jdmichaud

(window as any).console.image = (canvas: HTMLCanvasElement, targetWidth?: number, targetHeight?: number) => {
  const url = canvas.toDataURL();
  const image = new Image();
  image.src = url;

  image.onload = () => {
    // 1. Determine dimensions: Use target resolution if provided,
    // otherwise default to the canvas's own dimensions.
    const w = targetWidth ?? canvas.width;
    const h = targetHeight ?? canvas.height;

jdmichaud

Various solutions: https://wiki.openstreetmap.org/wiki/Street-level_imagery_services

The real open one seems to be panoramax.

panoramax instance: https://wiki.openstreetmap.org/wiki/Panoramax

France/IGN panoramax instance: https://panoramax.ign.fr

Info on out to take pictures: https://forum.geocommuns.fr/t/les-tutoriels-pour-contribuer-a-panoramax/1015

jdmichaud

List devices:

scanimage --list-devices

Get device capability:

scanimage --help --device-name=brother4

Do not use /dev path here but the first part of the device string provided by the list devices command:

jdmichaud

This is a very strong objection — and you’re right to push on it. Let me respond directly to your claim, not by retreating to “nested vectors”, and not by hand-waving about math libraries.

I’ll try to be precise about where matrices really do differ from vectors at the “core representation” level, and where they don’t. The key is that the difference is narrower than people often imply — but it’s still real.

---

1. You’re right about one big thing

“All this applied to vectors too.”

jdmichaud

Metadata

See metadata (and other data about the file):

ffmpeg -i input.mp4

Add metadata to video:

ffmpeg -i input.mp4 -metadata title="My Video Title" \

jdmichaud

const data = await (await fetch('Tv-test-pattern-146649_640.png')).bytes();
const decoder = new ImageDecoder({ type: "image/png", data });
const frame = await decoder.decode();
const bytes = new Uint8Array(frame.image.allocationSize());
await frame.image.copyTo(bytes.buffer);

jdmichaud

#include <sys/mman.h>
#include <string.h>
#include <stdio.h>

int main() {

    /*
        This is raw x86-64 machine code for a TINY *leaf function*:

            lea eax, [rdi + 1]    ; compute (arg + 1)

jdmichaud

This does not currently work.

• Render an image from a Dicom file (the last one in case the DICOM file contain an icon)
• resize the image using the Bicubic interpolation method
• apply the VOI Lut

convert isome.file[-1] -resize 50% -interpolate Bicubic -level 495,210 isome.file.png

jdmichaud

# python3 -mvenv venv && source venv/bin/activate && pip install matplotlib
# python3 make_polygon.py
import matplotlib.pyplot as plt
import math

def make_polygon(vertices):
    # compute centroid
    average_x = sum([vertex[0] for vertex in vertices]) / len(vertices)
    average_y = sum([vertex[1] for vertex in vertices]) / len(vertices)
    centroid = [average_x, average_y]

jdmichaud

-a archives recursively
-P shows progress and enables resuming interrupted download
-a compresses

rsync -azP -e 'ssh -p PORT' <user>@<IP>:/path/to/folder .