Emmanuel S. esgn

How to push a Docker image to Google Artifact Registry with a GitHub Action without using a Service Account

This how I managed to push a Docker image in a Google Cloud Artifact Registry using a GitHub Actions workflow without using a GCP Service Account.

Before using this tutorial make sure you have :

A GitHub repository containing a Dockerfile
A GCP project with sufficient permissions to make changes to it
The gcloud tool installed, authenticated and configured to use the GCP project. We will not use the --project flag in the following gcloud commands.

Create an Artifact Registry of Docker type

Extraire une zone du LiDAR HD à distance avec PDAL et PDAL Wrench

À partir d’une emprise exprimée en Lambert-93, ce tutoriel montre comment extraire depuis la Géoplateforme uniquement les points LiDAR HD nécessaires, même lorsque la zone couvre plusieurs dalles ou plusieurs blocs, sans télécharger intégralement les fichiers COPC sources en local.

Trouver les dalles LiDAR HD concernées sur la zone

Passage par un VPC (Virtual Point Cloud)

Le service de téléchargement de la Géoplateforme propose des fichiers VPC pour chaque bloc de LiDAR HD. Exemple : https://data.geopf.fr/chunk/telechargement/download/LiDARHD-NUALID/NUALHD_1-0__LAZ_LAMB93_PQ_2025-03-14/index.vpc

How to use .gitconfig files to handle multiple identities / git providers (github, gitlab, ...)

In the user home, define a parent .gitconfig file for your main account (eg: work account)

[user]
        name = John Doe
        email = john.doe@work.com
[credential]
        helper = store --file /home/johndoe/.git-credentials

	import sys
	import argparse
	import geopandas as gpd
	import pandas as pd
	import json

	def parse_args():
	parser = argparse.ArgumentParser(
	"Create JSON PDAL pipelines to download point cloud extract")
	parser.add_argument("--input_footprints", "-i",

	resource "google_storage_bucket" "osm_dl_planet" {
	project = var.project-id
	name = var.bucket-name
	uniform_bucket_level_access = true
	location = var.project-region
	storage_class = "REGIONAL"
	force_destroy = true
	}

	resource "google_service_account" "scheduler" {

	#!/usr/bin/env bash

	# Working area in EPSG:2154
	xmin=892000
	ymin=6249000
	xmax=894000
	ymax=6247000

	rm -rf footprints
	mkdir footprints

	#!/usr/bin/env python3

	import os
	import laspy
	import shutil

	file_ext = ".las"
	output_file = "output.las"

	if os.path.exists(output_file):

	import sys

	try:
	from osgeo import ogr, osr, gdal
	except:
	sys.exit('ERROR: cannot find GDAL/OGR modules')

	# create the spatial reference
	srs = osr.SpatialReference()
	srs.ImportFromEPSG(2154)

	#include <CGAL/Exact_predicates_inexact_constructions_kernel.h>
	#include <CGAL/IO/read_points.h>
	#include <CGAL/property_map.h>
	#include <CGAL/Surface_mesh.h>
	#include <CGAL/Shape_detection/Efficient_RANSAC.h>
	#include <CGAL/Polygonal_surface_reconstruction.h>
	#include <CGAL/Polygon_mesh_processing/triangulate_faces.h>
	#include <CGAL/Polygon_mesh_processing/orientation.h>
	#include <CGAL/Polygon_mesh_processing/polygon_mesh_to_polygon_soup.h>
	#include <CGAL/Polygon_mesh_processing/polygon_soup_to_polygon_mesh.h>