop to the Internet, allowing you to show your proof of concept to anyone with a simple url. However, some pieces of the puzzle have remained frustratingly difficult to place without relying on a paid service. For me, that piece was the persistence layer. I've always been a big fan of SQL, and Postgres in particular. Simple enough for the simplest MVP, and reliably consistent enough for business-critical data, it's a piece of tech I learned once and have used in every project since. Unfortunately, free database offerings are far less common than free static file hosts.

jcuffe

const pq = [];

function enqueue(value, priority) {
  const item = { value, priority };
  for (let i = 0; i < pq.length; i++) {
    const { value: pqValue, priority: pqPriority } = pq[i];
    // Maintain min-heap, sorted by (value, priority)
    if (value < pqValue || (value === pqValue && priority < pqPriority)) {
      pq.splice(i, 0, item);
      return;

jcuffe

const values = [5, 4, 4, 4, 4, 0, 0, 0, -1, -1, -1, -3];
const k = 20;

function main() {
  // Build a lookup of value frequencies, and calculate the best combo
  let bestCombo = 0;
  const frequencies = {};
  for (const value of values) {
    const absValue = Math.abs(value);
    if (!frequencies[absValue]) {

jcuffe

import { useStore, component$ } from '@builder.io/qwik';
import axios from 'axios';

export const Main = component$(async () => {
  const response = await axios.get('https://httpbin.org/')
  const state = useStore({ name: 'World', status: response.status });
  return (
    <p>Status: {state.status}</p>
  );
});

jcuffe

The various search patterns afforded to customers make it challenging to generally optimize the query performed by the /admin/record endpoint.

This gist explores several strategies that offer significant improvement for specific query patterns.

jcuffe

// New cursor format
const cursor = (record, columnName) => base64.encode(JSON.stringify({
  name: record.constructor.name,
  columnName,
  value: record[columnName],
  recordId: record.id
}))

// Paginate based on (potentially not unique) cursor column value, as well as (unique) record id
if (connection?.before) {

jcuffe

ARGS {
  input: [Object: null prototype] {
    eventId: '969356d5-6156-4064-a754-b4d6399186de',
    userId: 'auth0|609c4d9117bcf2007065def6',
    role: 'creator'
  }
}
HOST ROLE UPDATE
DEMOTED CREATOR auth0|5d9cf8731aa11a0c5bac758e
PROMOTED COHOST auth0|609c4d9117bcf2007065def6

jcuffe

// Query builders are not actually promises, but have a `.then()` method which matches the promise interface.
// As soon as you `await` or `.then()` the builder, the query is executed, and a new promise is generated that will contain the result.
const getQueryBuilder = () => ({
  then: (resolve) =>
    resolve(
      new Promise((resolve) =>
        setTimeout(() => resolve(["query", "results"]), 1000),
      ),
    ),
})