280 lines
No EOL
20 KiB
HTML
280 lines
No EOL
20 KiB
HTML
<?xml version="1.0" encoding="utf-8"?>
|
||
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
|
||
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
|
||
<html xmlns="http://www.w3.org/1999/xhtml" lang="fr" xml:lang="fr">
|
||
<head>
|
||
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
|
||
|
||
|
||
<meta name="keywords" content="arbre, HTML, script, ruby">
|
||
|
||
<link rel="shortcut icon" type="image/x-icon" href="/Scratch/img/favicon.ico" />
|
||
<link rel="stylesheet" type="text/css" href="/Scratch/assets/css/main.css" />
|
||
<link rel="stylesheet" type="text/css" href="/Scratch/css/twilight.css" />
|
||
<link rel="stylesheet" type="text/css" href="/Scratch/css/idc.css" />
|
||
<link rel="alternate" type="application/rss+xml" title="RSS" href="http://feeds.feedburner.com/yannespositocomfr"/>
|
||
|
||
<link rel="alternate" lang="fr" xml:lang="fr" title="Comment réparer un XML coupé ?" type="text/html" hreflang="fr" href="/Scratch/fr/blog/2010-05-19-How-to-cut-HTML-and-repair-it/" />
|
||
<link rel="alternate" lang="en" xml:lang="en" title="How to repair a cutted XML?" type="text/html" hreflang="en" href="/Scratch/en/blog/2010-05-19-How-to-cut-HTML-and-repair-it/" />
|
||
<script type="text/javascript" src="/Scratch/js/jquery-1.3.1.min.js"></script>
|
||
<script type="text/javascript" src="/Scratch/js/jquery.cookie.js"></script>
|
||
<script type="text/javascript" src="/Scratch/js/index.js"></script>
|
||
<!--[if lt IE 9]>
|
||
<script src="http://ie7-js.googlecode.com/svn/version/2.1(beta4)/IE9.js"></script>
|
||
<![endif]-->
|
||
<title>Comment réparer un XML coupé ?</title>
|
||
</head>
|
||
<body lang="fr" class="article">
|
||
<script type="text/javascript">// <![CDATA[
|
||
document.write('<div id="blackpage"><img src="/Scratch/img/loading.gif" alt="Chargement en cours..."/></div>');
|
||
// ]]>
|
||
</script>
|
||
|
||
<div id="content">
|
||
|
||
<div id="choix">
|
||
<div class="return"><a href="#entete">↓ Menu ↓</a></div>
|
||
<div id="choixlang">
|
||
<a href="/Scratch/en/blog/2010-05-19-How-to-cut-HTML-and-repair-it/" onclick="setLanguage('en')">in English</a>
|
||
</div>
|
||
<div class="flush"></div>
|
||
</div>
|
||
<div id="titre">
|
||
<h1>
|
||
Comment réparer un XML coupé ?
|
||
</h1>
|
||
|
||
<h2>
|
||
et comment s'en sortir sans parseur ?
|
||
</h2>
|
||
|
||
</div>
|
||
<div class="flush"></div>
|
||
|
||
|
||
|
||
|
||
|
||
<div class="flush"></div>
|
||
<div id="afterheader">
|
||
<div class="corps">
|
||
<p>Sur ma page d’accueil vous pouvez voir la liste des mes derniers articles avec le début de ceux-ci. Pour arriver à faire ça, j’ai besoin de couper le code XHTML de mes pages en plein milieu. Il m’a donc fallu trouver un moyen de les réparer.</p>
|
||
|
||
<p>Prenons un exemple :</p>
|
||
|
||
<pre class="twilight">
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">div</span> <span class="MetaTagAll">class</span>=<span class="String"><span class="String">"</span>corps<span class="String">"</span></span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">div</span> <span class="MetaTagAll">class</span>=<span class="String"><span class="String">"</span>intro<span class="String">"</span></span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>Introduction<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">div</span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>The first paragraph<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagInline"><span class="MetaTagInline"><</span><span class="MetaTagInline">img</span> <span class="MetaTagInline">src</span>=<span class="String"><span class="String">"</span>/img/img.png<span class="String">"</span></span> <span class="MetaTagInline">alt</span>=<span class="String"><span class="String">"</span>an image<span class="String">"</span></span>/<span class="MetaTagInline">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>Another long paragraph<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">div</span><span class="MetaTagAll">></span></span>
|
||
</pre>
|
||
|
||
<p>Après avoir coupé, j’obtiens :</p>
|
||
|
||
<pre class="twilight">
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">div</span> <span class="MetaTagAll">class</span>=<span class="String"><span class="String">"</span>corps<span class="String">"</span></span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">div</span> <span class="MetaTagAll">class</span>=<span class="String"><span class="String">"</span>intro<span class="String">"</span></span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>Introduction<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">div</span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>The first paragraph<span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
<span class="MetaTagInline"><span class="MetaTagInline"><</span><span class="MetaTagInline">img</span> <span class="MetaTagInline">src</span>=<span class="String"><span class="String">"</span>/img/im</span></span>
|
||
</pre>
|
||
|
||
<p>En plein milieu d’un tag <code><img></code> !</p>
|
||
|
||
<p>En réalité, ce n’est pas si difficile que celà peut paraître au premier abord. Le secret réside dans le fait de comprendre que l’on n’a pas besoin de conserver la structure complète de l’arbre pour le réparer, mais seulement la liste des parents non fermés.</p>
|
||
|
||
<p>Pour notre exemple, juste après le paragraphe <code>first paragraph</code> nous n’avons qu’à fermer un <code>div</code> pour la classe <code>corps</code> et le XML est réparé. Bien entendu, quand on est dans le cas où un tag est coupé au milieu, on a qu’à remonté juste avant le début de ce tag corrompu.</p>
|
||
|
||
<p>Donc, tout ce que nous avons à faire, c’est d’enregistrer la liste des parents dans une pile. Supposons que nous traitions le premier exemple complètement. La pile passera par les états suivants :</p>
|
||
|
||
<pre class="twilight">
|
||
[]
|
||
[div] <span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">div</span> <span class="MetaTagAll">class</span>=<span class="String"><span class="String">"</span>corps<span class="String">"</span></span><span class="MetaTagAll">></span></span>
|
||
[div, div] <span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">div</span> <span class="MetaTagAll">class</span>=<span class="String"><span class="String">"</span>intro<span class="String">"</span></span><span class="MetaTagAll">></span></span>
|
||
[div, div, p] <span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
Introduction
|
||
[div, div] <span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
[div] <span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">div</span><span class="MetaTagAll">></span></span>
|
||
[div, p] <span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
The first paragraph
|
||
[div] <span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
[div] <span class="MetaTagInline"><span class="MetaTagInline"><</span><span class="MetaTagInline">img</span> <span class="MetaTagInline">src</span>=<span class="String"><span class="String">"</span>/img/img.png<span class="String">"</span></span> <span class="MetaTagInline">alt</span>=<span class="String"><span class="String">"</span>an image<span class="String">"</span></span>/<span class="MetaTagInline">></span></span>
|
||
[div, p] <span class="MetaTagAll"><span class="MetaTagAll"><</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
Another long paragraph
|
||
[div] <span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">p</span><span class="MetaTagAll">></span></span>
|
||
[] <span class="MetaTagAll"><span class="MetaTagAll"></</span><span class="MetaTagAll">div</span><span class="MetaTagAll">></span></span>
|
||
</pre>
|
||
|
||
<p>L’algorithme est alors très simple :
|
||
<pre class="twilight">
|
||
let res be the XML as a string ;
|
||
read res and each time you encouter a tag:
|
||
if it is an opening one:
|
||
push it to the stack
|
||
else if it is a closing one:
|
||
pop the stack.</p>
|
||
|
||
<p>remove any malformed/cutted tag in the end of res
|
||
for each tag in the stack, pop it, and write:
|
||
res = res + closed tag</p>
|
||
|
||
<p>return res
|
||
</pre></p>
|
||
|
||
<p>Et <code>res</code> contiend le XML réparé.</p>
|
||
|
||
<p>Finallement, voici le code en ruby que j’utilise. La variable <code>xml</code> contient le XML coupé.</p>
|
||
|
||
<div class="code"><div class="file"><a href="/Scratch/fr/blog/2010-05-19-How-to-cut-HTML-and-repair-it/code/repair_xml.rb"> ➥ repair_xml.rb </a></div><div class="withfile">
|
||
<pre class="twilight">
|
||
<span class="Comment"><span class="Comment">#</span> repair cutted XML code by closing the tags</span>
|
||
<span class="Comment"><span class="Comment">#</span> work even if the XML is cut into a tag.</span>
|
||
<span class="Comment"><span class="Comment">#</span> example:</span>
|
||
<span class="Comment"><span class="Comment">#</span> transform '<div> <span> toto </span> <p> hello <a href="http://tur'</span>
|
||
<span class="Comment"><span class="Comment">#</span> into '<div> <span> toto </span> <p> hello </p></div>'</span>
|
||
<span class="Keyword">def</span> <span class="Entity">repair_xml</span>(<span class="Variable"> xml </span>)
|
||
parents<span class="Keyword">=</span>[]
|
||
depth<span class="Keyword">=</span><span class="Constant">0</span>
|
||
xml.<span class="Entity">scan</span>( <span class="StringRegexp"><span class="StringRegexp">%r{</span><<span class="StringRegexp"><span class="StringRegexp">(</span>/?<span class="StringRegexp">)</span></span><span class="StringRegexp"><span class="StringRegexp">(</span><span class="StringRegexpSpecial">\w</span>*<span class="StringRegexp">)</span></span><span class="StringRegexp"><span class="StringRegexp">[</span>^><span class="StringRegexp">]</span></span>*<span class="StringRegexp"><span class="StringRegexp">(</span>/?<span class="StringRegexp">)</span></span>><span class="StringRegexp">}</span></span> ).<span class="Entity">each</span> <span class="Keyword">do </span>|<span class="Variable">m</span>|
|
||
<span class="Keyword">if</span> m[<span class="Constant">2</span>] <span class="Keyword">==</span> <span class="String"><span class="String">"</span>/<span class="String">"</span></span>
|
||
<span class="Keyword">next</span>
|
||
<span class="Keyword">end</span>
|
||
<span class="Keyword">if</span> m[<span class="Constant">0</span>] <span class="Keyword">==</span> <span class="String"><span class="String">"</span><span class="String">"</span></span>
|
||
parents[depth]<span class="Keyword">=</span>m[<span class="Constant">1</span>]
|
||
depth<span class="Keyword">+=</span><span class="Constant">1</span>
|
||
<span class="Keyword">else</span>
|
||
depth<span class="Keyword">-=</span><span class="Constant">1</span>
|
||
<span class="Keyword">end</span>
|
||
<span class="Keyword">end</span>
|
||
res<span class="Keyword">=</span>xml.<span class="Entity">sub</span>(<span class="StringRegexp"><span class="StringRegexp">/</span></span><span class="StringRegexp"><<span class="StringRegexp"><span class="StringRegexp">[</span>^><span class="StringRegexp">]</span></span>*$</span><span class="StringRegexp"><span class="StringRegexp">/m</span></span>,<span class="String"><span class="String">'</span><span class="String">'</span></span>)
|
||
depth<span class="Keyword">-=</span><span class="Constant">1</span>
|
||
depth.<span class="Entity">downto</span>(<span class="Constant">0</span>).<span class="Entity">each</span> { |<span class="Variable">x</span>| res<span class="Keyword"><<=</span> <span class="String"><span class="String">%{</span></<span class="StringEmbeddedSource"><span class="StringEmbeddedSource">#{</span>parents<span class="StringEmbeddedSource">[</span>x<span class="StringEmbeddedSource">]</span><span class="StringEmbeddedSource">}</span></span>><span class="String">}</span></span> }
|
||
res
|
||
<span class="Keyword">end</span>
|
||
</pre>
|
||
</div></div>
|
||
|
||
<p>Je ne sais pas si ce code pourra vous être utile. Par contre le raisonnement pour y parvenir mérite d’être connu.</p>
|
||
|
||
</div>
|
||
|
||
|
||
|
||
<div id="choixrss">
|
||
<a id="rss" href="http://feeds.feedburner.com/yannespositocomfr">
|
||
s'abonner
|
||
</a>
|
||
</div>
|
||
<script type="text/javascript">
|
||
$(document).ready(function(){
|
||
$('#comment').hide();
|
||
$('#clickcomment').click(showComments);
|
||
});
|
||
function showComments() {
|
||
$('#comment').show();
|
||
$('#clickcomment').fadeOut();
|
||
}
|
||
document.write('<div id="clickcomment">Commentaires</div>');
|
||
</script>
|
||
<div class="flush"></div>
|
||
<div class="corps" id="comment">
|
||
<h2 class="first">commentaires</h2>
|
||
<noscript>
|
||
Vous devez activer javascript pour commenter.
|
||
</noscript>
|
||
|
||
<script type="text/javascript">
|
||
var idcomments_acct = 'a307f0044511ff1b5cfca573fc0a52e7';
|
||
var idcomments_post_id = '/Scratch/fr/blog/2010-05-19-How-to-cut-HTML-and-repair-it/';
|
||
var idcomments_post_url = 'http://yannesposito.com/Scratch/fr/blog/2010-05-19-How-to-cut-HTML-and-repair-it/';
|
||
</script>
|
||
<span id="IDCommentsPostTitle" style="display:none"></span>
|
||
<script type='text/javascript' src='/Scratch/js/genericCommentWrapperV2.js'></script>
|
||
|
||
</div>
|
||
|
||
<div id="entete" class="corps_spaced">
|
||
<div id="liens">
|
||
<ul><li><a href="/Scratch/fr/">Bienvenue</a></li>
|
||
<li><a href="/Scratch/fr/blog/">Blog</a></li>
|
||
<li><a href="/Scratch/fr/softwares/">Softwares</a></li>
|
||
<li><a href="/Scratch/fr/about/">À propos</a></li></ul>
|
||
</div>
|
||
<div class="flush"></div>
|
||
<hr/>
|
||
<div id="next_before_articles">
|
||
<div id="previous_articles">
|
||
articles précédents
|
||
|
||
<div class="previous_article">
|
||
<a href="/Scratch/fr/blog/2010-05-17-at-least-this-blog-revive/"><span class="nicer">«</span> Je reviens à la vie !</a>
|
||
</div>
|
||
|
||
|
||
<div class="previous_article">
|
||
<a href="/Scratch/fr/blog/2010-03-23-Encapsulate-git/"><span class="nicer">«</span> Encapsuler git</a>
|
||
</div>
|
||
|
||
|
||
<div class="previous_article">
|
||
<a href="/Scratch/fr/blog/2010-03-22-Git-Tips/"><span class="nicer">«</span> Astuces Git</a>
|
||
</div>
|
||
|
||
|
||
</div>
|
||
<div id="next_articles">
|
||
articles suivants
|
||
|
||
<div class="next_article">
|
||
<a href="/Scratch/fr/blog/2010-05-24-Trees--Pragmatism-and-Formalism/">Arbres ; Pragmatisme et Formalisme <span class="nicer">»</span></a>
|
||
</div>
|
||
|
||
|
||
<div class="next_article">
|
||
<a href="/Scratch/fr/blog/2010-06-14-multi-language-choices/">choix liés à l'écriture dans plusieurs langues <span class="nicer">»</span></a>
|
||
</div>
|
||
|
||
|
||
<div class="next_article">
|
||
<a href="/Scratch/fr/blog/2010-06-15-Get-my-blog-engine/">Récupérez mon système de blog <span class="nicer">»</span></a>
|
||
</div>
|
||
|
||
|
||
</div>
|
||
<div class="flush"></div>
|
||
</div>
|
||
</div>
|
||
|
||
|
||
<div id="bottom">
|
||
<div>
|
||
<a rel="license" href="http://creativecommons.org/licenses/by-sa/3.0/deed.fr">Droits de reproduction ©, Yann Esposito</a>
|
||
</div>
|
||
<div id="lastmod">
|
||
Écrit le : 19/05/2010
|
||
modifié le : 04/10/2010
|
||
</div>
|
||
<div>
|
||
Site entièrement réalisé avec
|
||
<a href="http://www.vim.org">Vim</a>
|
||
et
|
||
<a href="http://nanoc.stoneship.org">nanoc</a>
|
||
</div>
|
||
<div>
|
||
<a href="/Scratch/fr/validation/">Validation</a>
|
||
<a href="http://validator.w3.org/check?uri=referer"> [xhtml] </a>
|
||
.
|
||
<a href="http://jigsaw.w3.org/css-validator/check/referer?profile=css3"> [css] </a>
|
||
.
|
||
<a href="http://validator.w3.org/feed/check.cgi?url=http%3A//yannesposito.com/Scratch/fr/blog/feed/feed.xml">[rss]</a>
|
||
</div>
|
||
</div>
|
||
<div class="clear"></div>
|
||
</div>
|
||
</body>
|
||
</html> |