Testing PREG_NO_UTF8_CHECK

batchwords.sh

#!/bin/bash

if [[ ! -d "$1" ]]; then
  echo "Not a directory: $1"
  exit 1
fi

find "$1" -type f | while read name
do
  ./php-src/sapi/cli/php ./words.php "$2" < "$name" >/dev/null
  if [[ $? != 0 ]]; then
    echo "Crashed on: $name"
    exit 255
  fi
done

echo "Great success!"
exit 0

results.sh

$ ./php-src/sapi/cli/php ./words.php < ./php-src/sapi/cli/php > ./words.txt
$ tail -5 ./words.txt

===============================================================================
DONE, 326 errors
> PREG_INTERNAL_ERROR (326)
===============================================================================

$ ./php-src/sapi/cli/php ./words.php '(\p{L})(\1)+' < ./php-src/sapi/cli/php > ./words.txt
$ tail -4 ./words.txt

===============================================================================
DONE, no errors
===============================================================================

words.php

<?php

(new Program($argv))->execute();

class Program
{
  private $errors = [
    PREG_NO_ERROR => ['name' => "PREG_NO_ERROR", 'count' => 0],
    PREG_INTERNAL_ERROR => ['name' => "PREG_INTERNAL_ERROR", 'count' => 0],
    PREG_BACKTRACK_LIMIT_ERROR => ['name' => "PREG_BACKTRACK_LIMIT_ERROR", 'count' => 0],
    PREG_RECURSION_LIMIT_ERROR => ['name' => "PREG_RECURSION_LIMIT_ERROR", 'count' => 0],
    PREG_BAD_UTF8_ERROR => ['name' => "PREG_BAD_UTF8_ERROR", 'count' => 0],
    PREG_BAD_UTF8_OFFSET_ERROR => ['name' => "PREG_BAD_UTF8_OFFSET_ERROR", 'count' => 0],
  ];

  private $errcount = 0;

  public function __construct(array $argv)
  {
    $pattern = empty($argv[1]) ? '\w+' : $argv[1];
    $this->pattern = sprintf('/%s/u', $pattern);
  }

  public function execute()
  {
    $this->printHeader();

    do {
      $str = fread(STDIN, 4096);
      $res = preg_match_all($this->pattern, $str, $matches, PREG_NO_UTF8_CHECK);
      if ($res === false) {
        $this->errcount++;
        $this->errors[preg_last_error()]['count']++;
      } elseif ($res) {
        echo implode(PHP_EOL, $matches[0]), PHP_EOL;
      }
    } while(!feof(STDIN));

    $this->printFooter();
  }

  private function printHeader()
  {
    echo 'Searching for: ', $this->pattern, PHP_EOL;
    echo str_repeat('=', 79), PHP_EOL;
  }

  private function printFooter()
  {
    echo str_repeat(PHP_EOL, 3);
    echo str_repeat('=', 79), PHP_EOL;
    echo 'DONE, ', $this->errcount ?: 'no', ' errors', PHP_EOL;
    if ($this->errcount) {
      $errors = array_filter($this->errors, function ($info) { return $info['count']; });
      $errors = array_map(function ($info) {
        return sprintf('%s (%d)', $info['name'], $info['count']);
      }, $errors);
      echo implode(PHP_EOL, $errors), PHP_EOL;
    }
    echo str_repeat('=', 79), PHP_EOL;
  }
}