pcorpet

dotenv#1079 minimal repro

dotenv 18.0.9 parse used in a browser page under Vite dev, with vite-plugin-node-polyfills (buffer + Buffer global), as in the CO2 AI frontend.

Steps:

1. npm install
2. npm run dev and open the printed URL
3. The page throws Invalid or unexpected token.

What is going on: Vite's dep optimizer bundles dotenv/dist/index.cjs, which begins with #!/usr/bin/env node. The polyfill plugin injects import statements at the top of that bundle, so the shebang ends up on line 5 of an ES module, where it is a syntax error. With dotenv 17.4.2 the shebang is not present in the optimized output.

pcorpet

<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>Imperative vs Declarative Programming</title>
  <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/reveal.js/4.5.0/reveal.min.css">
  <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/reveal.js/4.5.0/theme/black.min.css">
  <style>
    :root {

pcorpet

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>yarn.lock in Git</title>
    <style>
        * {
            margin: 0;
            padding: 0;

pcorpet

#!/usr/bin/python3
# Script to help migrate Pylint config to Ruff.
#
# It lists enabled Pylint rules, Pylint rules that are enabled but already covered by Ruff rules,
# and the Ruff rules not enabled yet that match enabled Pylint rules.
#
# Largely based on https://github.com/astral-sh/ruff/issues/970
import re
import subprocess
from collections.abc import Mapping, Sequence

pcorpet

import libcst as cst
import os


def process_file(file_path):
    # Read the content of the file
    with open(file_path, "r") as file:
        code = file.read()

    # Parse the code file into a CST

pcorpet

"growth20122022": 0.11912225705329153,
"hasFreelancers": false,
"hollandCodeMajor": "S",
"hollandCodeMinor": "C",
"inAWorkplace": "dans une structure d'accompagnement",
"inDomain": "dans le social",
"is_diploma_strictly_required": true,
"jobBoards": [
 {
  "filters": [

pcorpet

// How one would talk about various companies in this job group:
// In French: "chez Axa, Groupama ou ...", "à la BNP, au Crédit agricole".
// This data was manually built by Bayes Impact's Content Team over the years.
string at_various_companies = 24;

// Feminine version of what_i_love_about field.
// This data was manually built by Bayes Impact's Content Team over the years.
string what_i_love_about_feminine = 25;

// The (at most 11) top departements in terms of offer/demand ratio, in descending order.

pcorpet

<WithNote
  hasComment={true}
  note={<React.Fragment>
    {userYou('Tu ne trouves pas ton', 'Vous ne trouvez pas votre')} métier&nbsp;?
    <ExternalLink style={{color: colors.BOB_BLUE, fontSize: 15, marginLeft: 3}}
      href="https://airtable.com/shreUw3GYqAwVAA27">
      Clique{userYou('', 'z')} ici pour l'ajouter
    </ExternalLink>
  </React.Fragment>}>
  <FieldSet

pcorpet

import hashlib
import sys


def hash_lines(inputfile, outputfile):
    """Hash all lines of the input file and populate the output."""
    count = 0
    with open(outputfile, 'wt') as output:
        with open(inputfile, 'r') as input_lines:
            for line in input_lines:

pcorpet

<html>
  <head>
    <script src="https://cdn.rawgit.com/emn178/js-sha1/c724643d/src/sha1.js"></script>
    <script>
      function hashIt() {
        var content = document.getElementById('content').value
        document.getElementById('result').innerText =
          content.split('\n').map(sha1).join('\n')
      }
    </script>