extranet/scripts/jobs/getBodaccPdf.php

<?php
/**
 * Download bodacc pdf file
 */
if ($argc != 4){
	echo $argv[0].' type annee num';
	echo "\n";
	exit;
}

$type = strtoupper($argv[1]);
$annee = $argv[2];

$num = $argv[3];
$c = strlen($num);
for($i=0;$i<4-$c;$i++){
	$num = '0'.$num;
}
$annonce = 1;

function getPage($url, $curl_data = '', $verbose=false)
{
	global $ckfile;

	//$user_agent = 'Mozilla/5.0 (Windows NT 5.1; rv:9.0.1) Gecko/20100101 Firefox/9.0.1';
	//$user_agent = 'Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)';
	$user_agent = 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0)';

	$post = false;
	$fields = '';
	if (is_array($curl_data) && count($curl_data)>0) {
		foreach($curl_data as $key=>$value) {
			$fields .= $key.'='.$value.'&';
		}
		rtrim($fields,'&');
		$post = true;
	}

	$options = array(
	CURLOPT_RETURNTRANSFER 	=> true,         // return web page
	CURLOPT_HEADER         	=> false,        // don't return headers
	CURLOPT_FOLLOWLOCATION 	=> true,        // follow redirects
	CURLOPT_ENCODING       	=> "",           // handle all encodings
	CURLOPT_USERAGENT      	=> $user_agent,  // who am i
	CURLOPT_AUTOREFERER    	=> true,         // set referer on redirect
	CURLOPT_CONNECTTIMEOUT 	=> 120,          // timeout on connect
	CURLOPT_TIMEOUT        	=> 120,          // timeout on response
	CURLOPT_MAXREDIRS      	=> 10,           // stop after 10 redirects
	CURLOPT_POST           	=> $post,        // i am sending post data
	CURLOPT_POSTFIELDS     	=> $fields,   	 // this are my post vars
	CURLOPT_SSL_VERIFYHOST 	=> 0,            // don't verify ssl
	CURLOPT_SSL_VERIFYPEER 	=> false,        //
	CURLOPT_VERBOSE        	=> $verbose ,    //
	//CURLOPT_COOKIESESSION 	=> true,
	CURLOPT_COOKIEFILE 		=> $ckfile,
	CURLOPT_COOKIEJAR 		=> $ckfile, 	// Stockage du cookie de session
	);

	$ch      = curl_init($url);
	curl_setopt_array($ch,$options);
	$content = curl_exec($ch);
	$err     = curl_errno($ch);
	$errmsg  = curl_error($ch) ;
	$header  = curl_getinfo($ch);
	curl_close($ch);

	//  $header['errno']   = $err;
	//  $header['errmsg']  = $errmsg;
	//  $header['content'] = $content;
	return array('header'=>$header, 'content'=>$content);
}

//==> Start
$debug = false;

$config = parse_ini_file(dirname(__FILE__).'/../../application/configs/application.ini', true);
$directory = $config['production']['profil.path.data'].'/bodacc/'.$type.'/'.$annee;
if ( !file_exists($directory) ) mkdir($directory, 0777, true);

//Define cookie file for storage
$ckfile = dirname(__FILE__).'/'.uniqid('cookies-');

$site = 'http://www.bodacc.fr';

//Recherche avancée
echo $url = $site.'/annonce/rechercheavancee';
echo "\n";
$result = getPage($url, $data);
if ($debug) file_put_contents('bodacc1.html', $result['content']);

echo $url = $site."/annonce/liste";
echo "\n";
$data = array(
    'categorieannonce' => 'tout',
    'commercant' => '',
    'datepublicationmax' => '',
    'datepublicationmin' => '',
    'motscles' => '',
    'numeroannonce' => $num,
    'numerodepartement' => 'tout',
    'numeroparution' =>	$annee.$num,
    'publication' => 'A',
    'registre' => '',
    'typeannonce' => 'tout',
);
$result = getPage($url, $data);
if ($debug) file_put_contents('bodacc2.html', $result['content']);
preg_match('/\<a href="\/annonce\/detail\/(.*)"\>/', $result['content'], $matches);

echo $url = $site.'/annonce/detail/'.$matches[1];
echo "\n";
if (empty($url)) {
    echo "Erreur téléchargement du PDF.";
    echo "\n";
    exit;
}

//Affichage de l'annonce
$result = getPage($url);
if ($debug) file_put_contents('bodacc3.html', $result['content']);
/*
<p>
<a href="/annonce/telecharger/EBODACC-A_20130013_0001_p000.pdf" title="Téléchargez le témoin de publication de l'annonce n°1 du Bodacc A n°20130013 du 18/01/2013 - Format pdf ">
<img src="/extension/dilabodacc/design/dilabodacc/images/deco/temoin.png" alt="" class="temoin">
Téléchargez le témoin de publication
</a>
</p>
*/
preg_match('/\<a href="\/annonce\/telecharger\/(.*)"/', $result['content'], $matches);

echo $url = $site.'/annonce/telecharger/'.$matches[1];
echo "\n";
if (empty($url)) {
    echo "Erreur téléchargement du PDF.";
    echo "\n";
    exit;
}

//Téléchargement du PDF
$result = getPage($url);
if (substr($result['content'],0,4)!='%PDF'){
	if ($debug) file_put_contents('bodacc4.html', $result['content']);
	echo "Erreur téléchargement du PDF.";
	echo "\n";
	exit;
}
file_put_contents($directory."/BODACC_".$type."_".$annee."_".$num.".pdf", $result['content']);
echo "\n";
if (file_exists($ckfile)) unlink($ckfile);