aldric

/**
 * ARCHITECTURAL BLUEPRINT: Cloudflare-Native RAG System
 * Use Case: Real-time compliance/manual lookup for 12k+ affiliate sites.
 */

import { Ai } from '@cloudflare/ai'

export interface Env {
  AI: any;
  VECTOR_INDEX: VectorizeIndex;

aldric

ADR-000: Architectural Decision Record (ADR) Manifesto

Status: Permanent | Architect: Aldric Gaudinot

Context

In ecosystems managing 12,000+ web properties (Blue Window) and €300M+ in financial transactions, "tribal knowledge" is a systemic risk. To ensure platform resilience and PCI-DSS compliance, every critical architectural shift must be codified.

The Decision

We mandate the use of ADRs for all platform-level changes. An ADR is not just a document; it is a historical record of intent that prevents "architectural drift" and ensures long-term maintainability.

Implementation Standard

aldric

curl -X POST "https://api.cloudflare.com/client/v4/zones/{ZONE_ID}/logpush/jobs" \
     -H "Authorization: Bearer {API_TOKEN}" \
     -H "Content-Type: application/json" \
     --data '{
  "name": "cloudflare-to-elastic",
  "destination_conf": "https://<ELASTIC_AGENT_HOST>:5044/cloudflare.logs?header=Authorization&value=Bearer%20{KEY}",
  "dataset": "http_requests",
  "logpull_options": "fields=ClientIP,ClientRequestHost,ClientRequestMethod,EdgeStartTimestamp,EdgeResponseStatus,WAFAction,BotScore",
  "enabled": true
}'

aldric

/*
The Vary: Accept Header: This is the secret sauce. It ensures that when a Chrome user (AVIF) and a legacy browser user (PNG) hit the same URL, Cloudflare's CDN caches and serves the correct version to each, without re-running the Worker.

Binding over Fetch: By using env.IMAGES_ENGINE instead of fetch('/cdn-cgi/image/...'), you avoid the "sub-request limit" and get significantly lower internal latency within the Cloudflare network.

Metadata Stripping: Affiliate banners often contain sensitive "Created By" or "GPS" metadata from design tools. metadata: 'none' automatically scrubs this while saving extra kilobytes.
*/
import { Hono } from 'hon o'

type Bindings = {

aldric

#!/usr/bin/env python
# ============================================================================
# Apache Beam - Batch Processing Pipeline
# Process files from Google Cloud Storage → Apache Beam → BigQuery
# 
# Features:
# - Read Avro/JSON files from GCS
# - Data validation and transformation
# - Deduplication by event_id
# - Write to BigQuery with WRITE_APPEND mode

aldric

# ============================================================================
# CLOUDFLARE ZERO TRUST - PRODUCTION CONFIGURATION
# Complete setup: SAML + Device Posture + Malware Blocking + Audit Logging
# ============================================================================

terraform {
  required_version = ">= 1.0"
  required_providers {
    cloudflare = {
      source  = "cloudflare/cloudflare"

aldric

    const filteredEdges = edges.reduce((acc, current) => {
      const x = acc.find(item => item.source === current.source && item.target === current.target);
      if (!x) {
        return acc.concat([current]);
      } else {
        return acc;
      }
    }, []);
	
	

aldric

#!/bin/bash

# Sample bash script to send any file (here server access logs), filter the content and send it to a ftp server

FTP_SERVER=''
FTP_USER=''
FTP_PASS=''
FTP_FOLDER=''

# Find the previous *complete* access file.

aldric

  import * as firebase from 'firebase/app';
  import campaigns from './reference-data/data.json';

  public batchUpdateby500() {
      const camps = campaigns
      .sort((a, b) => a.id - b.id)
      .map((c) => {
        const channel = 'channels';
        const brand = 'brands';

aldric

import * as firebase from 'firebase/app';

  public updateCountries() {
    this.getCountries().subscribe((countries) => {
      const batch = this.firestore.firestore.batch();
      countries.forEach(element => {
        const ref = this.firestore
          .collection('at-settings')
          .doc('countries')
          .collection('values')