xml.html revision 5173727616a7e9ef8e6a2d3c256685413593b1bc
1<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" 2 "http://www.w3.org/TR/html4/loose.dtd"> 3<html> 4<head> 5 <title>The XML C library for Gnome</title> 6 <meta name="GENERATOR" content="amaya V5.0"> 7 <meta http-equiv="Content-Type" content="text/html"> 8</head> 9 10<body bgcolor="#ffffff"> 11<h1 align="center">The XML C library for Gnome</h1> 12 13<h1>Note: this is the flat content of the <a href="index.html">web 14site</a></h1> 15 16<h1 style="text-align: center">libxml, a.k.a. gnome-xml</h1> 17 18<p></p> 19 20<p>Libxml is the XML C library developped for the Gnome project. XML itself 21is a metalanguage to design markup languages, i.e. text language where 22semantic and structure are added to the content using extra "markup" 23information enclosed between angle bracket. HTML is the most well-known 24markup language.</p> 25 26<p>Libxml2 implements a number of existing standards related to markup 27languages:</p> 28<ul> 29 <li>the XML standard: <a 30 href="http://www.w3.org/TR/REC-xml">http://www.w3.org/TR/REC-xml</a></li> 31 <li>Namespaces in XML: <a 32 href="http://www.w3.org/TR/REC-xml-names/">http://www.w3.org/TR/REC-xml-names/</a></li> 33 <li>XML Base: <a 34 href="http://www.w3.org/TR/xmlbase/">http://www.w3.org/TR/xmlbase/</a></li> 35 <li><a href="http://www.cis.ohio-state.edu/rfc/rfc2396.txt">RFC 2396</a> 36 : Uniform Resource Identifiers <a 37 href="http://www.ietf.org/rfc/rfc2396.txt">http://www.ietf.org/rfc/rfc2396.txt</a></li> 38 <li>XML Path Language (XPath) 1.0: <a 39 href="http://www.w3.org/TR/xpath">http://www.w3.org/TR/xpath</a></li> 40 <li>HTML4 parser: <a 41 href="http://www.w3.org/TR/html401/">http://www.w3.org/TR/html401/</a></li> 42 <li>most of XML Pointer Language (XPointer) Version 1.0: <a 43 href="http://www.w3.org/TR/xptr">http://www.w3.org/TR/xptr</a></li> 44 <li>XML Inclusions (XInclude) Version 1.0: <a 45 href="http://www.w3.org/TR/xinclude/">http://www.w3.org/TR/xinclude/</a></li> 46 <li>[ISO-8859-1], <a 47 href="http://www.cis.ohio-state.edu/rfc/rfc2044.txt">rfc2044</a> [UTF-8] 48 and <a href="http://www.cis.ohio-state.edu/rfc/rfc2781.txt">rfc2781</a> 49 [UTF-16] core encodings</li> 50 <li>part of SGML Open Technical Resolution TR9401:1997</li> 51 <li>XML Catalogs Working Draft 06 August 2001: <a 52 href="http://www.oasis-open.org/committees/entity/spec-2001-08-06.html">http://www.oasis-open.org/committees/entity/spec-2001-08-06.html</a></li> 53</ul> 54 55<p>In most cases libxml tries to implement the specifications in a relatively 56strict way. To some extent libxml2 provide some support for the following 57other specification but don't claim to implement them:</p> 58<ul> 59 <li>Document Object Model (DOM) <a 60 href="http://www.w3.org/TR/DOM-Level-2-Core/">http://www.w3.org/TR/DOM-Level-2-Core/</a> 61 it doesn't implement the API itself, gdome2 does this in top of 62 libxml2</li> 63 <li><a href="http://www.cis.ohio-state.edu/rfc/rfc959.txt">RFC 959</a> 64 : libxml implements a basic FTP client code</li> 65 <li><a href="http://www.cis.ohio-state.edu/rfc/rfc1945.txt">RFC 1945</a> 66 : HTTP/1.0, again a basic HTTP client code</li> 67 <li>SAX: a minimal SAX implementation compatible with early expat 68 versions</li> 69 <li>DocBook SGML v4: libxml2 includes a hackish parser to transition to 70 XML</li> 71</ul> 72 73<p>Libxml2 is known to be very portable, should build and work without 74serious troubles on a variety of systems (Linux, Unix, Windows, CygWin, 75MacOs, MacOsX, RISC Os, OS/2, VMS, QNX, MVS, ...)</p> 76 77<p>Separate documents:</p> 78<ul> 79 <li><a href="http://xmlsoft.org/XSLT/">the libxslt page</a> 80 providing an implementation of XSLT 1.0 and extensions on top of 81 libxml2</li> 82 <li><a href="http://www.cs.unibo.it/~casarini/gdome2/">the gdome2 page</a> 83 : a standard DOM2 implementation based on libxml2</li> 84</ul> 85 86<h2><a name="Introducti">Introduction</a></h2> 87 88<p>This document describes libxml, the <a 89href="http://www.w3.org/XML/">XML</a> C library developped for the <a 90href="http://www.gnome.org/">Gnome</a> project. <a 91href="http://www.w3.org/XML/">XML is a standard</a> for building tag-based 92structured documents/data.</p> 93 94<p>Here are some key points about libxml:</p> 95<ul> 96 <li>Libxml exports Push and Pull type parser interfaces for both XML and 97 HTML.</li> 98 <li>Libxml can do DTD validation at parse time, using a parsed document 99 instance, or with an arbitrary DTD.</li> 100 <li>Libxml now includes nearly complete <a 101 href="http://www.w3.org/TR/xpath">XPath</a>, <a 102 href="http://www.w3.org/TR/xptr">XPointer</a> and <a 103 href="http://www.w3.org/TR/xinclude">XInclude</a> implementations.</li> 104 <li>It is written in plain C, making as few assumptions as possible, and 105 sticking closely to ANSI C/POSIX for easy embedding. Works on 106 Linux/Unix/Windows, ported to a number of other platforms.</li> 107 <li>Basic support for HTTP and FTP client allowing aplications to fetch 108 remote resources</li> 109 <li>The design is modular, most of the extensions can be compiled out.</li> 110 <li>The internal document repesentation is as close as possible to the <a 111 href="http://www.w3.org/DOM/">DOM</a> interfaces.</li> 112 <li>Libxml also has a <a href="http://www.megginson.com/SAX/index.html">SAX 113 like interface</a>; the interface is designed to be compatible with <a 114 href="http://www.jclark.com/xml/expat.html">Expat</a>.</li> 115 <li>This library is released both under the <a 116 href="http://www.w3.org/Consortium/Legal/copyright-software-19980720.html">W3C 117 IPR</a> and the <a href="http://www.gnu.org/copyleft/lesser.html">GNU 118 LGPL</a>. Use either at your convenience, basically this should make 119 everybody happy, if not, drop me a mail.</li> 120</ul> 121 122<p>Warning: unless you are forced to because your application links with a 123Gnome library requiring it, <strong><span 124style="background-color: #FF0000">Do Not Use libxml1</span></strong>, use 125libxml2</p> 126 127<h2><a name="FAQ">FAQ</a></h2> 128 129<p>Table of Content:</p> 130<ul> 131 <li><a href="FAQ.html#Licence">Licence(s)</a></li> 132 <li><a href="FAQ.html#Installati">Installation</a></li> 133 <li><a href="FAQ.html#Compilatio">Compilation</a></li> 134 <li><a href="FAQ.html#Developer">Developer corner</a></li> 135</ul> 136 137<h3><a name="Licence">Licence</a>(s)</h3> 138<ol> 139 <li><em>Licensing Terms for libxml</em> 140 <p>libxml is released under 2 (compatible) licences:</p> 141 <ul> 142 <li>the <a href="http://www.gnu.org/copyleft/lgpl.html">LGPL</a>: GNU 143 Library General Public License</li> 144 <li>the <a 145 href="http://www.w3.org/Consortium/Legal/copyright-software-19980720.html">W3C 146 IPR</a>: very similar to the XWindow licence</li> 147 </ul> 148 </li> 149 <li><em>Can I embed libxml in a proprietary application ?</em> 150 <p>Yes. The W3C IPR allows you to also keep proprietary the changes you 151 made to libxml, but it would be graceful to provide back bugfixes and 152 improvements as patches for possible incorporation in the main 153 development tree</p> 154 </li> 155</ol> 156 157<h3><a name="Installati">Installation</a></h3> 158<ol> 159 <li>Unless you are forced to because your application links with a Gnome 160 library requiring it, <strong><span style="background-color: #FF0000">Do 161 Not Use libxml1</span></strong>, use libxml2</li> 162 <li><em>Where can I get libxml</em> 163 ? 164 <p>The original distribution comes from <a 165 href="ftp://rpmfind.net/pub/libxml/">rpmfind.net</a> or <a 166 href="ftp://ftp.gnome.org/pub/GNOME/stable/sources/libxml/">gnome.org</a></p> 167 <p>Most linux and Bsd distribution includes libxml, this is probably the 168 safer way for end-users</p> 169 <p>David Doolin provides precompiled Windows versions at <a 170 href="http://www.ce.berkeley.edu/~doolin/code/libxmlwin32/ ">http://www.ce.berkeley.edu/~doolin/code/libxmlwin32/</a></p> 171 </li> 172 <li><em>I see libxml and libxml2 releases, which one should I install ?</em> 173 <ul> 174 <li>If you are not concerned by any existing backward compatibility 175 with existing application, install libxml2 only</li> 176 <li>If you are not doing development, you can safely install both. 177 usually the packages <a 178 href="http://rpmfind.net/linux/RPM/libxml.html">libxml</a> and <a 179 href="http://rpmfind.net/linux/RPM/libxml2.html">libxml2</a> are 180 compatible (this is not the case for development packages)</li> 181 <li>If you are a developer and your system provides separate packaging 182 for shared libraries and the development components, it is possible 183 to install libxml and libxml2, and also <a 184 href="http://rpmfind.net/linux/RPM/libxml-devel.html">libxml-devel</a> 185 and <a 186 href="http://rpmfind.net/linux/RPM/libxml2-devel.html">libxml2-devel</a> 187 too for libxml2 >= 2.3.0</li> 188 <li>If you are developing a new application, please develop against 189 libxml2(-devel)</li> 190 </ul> 191 </li> 192 <li><em>I can't install the libxml package it conflicts with libxml0</em> 193 <p>You probably have an old libxml0 package used to provide the shared 194 library for libxml.so.0, you can probably safely remove it. Anyway the 195 libxml packages provided on <a 196 href="ftp://rpmfind.net/pub/libxml/">rpmfind.net</a> provides 197 libxml.so.0</p> 198 </li> 199 <li><em>I can't install the libxml(2) RPM package due to failed 200 dependancies</em> 201 <p>The most generic solution is to refetch the latest src.rpm , and 202 rebuild it locally with</p> 203 <p><code>rpm --rebuild libxml(2)-xxx.src.rpm</code></p> 204 <p>if everything goes well it will generate two binary rpm (one providing 205 the shared libs and xmllint, and the other one, the -devel package 206 providing includes, static libraries and scripts needed to build 207 applications with libxml(2)) that you can install locally.</p> 208 </li> 209</ol> 210 211<h3><a name="Compilatio">Compilation</a></h3> 212<ol> 213 <li><em>What is the process to compile libxml ?</em> 214 <p>As most UNIX libraries libxml follows the "standard":</p> 215 <p><code>gunzip -c xxx.tar.gz | tar xvf -</code></p> 216 <p><code>cd libxml-xxxx</code></p> 217 <p><code>/configure --help</code></p> 218 <p>to see the options, then the compilation/installation proper</p> 219 <p><code>/configure [possible options]</code></p> 220 <p><code>make</code></p> 221 <p><code>make install</code></p> 222 <p>At that point you may have to rerun ldconfig or similar utility to 223 update your list of installed shared libs.</p> 224 </li> 225 <li><em>What other libraries are needed to compile/install libxml ?</em> 226 <p>Libxml does not requires any other library, the normal C ANSI API 227 should be sufficient (please report any violation to this rule you may 228 find).</p> 229 <p>However if found at configuration time libxml will detect and use the 230 following libs:</p> 231 <ul> 232 <li><a href="http://www.info-zip.org/pub/infozip/zlib/">libz</a> 233 : a highly portable and available widely compression library</li> 234 <li>iconv: a powerful character encoding conversion library. It's 235 included by default on recent glibc libraries, so it doesn't need to 236 be installed specifically on linux. It seems it's now <a 237 href="http://www.opennc.org/onlinepubs/7908799/xsh/iconv.html">part 238 of the official UNIX</a> specification. Here is one <a 239 href="http://clisp.cons.org/~haible/packages-libiconv.html">implementation 240 of the library</a> which source can be found <a 241 href="ftp://ftp.ilog.fr/pub/Users/haible/gnu/">here</a>.</li> 242 </ul> 243 </li> 244 <li><em>make check fails on some platforms</em> 245 <p>Sometime the regression tests results don't completely match the value 246 produced by the parser, and the makefile uses diff to print the delta. On 247 some platforms the diff return breaks the compilation process, if the 248 diff is small this is probably not a serious problem</p> 249 </li> 250 <li><em>I use the CVS version and there is no configure script</em> 251 <p>The configure (and other Makefiles) are generated. Use the autogen.sh 252 script to regenerate the configure and Makefiles, like:</p> 253 <p><code>/autogen.sh --prefix=/usr --disable-shared</code></p> 254 </li> 255 <li><em>I have troubles when running make tests with gcc-3.0</em> 256 <p>It seems the initial release of gcc-3.0 has a problem with the 257 optimizer which miscompiles the URI module. Please use another 258 compiler</p> 259 </li> 260</ol> 261 262<h3><a name="Developer">Developer</a> corner</h3> 263<ol> 264 <li><em>xmlDocDump() generates output on one line</em> 265 <p>libxml will not <strong>invent</strong> spaces in the content of a 266 document since <strong>all spaces in the content of a document are 267 significant</strong>. If you build a tree from the API and want 268 indentation:</p> 269 <ol> 270 <li>the correct way is to generate those yourself too</li> 271 <li>the dangerous way is to ask libxml to add those blanks to your 272 content <strong>modifying the content of your document in the 273 process</strong>. The result may not be what you expect. There is 274 <strong>NO</strong> way to guarantee that such a modification won't 275 impact other part of the content of your document. See <a 276 href="http://xmlsoft.org/html/libxml-parser.html#XMLKEEPBLANKSDEFAULT">xmlKeepBlanksDefault 277 ()</a> and <a 278 href="http://xmlsoft.org/html/libxml-tree.html#XMLSAVEFORMATFILE">xmlSaveFormatFile 279 ()</a></li> 280 </ol> 281 </li> 282 <li>Extra nodes in the document: 283 <p><em>For a XML file as below:</em></p> 284 <pre><?xml version="1.0"?> 285<PLAN xmlns="http://www.argus.ca/autotest/1.0/"> 286<NODE CommFlag="0"/> 287<NODE CommFlag="1"/> 288</PLAN></pre> 289 <p><em>after parsing it with the function 290 pxmlDoc=xmlParseFile(...);</em></p> 291 <p><em>I want to the get the content of the first node (node with the 292 CommFlag="0")</em></p> 293 <p><em>so I did it as following;</em></p> 294 <pre>xmlNodePtr pode; 295pnode=pxmlDoc->children->children;</pre> 296 <p><em>but it does not work. If I change it to</em></p> 297 <pre>pnode=pxmlDoc->children->children->next;</pre> 298 <p><em>then it works. Can someone explain it to me.</em></p> 299 <p></p> 300 <p>In XML all characters in the content of the document are significant 301 <strong>including blanks and formatting line breaks</strong>.</p> 302 <p>The extra nodes you are wondering about are just that, text nodes with 303 the formatting spaces wich are part of the document but that people tend 304 to forget. There is a function <a 305 href="http://xmlsoft.org/html/libxml-parser.html">xmlKeepBlanksDefault 306 ()</a> to remove those at parse time, but that's an heuristic, and its 307 use should be limited to case where you are sure there is no 308 mixed-content in the document.</p> 309 </li> 310 <li><em>I get compilation errors of existing code like when accessing 311 <strong>root</strong> or <strong>childs fields</strong> of nodes</em> 312 <p>You are compiling code developed for libxml version 1 and using a 313 libxml2 development environment. Either switch back to libxml v1 devel or 314 even better fix the code to compile with libxml2 (or both) by <a 315 href="upgrade.html">following the instructions</a>.</p> 316 </li> 317 <li><em>I get compilation errors about non existing 318 <strong>xmlRootNode</strong> or <strong>xmlChildrenNode</strong> 319 fields</em> 320 <p>The source code you are using has been <a 321 href="upgrade.html">upgraded</a> to be able to compile with both libxml 322 and libxml2, but you need to install a more recent version: 323 libxml(-devel) >= 1.8.8 or libxml2(-devel) >= 2.1.0</p> 324 </li> 325 <li><em>XPath implementation looks seriously broken</em> 326 <p>XPath implementation prior to 2.3.0 was really incomplete, upgrade to 327 a recent version, the implementation and debug of libxslt generated fixes 328 for most obvious problems.</p> 329 </li> 330 <li><em>The example provided in the web page does not compile</em> 331 <p>It's hard to maintain the documentation in sync with the code 332 <grin/> ...</p> 333 <p>Check the previous points 1/ and 2/ raised before, and send 334 patches.</p> 335 </li> 336 <li><em>Where can I get more examples and informations than in the web 337 page</em> 338 <p>Ideally a libxml book would be nice. I have no such plan ... But you 339 can:</p> 340 <ul> 341 <li>check more deeply the <a href="html/libxml-lib.html">existing 342 generated doc</a></li> 343 <li>looks for examples of use for libxml function using the Gnome code 344 for example the following will query the full Gnome CVs base for the 345 use of the <strong>xmlAddChild()</strong> function: 346 <p><a 347 href="http://cvs.gnome.org/lxr/search?string=xmlAddChild">http://cvs.gnome.org/lxr/search?string=xmlAddChild</a></p> 348 <p>This may be slow, a large hardware donation to the gnome project 349 could cure this :-)</p> 350 </li> 351 <li><a 352 href="http://cvs.gnome.org/bonsai/rview.cgi?cvsroot=/cvs/gnome&dir=gnome-xml">Browse 353 the libxml source</a> 354 , I try to write code as clean and documented as possible, so 355 looking at it may be helpful</li> 356 </ul> 357 </li> 358 <li>What about C++ ? 359 <p>libxml is written in pure C in order to allow easy reuse on a number 360 of platforms, including embedded systems. I don't intend to convert to 361 C++.</p> 362 <p>There is however a C++ wrapper provided by Ari Johnson 363 <ari@btigate.com> which may fullfill your needs:</p> 364 <p>Website: <a 365 href="http://lusis.org/~ari/xml++/">http://lusis.org/~ari/xml++/</a></p> 366 <p>Download: <a 367 href="http://lusis.org/~ari/xml++/libxml++.tar.gz">http://lusis.org/~ari/xml++/libxml++.tar.gz</a></p> 368 </li> 369 <li>How to validate a document a posteriori ? 370 <p>It is possible to validate documents which had not been validated at 371 initial parsing time or documents who have been built from scratch using 372 the API. Use the <a 373 href="http://xmlsoft.org/html/libxml-valid.html#XMLVALIDATEDTD">xmlValidateDtd()</a> 374 function. It is also possible to simply add a Dtd to an existing 375 document:</p> 376 <pre>xmlDocPtr doc; /* your existing document */ 377 xmlDtdPtr dtd = xmlParseDTD(NULL, filename_of_dtd); /* parse the DTD */ 378 dtd->name = xmlStrDup((xmlChar*)"root_name"); /* use the given root */ 379 380 doc->intSubset = dtd; 381 if (doc->children == NULL) xmlAddChild((xmlNodePtr)doc, (xmlNodePtr)dtd); 382 else xmlAddPrevSibling(doc->children, (xmlNodePtr)dtd); 383 </pre> 384 </li> 385 <li>etc ...</li> 386</ol> 387 388<p></p> 389 390<h2><a name="Documentat">Documentation</a></h2> 391 392<p>There are some on-line resources about using libxml:</p> 393<ol> 394 <li>Check the <a href="FAQ.html">FAQ</a></li> 395 <li>Check the <a href="http://xmlsoft.org/html/libxml-lib.html">extensive 396 documentation</a> automatically extracted from code comments (using <a 397 href="http://cvs.gnome.org/bonsai/rview.cgi?cvsroot=/cvs/gnome&dir=gtk-doc">gtk 398 doc</a>).</li> 399 <li>Look at the documentation about <a href="encoding.html">libxml 400 internationalization support</a></li> 401 <li>This page provides a global overview and <a href="example.html">some 402 examples</a> on how to use libxml.</li> 403 <li><a href="mailto:james@daa.com.au">James Henstridge</a> 404 wrote <a 405 href="http://www.daa.com.au/~james/gnome/xml-sax/xml-sax.html">some nice 406 documentation</a> explaining how to use the libxml SAX interface.</li> 407 <li>George Lebl wrote <a 408 href="http://www-4.ibm.com/software/developer/library/gnome3/">an article 409 for IBM developerWorks</a> about using libxml.</li> 410 <li>Check <a href="http://cvs.gnome.org/lxr/source/gnome-xml/TODO">the TODO 411 file</a></li> 412 <li>Read the <a href="upgrade.html">1.x to 2.x upgrade path</a>. If you are 413 starting a new project using libxml you should really use the 2.x 414 version.</li> 415 <li>And don't forget to look at the <a 416 href="http://mail.gnome.org/archives/xml/">mailing-list archive</a>.</li> 417</ol> 418 419<h2><a name="Reporting">Reporting bugs and getting help</a></h2> 420 421<p>Well, bugs or missing features are always possible, and I will make a 422point of fixing them in a timely fashion. The best way to report a bug is to 423use the <a href="http://bugzilla.gnome.org/buglist.cgi?product=libxml">Gnome 424bug tracking database</a> (make sure to use the "libxml" module name). I look 425at reports there regularly and it's good to have a reminder when a bug is 426still open. Be sure to specify that the bug is for the package libxml.</p> 427 428<p>There is also a mailing-list <a 429href="mailto:xml@gnome.org">xml@gnome.org</a> for libxml, with an <a 430href="http://mail.gnome.org/archives/xml/">on-line archive</a> (<a 431href="http://xmlsoft.org/messages">old</a>). To subscribe to this list, 432please visit the <a 433href="http://mail.gnome.org/mailman/listinfo/xml">associated Web</a> page and 434follow the instructions. <strong>Do not send code, I won't debug it</strong> 435(but patches are really appreciated!).</p> 436 437<p>Check the following <strong><span style="color: #FF0000">before 438posting</span></strong>:</p> 439<ul> 440 <li>read the <a href="FAQ.html">FAQ</a></li> 441 <li>make sure you are <a href="ftp://xmlsoft.org/">using a recent 442 version</a>, and that the problem still shows up in those</li> 443 <li>check the <a href="http://mail.gnome.org/archives/xml/">list 444 archives</a> to see if the problem was reported already, in this case 445 there is probably a fix available, similary check the <a 446 href="http://bugzilla.gnome.org/buglist.cgi?product=libxml">registered 447 open bugs</a></li> 448 <li>make sure you can reproduce the bug with xmllint or one of the test 449 programs found in source in the distribution</li> 450 <li>Please send the command showing the error as well as the input (as an 451 attachement)</li> 452</ul> 453 454<p>Then send the bug with associated informations to reproduce it to the <a 455href="mailto:xml@gnome.org">xml@gnome.org</a> list; if it's really libxml 456related I will approve it.. Please do not send me mail directly, it makes 457things really harder to track and in some cases I'm not the best person to 458answer a given question, ask the list instead.</p> 459 460<p>Of course, bugs reported with a suggested patch for fixing them will 461probably be processed faster.</p> 462 463<p>If you're looking for help, a quick look at <a 464href="http://mail.gnome.org/archives/xml/">the list archive</a> may actually 465provide the answer, I usually send source samples when answering libxml usage 466questions. The <a href="http://xmlsoft.org/html/book1.html">auto-generated 467documentantion</a> is not as polished as I would like (i need to learn more 468about Docbook), but it's a good starting point.</p> 469 470<h2><a name="help">How to help</a></h2> 471 472<p>You can help the project in various ways, the best thing to do first is to 473subscribe to the mailing-list as explained before, check the <a 474href="http://mail.gnome.org/archives/xml/">archives </a>and the <a 475href="http://bugzilla.gnome.org/buglist.cgi?product=libxml">Gnome bug 476database:</a>:</p> 477<ol> 478 <li>provide patches when you find problems</li> 479 <li>provide the diffs when you port libxml to a new platform. They may not 480 be integrated in all cases but help pinpointing portability problems 481 and</li> 482 <li>provide documentation fixes (either as patches to the code comments or 483 as HTML diffs).</li> 484 <li>provide new documentations pieces (translations, examples, etc ...)</li> 485 <li>Check the TODO file and try to close one of the items</li> 486 <li>take one of the points raised in the archive or the bug database and 487 provide a fix. <a href="mailto:daniel@veillard.com">Get in touch with me 488 </a>before to avoid synchronization problems and check that the suggested 489 fix will fit in nicely :-)</li> 490</ol> 491 492<h2><a name="Downloads">Downloads</a></h2> 493 494<p>The latest versions of libxml can be found on <a 495href="ftp://xmlsoft.org/">xmlsoft.org</a> (<a 496href="ftp://speakeasy.rpmfind.net/pub/libxml/">Seattle</a>, <a 497href="ftp://fr.rpmfind.net/pub/libxml/">France</a>) or on the <a 498href="ftp://ftp.gnome.org/pub/GNOME/MIRRORS.html">Gnome FTP server</a> either 499as a <a href="ftp://ftp.gnome.org/pub/GNOME/stable/sources/libxml/">source 500archive</a> or <a 501href="ftp://ftp.gnome.org/pub/GNOME/stable/redhat/i386/libxml/">RPM 502packages</a>. (NOTE that you need both the <a 503href="http://rpmfind.net/linux/RPM/libxml2.html">libxml(2)</a> and <a 504href="http://rpmfind.net/linux/RPM/libxml2-devel.html">libxml(2)-devel</a> 505packages installed to compile applications using libxml.) <a 506href="mailto:izlatkovic@daenet.de">Igor Zlatkovic</a> is now the maintainer 507of the Windows port, <a 508href="http://www.fh-frankfurt.de/~igor/projects/libxml/index.html">he 509provides binaries</a>. <a href="mailto:Gary.Pennington@sun.com">Gary 510Pennington</a> provides <a href="http://garypennington.net/libxml2/">Solaris 511binaries</a>.</p> 512 513<p><a name="Snapshot">Snapshot:</a></p> 514<ul> 515 <li>Code from the W3C cvs base libxml <a 516 href="ftp://xmlsoft.org/cvs-snapshot.tar.gz">cvs-snapshot.tar.gz</a></li> 517 <li>Docs, content of the web site, the list archive included <a 518 href="ftp://xmlsoft.org/libxml-docs.tar.gz">libxml-docs.tar.gz</a></li> 519</ul> 520 521<p><a name="Contribs">Contributions:</a></p> 522 523<p>I do accept external contributions, especially if compiling on another 524platform, get in touch with me to upload the package, wrappers for various 525languages have been provided, and can be found in the <a 526href="contribs.html">contrib section</a></p> 527 528<p>Libxml is also available from CVS:</p> 529<ul> 530 <li><p>The <a 531 href="http://cvs.gnome.org/bonsai/rview.cgi?cvsroot=/cvs/gnome&dir=gnome-xml">Gnome 532 CVS base</a>. Check the <a 533 href="http://developer.gnome.org/tools/cvs.html">Gnome CVS Tools</a> 534 page; the CVS module is <b>gnome-xml</b>.</p> 535 </li> 536 <li>The <strong>libxslt</strong> module is also present there</li> 537</ul> 538 539<h2><a name="News">News</a></h2> 540 541<h3>CVS only : check the <a 542href="http://cvs.gnome.org/lxr/source/gnome-xml/ChangeLog">Changelog</a> file 543for a really accurate description</h3> 544 545<p>Items floating around but not actively worked on, get in touch with me if 546you want to test those</p> 547<ul> 548 <li>Implementing <a href="http://xmlsoft.org/XSLT">XSLT</a>, this is done 549 as a separate C library on top of libxml called libxslt</li> 550 <li>Finishing up <a href="http://www.w3.org/TR/xptr">XPointer</a> and <a 551 href="http://www.w3.org/TR/xinclude">XInclude</a></li> 552 <li>(seeems working but delayed from release) parsing/import of Docbook 553 SGML docs</li> 554</ul> 555 556<h3>2.4.13: Jan 14 2001</h3> 557<ul> 558 <li>update of the documentation: John Fleck and Charlie Bozeman</li> 559 <li>cleanup of timing code from Justin Fletcher</li> 560 <li>fixes for Windows and initial thread support on Win32: Igor and Serguei 561 Narojnyi</li> 562 <li>Cygwin patch from Robert Collins</li> 563 <li>added xmlSetEntityReferenceFunc() for Keith Isdale work on xsldbg</li> 564</ul> 565 566<h3>2.4.12: Dec 7 2001</h3> 567<ul> 568 <li>a few bug fixes: thread (Gary Pennington), xmllint (Geert Kloosterman), 569 XML parser (Robin Berjon), XPointer (Danny Jamshy), I/O cleanups 570 (robert)</li> 571 <li>Eric Lavigne contributed project files for MacOS</li> 572 <li>some makefiles cleanups</li> 573</ul> 574 575<h3>2.4.11: Nov 26 2001</h3> 576<ul> 577 <li>fixed a couple of errors in the includes, fixed a few bugs, some code 578 cleanups</li> 579 <li>xmllint man pages improvement by Heiko Rupp</li> 580 <li>updated VMS build instructions from John A Fotheringham</li> 581 <li>Windows Makefiles updates from Igor</li> 582</ul> 583 584<h3>2.4.10: Nov 10 2001</h3> 585<ul> 586 <li>URI escaping fix (Joel Young)</li> 587 <li>added xmlGetNodePath() (for paths or XPointers generation)</li> 588 <li>Fixes namespace handling problems when using DTD and validation</li> 589 <li>improvements on xmllint: Morus Walter patches for --format and 590 --encode, Stefan Kost and Heiko Rupp improvements on the --shell</li> 591 <li>fixes for xmlcatalog linking pointed by Weiqi Gao</li> 592 <li>fixes to the HTML parser</li> 593</ul> 594 595<h3>2.4.9: Nov 6 2001</h3> 596<ul> 597 <li>fixes more catalog bugs</li> 598 <li>avoid a compilation problem, improve xmlGetLineNo()</li> 599</ul> 600 601<h3>2.4.8: Nov 4 2001</h3> 602<ul> 603 <li>fixed SGML catalogs broken in previous release, updated xmlcatalog 604 tool</li> 605 <li>fixed a compile errors and some includes troubles.</li> 606</ul> 607 608<h3>2.4.7: Oct 30 2001</h3> 609<ul> 610 <li>exported some debugging interfaces</li> 611 <li>serious rewrite of the catalog code</li> 612 <li>integrated Gary Pennington thread safety patch, added configure option 613 and regression tests</li> 614 <li>removed an HTML parser bug</li> 615 <li>fixed a couple of potentially serious validation bugs</li> 616 <li>integrated the SGML DocBook support in xmllint</li> 617 <li>changed the nanoftp anonymous login passwd</li> 618 <li>some I/O cleanup and a couple of interfaces for Perl wrapper</li> 619 <li>general bug fixes</li> 620 <li>updated xmllint man page by John Fleck</li> 621 <li>some VMS and Windows updates</li> 622</ul> 623 624<h3>2.4.6: Oct 10 2001</h3> 625<ul> 626 <li>added an updated man pages by John Fleck</li> 627 <li>portability and configure fixes</li> 628 <li>an infinite loop on the HTML parser was removed (William)</li> 629 <li>Windows makefile patches from Igor</li> 630 <li>fixed half a dozen bugs reported fof libxml or libxslt</li> 631 <li>updated xmlcatalog to be able to modify SGML super catalogs</li> 632</ul> 633 634<h3>2.4.5: Sep 14 2001</h3> 635<ul> 636 <li>Remove a few annoying bugs in 2.4.4</li> 637 <li>forces the HTML serializer to output decimal charrefs since some 638 version of Netscape can't handle hexadecimal ones</li> 639</ul> 640 641<h3>1.8.16: Sep 14 2001</h3> 642<ul> 643 <li>maintenance release of the old libxml1 branch, couple of bug and 644 portability fixes</li> 645</ul> 646 647<h3>2.4.4: Sep 12 2001</h3> 648<ul> 649 <li>added --convert to xmlcatalog, bug fixes and cleanups of XML 650 Catalog</li> 651 <li>a few bug fixes and some portability changes</li> 652 <li>some documentation cleanups</li> 653</ul> 654 655<h3>2.4.3: Aug 23 2001</h3> 656<ul> 657 <li>XML Catalog support see the doc</li> 658 <li>New NaN/Infinity floating point code</li> 659 <li>A few bug fixes</li> 660</ul> 661 662<h3>2.4.2: Aug 15 2001</h3> 663<ul> 664 <li>adds xmlLineNumbersDefault() to control line number generation</li> 665 <li>lot of bug fixes</li> 666 <li>the Microsoft MSC projects files shuld now be up to date</li> 667 <li>inheritance of namespaces from DTD defaulted attributes</li> 668 <li>fixes a serious potential security bug</li> 669 <li>added a --format option to xmllint</li> 670</ul> 671 672<h3>2.4.1: July 24 2001</h3> 673<ul> 674 <li>possibility to keep line numbers in the tree</li> 675 <li>some computation NaN fixes</li> 676 <li>extension of the XPath API</li> 677 <li>cleanup for alpha and ia64 targets</li> 678 <li>patch to allow saving through HTTP PUT or POST</li> 679</ul> 680 681<h3>2.4.0: July 10 2001</h3> 682<ul> 683 <li>Fixed a few bugs in XPath, validation, and tree handling.</li> 684 <li>Fixed XML Base implementation, added a coupel of examples to the 685 regression tests</li> 686 <li>A bit of cleanup</li> 687</ul> 688 689<h3>2.3.14: July 5 2001</h3> 690<ul> 691 <li>fixed some entities problems and reduce mem requirement when 692 substituing them</li> 693 <li>lots of improvements in the XPath queries interpreter can be 694 substancially faster</li> 695 <li>Makefiles and configure cleanups</li> 696 <li>Fixes to XPath variable eval, and compare on empty node set</li> 697 <li>HTML tag closing bug fixed</li> 698 <li>Fixed an URI reference computating problem when validating</li> 699</ul> 700 701<h3>2.3.13: June 28 2001</h3> 702<ul> 703 <li>2.3.12 configure.in was broken as well as the push mode XML parser</li> 704 <li>a few more fixes for compilation on Windows MSC by Yon Derek</li> 705</ul> 706 707<h3>1.8.14: June 28 2001</h3> 708<ul> 709 <li>Zbigniew Chyla gave a patch to use the old XML parser in push mode</li> 710 <li>Small Makefile fix</li> 711</ul> 712 713<h3>2.3.12: June 26 2001</h3> 714<ul> 715 <li>lots of cleanup</li> 716 <li>a couple of validation fix</li> 717 <li>fixed line number counting</li> 718 <li>fixed serious problems in the XInclude processing</li> 719 <li>added support for UTF8 BOM at beginning of entities</li> 720 <li>fixed a strange gcc optimizer bugs in xpath handling of float, gcc-3.0 721 miscompile uri.c (William), Thomas Leitner provided a fix for the 722 optimizer on Tru64</li> 723 <li>incorporated Yon Derek and Igor Zlatkovic fixes and improvements for 724 compilation on Windows MSC</li> 725 <li>update of libxml-doc.el (Felix Natter)</li> 726 <li>fixed 2 bugs in URI normalization code</li> 727</ul> 728 729<h3>2.3.11: June 17 2001</h3> 730<ul> 731 <li>updates to trio, Makefiles and configure should fix some portability 732 problems (alpha)</li> 733 <li>fixed some HTML serialization problems (pre, script, and block/inline 734 handling), added encoding aware APIs, cleanup of this code</li> 735 <li>added xmlHasNsProp()</li> 736 <li>implemented a specific PI for encoding support in the DocBook SGML 737 parser</li> 738 <li>some XPath fixes (-Infinity, / as a function parameter and namespaces 739 node selection)</li> 740 <li>fixed a performance problem and an error in the validation code</li> 741 <li>fixed XInclude routine to implement the recursive behaviour</li> 742 <li>fixed xmlFreeNode problem when libxml is included statically twice</li> 743 <li>added --version to xmllint for bug reports</li> 744</ul> 745 746<h3>2.3.10: June 1 2001</h3> 747<ul> 748 <li>fixed the SGML catalog support</li> 749 <li>a number of reported bugs got fixed, in XPath, iconv detection, 750 XInclude processing</li> 751 <li>XPath string function should now handle unicode correctly</li> 752</ul> 753 754<h3>2.3.9: May 19 2001</h3> 755 756<p>Lots of bugfixes, and added a basic SGML catalog support:</p> 757<ul> 758 <li>HTML push bugfix #54891 and another patch from Jonas Borgstr�m</li> 759 <li>some serious speed optimisation again</li> 760 <li>some documentation cleanups</li> 761 <li>trying to get better linking on solaris (-R)</li> 762 <li>XPath API cleanup from Thomas Broyer</li> 763 <li>Validation bug fixed #54631, added a patch from Gary Pennington, fixed 764 xmlValidGetValidElements()</li> 765 <li>Added an INSTALL file</li> 766 <li>Attribute removal added to API: #54433</li> 767 <li>added a basic support for SGML catalogs</li> 768 <li>fixed xmlKeepBlanksDefault(0) API</li> 769 <li>bugfix in xmlNodeGetLang()</li> 770 <li>fixed a small configure portability problem</li> 771 <li>fixed an inversion of SYSTEM and PUBLIC identifier in HTML document</li> 772</ul> 773 774<h3>1.8.13: May 14 2001</h3> 775<ul> 776 <li>bugfixes release of the old libxml1 branch used by Gnome</li> 777</ul> 778 779<h3>2.3.8: May 3 2001</h3> 780<ul> 781 <li>Integrated an SGML DocBook parser for the Gnome project</li> 782 <li>Fixed a few things in the HTML parser</li> 783 <li>Fixed some XPath bugs raised by XSLT use, tried to fix the floating 784 point portability issue</li> 785 <li>Speed improvement (8M/s for SAX, 3M/s for DOM, 1.5M/s for 786 DOM+validation using the XML REC as input and a 700MHz celeron).</li> 787 <li>incorporated more Windows cleanup</li> 788 <li>added xmlSaveFormatFile()</li> 789 <li>fixed problems in copying nodes with entities references (gdome)</li> 790 <li>removed some troubles surrounding the new validation module</li> 791</ul> 792 793<h3>2.3.7: April 22 2001</h3> 794<ul> 795 <li>lots of small bug fixes, corrected XPointer</li> 796 <li>Non determinist content model validation support</li> 797 <li>added xmlDocCopyNode for gdome2</li> 798 <li>revamped the way the HTML parser handles end of tags</li> 799 <li>XPath: corrctions of namespacessupport and number formatting</li> 800 <li>Windows: Igor Zlatkovic patches for MSC compilation</li> 801 <li>HTML ouput fixes from P C Chow and William M. Brack</li> 802 <li>Improved validation speed sensible for DocBook</li> 803 <li>fixed a big bug with ID declared in external parsed entities</li> 804 <li>portability fixes, update of Trio from Bjorn Reese</li> 805</ul> 806 807<h3>2.3.6: April 8 2001</h3> 808<ul> 809 <li>Code cleanup using extreme gcc compiler warning options, found and 810 cleared half a dozen potential problem</li> 811 <li>the Eazel team found an XML parser bug</li> 812 <li>cleaned up the user of some of the string formatting function. used the 813 trio library code to provide the one needed when the platform is missing 814 them</li> 815 <li>xpath: removed a memory leak and fixed the predicate evaluation 816 problem, extended the testsuite and cleaned up the result. XPointer seems 817 broken ...</li> 818</ul> 819 820<h3>2.3.5: Mar 23 2001</h3> 821<ul> 822 <li>Biggest change is separate parsing and evaluation of XPath expressions, 823 there is some new APIs for this too</li> 824 <li>included a number of bug fixes(XML push parser, 51876, notations, 825 52299)</li> 826 <li>Fixed some portability issues</li> 827</ul> 828 829<h3>2.3.4: Mar 10 2001</h3> 830<ul> 831 <li>Fixed bugs #51860 and #51861</li> 832 <li>Added a global variable xmlDefaultBufferSize to allow default buffer 833 size to be application tunable.</li> 834 <li>Some cleanup in the validation code, still a bug left and this part 835 should probably be rewritten to support ambiguous content model :-\</li> 836 <li>Fix a couple of serious bugs introduced or raised by changes in 2.3.3 837 parser</li> 838 <li>Fixed another bug in xmlNodeGetContent()</li> 839 <li>Bjorn fixed XPath node collection and Number formatting</li> 840 <li>Fixed a loop reported in the HTML parsing</li> 841 <li>blank space are reported even if the Dtd content model proves that they 842 are formatting spaces, this is for XmL conformance</li> 843</ul> 844 845<h3>2.3.3: Mar 1 2001</h3> 846<ul> 847 <li>small change in XPath for XSLT</li> 848 <li>documentation cleanups</li> 849 <li>fix in validation by Gary Pennington</li> 850 <li>serious parsing performances improvements</li> 851</ul> 852 853<h3>2.3.2: Feb 24 2001</h3> 854<ul> 855 <li>chasing XPath bugs, found a bunch, completed some TODO</li> 856 <li>fixed a Dtd parsing bug</li> 857 <li>fixed a bug in xmlNodeGetContent</li> 858 <li>ID/IDREF support partly rewritten by Gary Pennington</li> 859</ul> 860 861<h3>2.3.1: Feb 15 2001</h3> 862<ul> 863 <li>some XPath and HTML bug fixes for XSLT</li> 864 <li>small extension of the hash table interfaces for DOM gdome2 865 implementation</li> 866 <li>A few bug fixes</li> 867</ul> 868 869<h3>2.3.0: Feb 8 2001 (2.2.12 was on 25 Jan but I didn't kept track)</h3> 870<ul> 871 <li>Lots of XPath bug fixes</li> 872 <li>Add a mode with Dtd lookup but without validation error reporting for 873 XSLT</li> 874 <li>Add support for text node without escaping (XSLT)</li> 875 <li>bug fixes for xmlCheckFilename</li> 876 <li>validation code bug fixes from Gary Pennington</li> 877 <li>Patch from Paul D. Smith correcting URI path normalization</li> 878 <li>Patch to allow simultaneous install of libxml-devel and 879 libxml2-devel</li> 880 <li>the example Makefile is now fixed</li> 881 <li>added HTML to the RPM packages</li> 882 <li>tree copying bugfixes</li> 883 <li>updates to Windows makefiles</li> 884 <li>optimisation patch from Bjorn Reese</li> 885</ul> 886 887<h3>2.2.11: Jan 4 2001</h3> 888<ul> 889 <li>bunch of bug fixes (memory I/O, xpath, ftp/http, ...)</li> 890 <li>added htmlHandleOmittedElem()</li> 891 <li>Applied Bjorn Reese's IPV6 first patch</li> 892 <li>Applied Paul D. Smith patches for validation of XInclude results</li> 893 <li>added XPointer xmlns() new scheme support</li> 894</ul> 895 896<h3>2.2.10: Nov 25 2000</h3> 897<ul> 898 <li>Fix the Windows problems of 2.2.8</li> 899 <li>integrate OpenVMS patches</li> 900 <li>better handling of some nasty HTML input</li> 901 <li>Improved the XPointer implementation</li> 902 <li>integrate a number of provided patches</li> 903</ul> 904 905<h3>2.2.9: Nov 25 2000</h3> 906<ul> 907 <li>erroneous release :-(</li> 908</ul> 909 910<h3>2.2.8: Nov 13 2000</h3> 911<ul> 912 <li>First version of <a href="http://www.w3.org/TR/xinclude">XInclude</a> 913 support</li> 914 <li>Patch in conditional section handling</li> 915 <li>updated MS compiler project</li> 916 <li>fixed some XPath problems</li> 917 <li>added an URI escaping function</li> 918 <li>some other bug fixes</li> 919</ul> 920 921<h3>2.2.7: Oct 31 2000</h3> 922<ul> 923 <li>added message redirection</li> 924 <li>XPath improvements (thanks TOM !)</li> 925 <li>xmlIOParseDTD() added</li> 926 <li>various small fixes in the HTML, URI, HTTP and XPointer support</li> 927 <li>some cleanup of the Makefile, autoconf and the distribution content</li> 928</ul> 929 930<h3>2.2.6: Oct 25 2000:</h3> 931<ul> 932 <li>Added an hash table module, migrated a number of internal structure to 933 those</li> 934 <li>Fixed a posteriori validation problems</li> 935 <li>HTTP module cleanups</li> 936 <li>HTML parser improvements (tag errors, script/style handling, attribute 937 normalization)</li> 938 <li>coalescing of adjacent text nodes</li> 939 <li>couple of XPath bug fixes, exported the internal API</li> 940</ul> 941 942<h3>2.2.5: Oct 15 2000:</h3> 943<ul> 944 <li>XPointer implementation and testsuite</li> 945 <li>Lot of XPath fixes, added variable and functions registration, more 946 tests</li> 947 <li>Portability fixes, lots of enhancements toward an easy Windows build 948 and release</li> 949 <li>Late validation fixes</li> 950 <li>Integrated a lot of contributed patches</li> 951 <li>added memory management docs</li> 952 <li>a performance problem when using large buffer seems fixed</li> 953</ul> 954 955<h3>2.2.4: Oct 1 2000:</h3> 956<ul> 957 <li>main XPath problem fixed</li> 958 <li>Integrated portability patches for Windows</li> 959 <li>Serious bug fixes on the URI and HTML code</li> 960</ul> 961 962<h3>2.2.3: Sep 17 2000</h3> 963<ul> 964 <li>bug fixes</li> 965 <li>cleanup of entity handling code</li> 966 <li>overall review of all loops in the parsers, all sprintf usage has been 967 checked too</li> 968 <li>Far better handling of larges Dtd. Validating against Docbook XML Dtd 969 works smoothly now.</li> 970</ul> 971 972<h3>1.8.10: Sep 6 2000</h3> 973<ul> 974 <li>bug fix release for some Gnome projects</li> 975</ul> 976 977<h3>2.2.2: August 12 2000</h3> 978<ul> 979 <li>mostly bug fixes</li> 980 <li>started adding routines to access xml parser context options</li> 981</ul> 982 983<h3>2.2.1: July 21 2000</h3> 984<ul> 985 <li>a purely bug fixes release</li> 986 <li>fixed an encoding support problem when parsing from a memory block</li> 987 <li>fixed a DOCTYPE parsing problem</li> 988 <li>removed a bug in the function allowing to override the memory 989 allocation routines</li> 990</ul> 991 992<h3>2.2.0: July 14 2000</h3> 993<ul> 994 <li>applied a lot of portability fixes</li> 995 <li>better encoding support/cleanup and saving (content is now always 996 encoded in UTF-8)</li> 997 <li>the HTML parser now correctly handles encodings</li> 998 <li>added xmlHasProp()</li> 999 <li>fixed a serious problem with &#38;</li> 1000 <li>propagated the fix to FTP client</li> 1001 <li>cleanup, bugfixes, etc ...</li> 1002 <li>Added a page about <a href="encoding.html">libxml Internationalization 1003 support</a></li> 1004</ul> 1005 1006<h3>1.8.9: July 9 2000</h3> 1007<ul> 1008 <li>fixed the spec the RPMs should be better</li> 1009 <li>fixed a serious bug in the FTP implementation, released 1.8.9 to solve 1010 rpmfind users problem</li> 1011</ul> 1012 1013<h3>2.1.1: July 1 2000</h3> 1014<ul> 1015 <li>fixes a couple of bugs in the 2.1.0 packaging</li> 1016 <li>improvements on the HTML parser</li> 1017</ul> 1018 1019<h3>2.1.0 and 1.8.8: June 29 2000</h3> 1020<ul> 1021 <li>1.8.8 is mostly a comodity package for upgrading to libxml2 accoding to 1022 <a href="upgrade.html">new instructions</a>. It fixes a nasty problem 1023 about &#38; charref parsing</li> 1024 <li>2.1.0 also ease the upgrade from libxml v1 to the recent version. it 1025 also contains numerous fixes and enhancements: 1026 <ul> 1027 <li>added xmlStopParser() to stop parsing</li> 1028 <li>improved a lot parsing speed when there is large CDATA blocs</li> 1029 <li>includes XPath patches provided by Picdar Technology</li> 1030 <li>tried to fix as much as possible DtD validation and namespace 1031 related problems</li> 1032 <li>output to a given encoding has been added/tested</li> 1033 <li>lot of various fixes</li> 1034 </ul> 1035 </li> 1036</ul> 1037 1038<h3>2.0.0: Apr 12 2000</h3> 1039<ul> 1040 <li>First public release of libxml2. If you are using libxml, it's a good 1041 idea to check the 1.x to 2.x upgrade instructions. NOTE: while initally 1042 scheduled for Apr 3 the relase occured only on Apr 12 due to massive 1043 workload.</li> 1044 <li>The include are now located under $prefix/include/libxml (instead of 1045 $prefix/include/gnome-xml), they also are referenced by 1046 <pre>#include <libxml/xxx.h></pre> 1047 <p>instead of</p> 1048 <pre>#include "xxx.h"</pre> 1049 </li> 1050 <li>a new URI module for parsing URIs and following strictly RFC 2396</li> 1051 <li>the memory allocation routines used by libxml can now be overloaded 1052 dynamically by using xmlMemSetup()</li> 1053 <li>The previously CVS only tool tester has been renamed 1054 <strong>xmllint</strong> and is now installed as part of the libxml2 1055 package</li> 1056 <li>The I/O interface has been revamped. There is now ways to plug in 1057 specific I/O modules, either at the URI scheme detection level using 1058 xmlRegisterInputCallbacks() or by passing I/O functions when creating a 1059 parser context using xmlCreateIOParserCtxt()</li> 1060 <li>there is a C preprocessor macro LIBXML_VERSION providing the version 1061 number of the libxml module in use</li> 1062 <li>a number of optional features of libxml can now be excluded at 1063 configure time (FTP/HTTP/HTML/XPath/Debug)</li> 1064</ul> 1065 1066<h3>2.0.0beta: Mar 14 2000</h3> 1067<ul> 1068 <li>This is a first Beta release of libxml version 2</li> 1069 <li>It's available only from<a href="ftp://xmlsoft.org/">xmlsoft.org 1070 FTP</a>, it's packaged as libxml2-2.0.0beta and available as tar and 1071 RPMs</li> 1072 <li>This version is now the head in the Gnome CVS base, the old one is 1073 available under the tag LIB_XML_1_X</li> 1074 <li>This includes a very large set of changes. Froma programmatic point of 1075 view applications should not have to be modified too much, check the <a 1076 href="upgrade.html">upgrade page</a></li> 1077 <li>Some interfaces may changes (especially a bit about encoding).</li> 1078 <li>the updates includes: 1079 <ul> 1080 <li>fix I18N support. ISO-Latin-x/UTF-8/UTF-16 (nearly) seems correctly 1081 handled now</li> 1082 <li>Better handling of entities, especially well formedness checking 1083 and proper PEref extensions in external subsets</li> 1084 <li>DTD conditional sections</li> 1085 <li>Validation now correcly handle entities content</li> 1086 <li><a href="http://rpmfind.net/tools/gdome/messages/0039.html">change 1087 structures to accomodate DOM</a></li> 1088 </ul> 1089 </li> 1090 <li>Serious progress were made toward compliance, <a 1091 href="conf/result.html">here are the result of the test</a> against the 1092 OASIS testsuite (except the japanese tests since I don't support that 1093 encoding yet). This URL is rebuilt every couple of hours using the CVS 1094 head version.</li> 1095</ul> 1096 1097<h3>1.8.7: Mar 6 2000</h3> 1098<ul> 1099 <li>This is a bug fix release:</li> 1100 <li>It is possible to disable the ignorable blanks heuristic used by 1101 libxml-1.x, a new function xmlKeepBlanksDefault(0) will allow this. Note 1102 that for adherence to XML spec, this behaviour will be disabled by 1103 default in 2.x . The same function will allow to keep compatibility for 1104 old code.</li> 1105 <li>Blanks in <a> </a> constructs are not ignored anymore, 1106 avoiding heuristic is really the Right Way :-\</li> 1107 <li>The unchecked use of snprintf which was breaking libxml-1.8.6 1108 compilation on some platforms has been fixed</li> 1109 <li>nanoftp.c nanohttp.c: Fixed '#' and '?' stripping when processing 1110 URIs</li> 1111</ul> 1112 1113<h3>1.8.6: Jan 31 2000</h3> 1114<ul> 1115 <li>added a nanoFTP transport module, debugged until the new version of <a 1116 href="http://rpmfind.net/linux/rpm2html/rpmfind.html">rpmfind</a> can use 1117 it without troubles</li> 1118</ul> 1119 1120<h3>1.8.5: Jan 21 2000</h3> 1121<ul> 1122 <li>adding APIs to parse a well balanced chunk of XML (production <a 1123 href="http://www.w3.org/TR/REC-xml#NT-content">[43] content</a> of the 1124 XML spec)</li> 1125 <li>fixed a hideous bug in xmlGetProp pointed by Rune.Djurhuus@fast.no</li> 1126 <li>Jody Goldberg <jgoldberg@home.com> provided another patch trying 1127 to solve the zlib checks problems</li> 1128 <li>The current state in gnome CVS base is expected to ship as 1.8.5 with 1129 gnumeric soon</li> 1130</ul> 1131 1132<h3>1.8.4: Jan 13 2000</h3> 1133<ul> 1134 <li>bug fixes, reintroduced xmlNewGlobalNs(), fixed xmlNewNs()</li> 1135 <li>all exit() call should have been removed from libxml</li> 1136 <li>fixed a problem with INCLUDE_WINSOCK on WIN32 platform</li> 1137 <li>added newDocFragment()</li> 1138</ul> 1139 1140<h3>1.8.3: Jan 5 2000</h3> 1141<ul> 1142 <li>a Push interface for the XML and HTML parsers</li> 1143 <li>a shell-like interface to the document tree (try tester --shell :-)</li> 1144 <li>lots of bug fixes and improvement added over XMas hollidays</li> 1145 <li>fixed the DTD parsing code to work with the xhtml DTD</li> 1146 <li>added xmlRemoveProp(), xmlRemoveID() and xmlRemoveRef()</li> 1147 <li>Fixed bugs in xmlNewNs()</li> 1148 <li>External entity loading code has been revamped, now it uses 1149 xmlLoadExternalEntity(), some fix on entities processing were added</li> 1150 <li>cleaned up WIN32 includes of socket stuff</li> 1151</ul> 1152 1153<h3>1.8.2: Dec 21 1999</h3> 1154<ul> 1155 <li>I got another problem with includes and C++, I hope this issue is fixed 1156 for good this time</li> 1157 <li>Added a few tree modification functions: xmlReplaceNode, 1158 xmlAddPrevSibling, xmlAddNextSibling, xmlNodeSetName and 1159 xmlDocSetRootElement</li> 1160 <li>Tried to improve the HTML output with help from <a 1161 href="mailto:clahey@umich.edu">Chris Lahey</a></li> 1162</ul> 1163 1164<h3>1.8.1: Dec 18 1999</h3> 1165<ul> 1166 <li>various patches to avoid troubles when using libxml with C++ compilers 1167 the "namespace" keyword and C escaping in include files</li> 1168 <li>a problem in one of the core macros IS_CHAR was corrected</li> 1169 <li>fixed a bug introduced in 1.8.0 breaking default namespace processing, 1170 and more specifically the Dia application</li> 1171 <li>fixed a posteriori validation (validation after parsing, or by using a 1172 Dtd not specified in the original document)</li> 1173 <li>fixed a bug in</li> 1174</ul> 1175 1176<h3>1.8.0: Dec 12 1999</h3> 1177<ul> 1178 <li>cleanup, especially memory wise</li> 1179 <li>the parser should be more reliable, especially the HTML one, it should 1180 not crash, whatever the input !</li> 1181 <li>Integrated various patches, especially a speedup improvement for large 1182 dataset from <a href="mailto:cnygard@bellatlantic.net">Carl Nygard</a>, 1183 configure with --with-buffers to enable them.</li> 1184 <li>attribute normalization, oops should have been added long ago !</li> 1185 <li>attributes defaulted from Dtds should be available, xmlSetProp() now 1186 does entities escapting by default.</li> 1187</ul> 1188 1189<h3>1.7.4: Oct 25 1999</h3> 1190<ul> 1191 <li>Lots of HTML improvement</li> 1192 <li>Fixed some errors when saving both XML and HTML</li> 1193 <li>More examples, the regression tests should now look clean</li> 1194 <li>Fixed a bug with contiguous charref</li> 1195</ul> 1196 1197<h3>1.7.3: Sep 29 1999</h3> 1198<ul> 1199 <li>portability problems fixed</li> 1200 <li>snprintf was used unconditionnally, leading to link problems on system 1201 were it's not available, fixed</li> 1202</ul> 1203 1204<h3>1.7.1: Sep 24 1999</h3> 1205<ul> 1206 <li>The basic type for strings manipulated by libxml has been renamed in 1207 1.7.1 from <strong>CHAR</strong> to <strong>xmlChar</strong>. The reason 1208 is that CHAR was conflicting with a predefined type on Windows. However 1209 on non WIN32 environment, compatibility is provided by the way of a 1210 <strong>#define </strong>.</li> 1211 <li>Changed another error : the use of a structure field called errno, and 1212 leading to troubles on platforms where it's a macro</li> 1213</ul> 1214 1215<h3>1.7.0: sep 23 1999</h3> 1216<ul> 1217 <li>Added the ability to fetch remote DTD or parsed entities, see the <a 1218 href="html/libxml-nanohttp.html">nanohttp</a> module.</li> 1219 <li>Added an errno to report errors by another mean than a simple printf 1220 like callback</li> 1221 <li>Finished ID/IDREF support and checking when validation</li> 1222 <li>Serious memory leaks fixed (there is now a <a 1223 href="html/libxml-xmlmemory.html">memory wrapper</a> module)</li> 1224 <li>Improvement of <a href="http://www.w3.org/TR/xpath">XPath</a> 1225 implementation</li> 1226 <li>Added an HTML parser front-end</li> 1227</ul> 1228 1229<h2><a name="XML">XML</a></h2> 1230 1231<p><a href="http://www.w3.org/TR/REC-xml">XML is a standard</a> for 1232markup-based structured documents. Here is <a name="example">an example XML 1233document</a>:</p> 1234<pre><?xml version="1.0"?> 1235<EXAMPLE prop1="gnome is great" prop2="&amp; linux too"> 1236 <head> 1237 <title>Welcome to Gnome</title> 1238 </head> 1239 <chapter> 1240 <title>The Linux adventure</title> 1241 <p>bla bla bla ...</p> 1242 <image href="linus.gif"/> 1243 <p>...</p> 1244 </chapter> 1245</EXAMPLE></pre> 1246 1247<p>The first line specifies that it's an XML document and gives useful 1248information about its encoding. Then the document is a text format whose 1249structure is specified by tags between brackets. <strong>Each tag opened has 1250to be closed</strong>. XML is pedantic about this. However, if a tag is empty 1251(no content), a single tag can serve as both the opening and closing tag if 1252it ends with <code>/></code> rather than with <code>></code>. Note 1253that, for example, the image tag has no content (just an attribute) and is 1254closed by ending the tag with <code>/></code>.</p> 1255 1256<p>XML can be applied sucessfully to a wide range of uses, from long term 1257structured document maintenance (where it follows the steps of SGML) to 1258simple data encoding mechanisms like configuration file formatting (glade), 1259spreadsheets (gnumeric), or even shorter lived documents such as WebDAV where 1260it is used to encode remote calls between a client and a server.</p> 1261 1262<h2><a name="XSLT">XSLT</a></h2> 1263 1264<p>Check <a href="http://xmlsoft.org/XSLT">the separate libxslt page</a></p> 1265 1266<p><a href="http://www.w3.org/TR/xslt">XSL Transformations</a>, is a 1267language for transforming XML documents into other XML documents (or 1268HTML/textual output).</p> 1269 1270<p>A separate library called libxslt is being built on top of libxml2. This 1271module "libxslt" can be found in the Gnome CVS base too.</p> 1272 1273<p>You can check the <a 1274href="http://cvs.gnome.org/lxr/source/libxslt/FEATURES">features</a> 1275supported and the progresses on the <a 1276href="http://cvs.gnome.org/lxr/source/libxslt/ChangeLog">Changelog</a></p> 1277 1278<h2><a name="architecture">libxml architecture</a></h2> 1279 1280<p>Libxml is made of multiple components; some of them are optional, and most 1281of the block interfaces are public. The main components are:</p> 1282<ul> 1283 <li>an Input/Output layer</li> 1284 <li>FTP and HTTP client layers (optional)</li> 1285 <li>an Internationalization layer managing the encodings support</li> 1286 <li>a URI module</li> 1287 <li>the XML parser and its basic SAX interface</li> 1288 <li>an HTML parser using the same SAX interface (optional)</li> 1289 <li>a SAX tree module to build an in-memory DOM representation</li> 1290 <li>a tree module to manipulate the DOM representation</li> 1291 <li>a validation module using the DOM representation (optional)</li> 1292 <li>an XPath module for global lookup in a DOM representation 1293 (optional)</li> 1294 <li>a debug module (optional)</li> 1295</ul> 1296 1297<p>Graphically this gives the following:</p> 1298 1299<p><img src="libxml.gif" alt="a graphical view of the various"></p> 1300 1301<p></p> 1302 1303<h2><a name="tree">The tree output</a></h2> 1304 1305<p>The parser returns a tree built during the document analysis. The value 1306returned is an <strong>xmlDocPtr</strong> (i.e., a pointer to an 1307<strong>xmlDoc</strong> structure). This structure contains information such 1308as the file name, the document type, and a <strong>children</strong> pointer 1309which is the root of the document (or more exactly the first child under the 1310root which is the document). The tree is made of <strong>xmlNode</strong>s, 1311chained in double-linked lists of siblings and with a children<->parent 1312relationship. An xmlNode can also carry properties (a chain of xmlAttr 1313structures). An attribute may have a value which is a list of TEXT or 1314ENTITY_REF nodes.</p> 1315 1316<p>Here is an example (erroneous with respect to the XML spec since there 1317should be only one ELEMENT under the root):</p> 1318 1319<p><img src="structure.gif" alt=" structure.gif "></p> 1320 1321<p>In the source package there is a small program (not installed by default) 1322called <strong>xmllint</strong> which parses XML files given as argument and 1323prints them back as parsed. This is useful for detecting errors both in XML 1324code and in the XML parser itself. It has an option <strong>--debug</strong> 1325which prints the actual in-memory structure of the document; here is the 1326result with the <a href="#example">example</a> given before:</p> 1327<pre>DOCUMENT 1328version=1.0 1329standalone=true 1330 ELEMENT EXAMPLE 1331 ATTRIBUTE prop1 1332 TEXT 1333 content=gnome is great 1334 ATTRIBUTE prop2 1335 ENTITY_REF 1336 TEXT 1337 content= linux too 1338 ELEMENT head 1339 ELEMENT title 1340 TEXT 1341 content=Welcome to Gnome 1342 ELEMENT chapter 1343 ELEMENT title 1344 TEXT 1345 content=The Linux adventure 1346 ELEMENT p 1347 TEXT 1348 content=bla bla bla ... 1349 ELEMENT image 1350 ATTRIBUTE href 1351 TEXT 1352 content=linus.gif 1353 ELEMENT p 1354 TEXT 1355 content=...</pre> 1356 1357<p>This should be useful for learning the internal representation model.</p> 1358 1359<h2><a name="interface">The SAX interface</a></h2> 1360 1361<p>Sometimes the DOM tree output is just too large to fit reasonably into 1362memory. In that case (and if you don't expect to save back the XML document 1363loaded using libxml), it's better to use the SAX interface of libxml. SAX is 1364a <strong>callback-based interface</strong> to the parser. Before parsing, 1365the application layer registers a customized set of callbacks which are 1366called by the library as it progresses through the XML input.</p> 1367 1368<p>To get more detailed step-by-step guidance on using the SAX interface of 1369libxml, see the <a 1370href="http://www.daa.com.au/~james/gnome/xml-sax/xml-sax.html">nice 1371documentation</a>.written by <a href="mailto:james@daa.com.au">James 1372Henstridge</a>.</p> 1373 1374<p>You can debug the SAX behaviour by using the <strong>testSAX</strong> 1375program located in the gnome-xml module (it's usually not shipped in the 1376binary packages of libxml, but you can find it in the tar source 1377distribution). Here is the sequence of callbacks that would be reported by 1378testSAX when parsing the example XML document shown earlier:</p> 1379<pre>SAX.setDocumentLocator() 1380SAX.startDocument() 1381SAX.getEntity(amp) 1382SAX.startElement(EXAMPLE, prop1='gnome is great', prop2='&amp; linux too') 1383SAX.characters( , 3) 1384SAX.startElement(head) 1385SAX.characters( , 4) 1386SAX.startElement(title) 1387SAX.characters(Welcome to Gnome, 16) 1388SAX.endElement(title) 1389SAX.characters( , 3) 1390SAX.endElement(head) 1391SAX.characters( , 3) 1392SAX.startElement(chapter) 1393SAX.characters( , 4) 1394SAX.startElement(title) 1395SAX.characters(The Linux adventure, 19) 1396SAX.endElement(title) 1397SAX.characters( , 4) 1398SAX.startElement(p) 1399SAX.characters(bla bla bla ..., 15) 1400SAX.endElement(p) 1401SAX.characters( , 4) 1402SAX.startElement(image, href='linus.gif') 1403SAX.endElement(image) 1404SAX.characters( , 4) 1405SAX.startElement(p) 1406SAX.characters(..., 3) 1407SAX.endElement(p) 1408SAX.characters( , 3) 1409SAX.endElement(chapter) 1410SAX.characters( , 1) 1411SAX.endElement(EXAMPLE) 1412SAX.endDocument()</pre> 1413 1414<p>Most of the other interfaces of libxml are based on the DOM tree-building 1415facility, so nearly everything up to the end of this document presupposes the 1416use of the standard DOM tree build. Note that the DOM tree itself is built by 1417a set of registered default callbacks, without internal specific 1418interface.</p> 1419 1420<h2><a name="Validation">Validation & DTDs</a></h2> 1421 1422<p>Table of Content:</p> 1423<ol> 1424 <li><a href="#General5">General overview</a></li> 1425 <li><a href="#definition">The definition</a></li> 1426 <li><a href="#Simple">Simple rules</a> 1427 <ol> 1428 <li><a href="#reference">How to reference a DTD from a document</a></li> 1429 <li><a href="#Declaring">Declaring elements</a></li> 1430 <li><a href="#Declaring1">Declaring attributes</a></li> 1431 </ol> 1432 </li> 1433 <li><a href="#Some">Some examples</a></li> 1434 <li><a href="#validate">How to validate</a></li> 1435 <li><a href="#Other">Other resources</a></li> 1436</ol> 1437 1438<h3><a name="General5">General overview</a></h3> 1439 1440<p>Well what is validation and what is a DTD ?</p> 1441 1442<p>DTD is the acronym for Document Type Definition. This is a description of 1443the content for a familly of XML files. This is part of the XML 1.0 1444specification, and alows to describe and check that a given document instance 1445conforms to a set of rules detailing its structure and content.</p> 1446 1447<p>Validation is the process of checking a document against a DTD (more 1448generally against a set of construction rules).</p> 1449 1450<p>The validation process and building DTDs are the two most difficult parts 1451of the XML life cycle. Briefly a DTD defines all the possibles element to be 1452found within your document, what is the formal shape of your document tree 1453(by defining the allowed content of an element, either text, a regular 1454expression for the allowed list of children, or mixed content i.e. both text 1455and children). The DTD also defines the allowed attributes for all elements 1456and the types of the attributes.</p> 1457 1458<h3><a name="definition1">The definition</a></h3> 1459 1460<p>The <a href="http://www.w3.org/TR/REC-xml">W3C XML Recommendation</a> (<a 1461href="http://www.xml.com/axml/axml.html">Tim Bray's annotated version of 1462Rev1</a>):</p> 1463<ul> 1464 <li><a href="http://www.w3.org/TR/REC-xml#elemdecls">Declaring 1465 elements</a></li> 1466 <li><a href="http://www.w3.org/TR/REC-xml#attdecls">Declaring 1467 attributes</a></li> 1468</ul> 1469 1470<p>(unfortunately) all this is inherited from the SGML world, the syntax is 1471ancient...</p> 1472 1473<h3><a name="Simple1">Simple rules</a></h3> 1474 1475<p>Writing DTD can be done in multiple ways, the rules to build them if you 1476need something fixed or something which can evolve over time can be radically 1477different. Really complex DTD like Docbook ones are flexible but quite harder 1478to design. I will just focuse on DTDs for a formats with a fixed simple 1479structure. It is just a set of basic rules, and definitely not exhaustive nor 1480useable for complex DTD design.</p> 1481 1482<h4><a name="reference1">How to reference a DTD from a document</a>:</h4> 1483 1484<p>Assuming the top element of the document is <code>spec</code> and the dtd 1485is placed in the file <code>mydtd</code> in the subdirectory 1486<code>dtds</code> of the directory from where the document were loaded:</p> 1487 1488<p><code><!DOCTYPE spec SYSTEM "dtds/mydtd"></code></p> 1489 1490<p>Notes:</p> 1491<ul> 1492 <li>the system string is actually an URI-Reference (as defined in <a 1493 href="http://www.ietf.org/rfc/rfc2396.txt">RFC 2396</a>) so you can use a 1494 full URL string indicating the location of your DTD on the Web, this is a 1495 really good thing to do if you want others to validate your document</li> 1496 <li>it is also possible to associate a <code>PUBLIC</code> identifier (a 1497 magic string) so that the DTd is looked up in catalogs on the client side 1498 without having to locate it on the web</li> 1499 <li>a dtd contains a set of elements and attributes declarations, but they 1500 don't define what the root of the document should be. This is explicitely 1501 told to the parser/validator as the first element of the 1502 <code>DOCTYPE</code> declaration.</li> 1503</ul> 1504 1505<h4><a name="Declaring2">Declaring elements</a>:</h4> 1506 1507<p>The following declares an element <code>spec</code>:</p> 1508 1509<p><code><!ELEMENT spec (front, body, back?)></code></p> 1510 1511<p>it also expresses that the spec element contains one <code>front</code>, 1512one <code>body</code> and one optionnal <code>back</code> children elements 1513in this order. The declaration of one element of the structure and its 1514content are done in a single declaration. Similary the following declares 1515<code>div1</code> elements:</p> 1516 1517<p><code><!ELEMENT div1 (head, (p | list | note)*, div2?)></code></p> 1518 1519<p>means div1 contains one <code>head</code> then a series of optional 1520<code>p</code>, <code>list</code>s and <code>note</code>s and then an 1521optional <code>div2</code>. And last but not least an element can contain 1522text:</p> 1523 1524<p><code><!ELEMENT b (#PCDATA)></code></p> 1525 1526<p><code>b</code> contains text or being of mixed content (text and elements 1527in no particular order):</p> 1528 1529<p><code><!ELEMENT p (#PCDATA|a|ul|b|i|em)*></code></p> 1530 1531<p><code>p </code>can contain text or <code>a</code>, <code>ul</code>, 1532<code>b</code>, <code>i </code>or <code>em</code> elements in no particular 1533order.</p> 1534 1535<h4><a name="Declaring1">Declaring attributes</a>:</h4> 1536 1537<p>again the attributes declaration includes their content definition:</p> 1538 1539<p><code><!ATTLIST termdef name CDATA #IMPLIED></code></p> 1540 1541<p>means that the element <code>termdef</code> can have a <code>name</code> 1542attribute containing text (<code>CDATA</code>) and which is optionnal 1543(<code>#IMPLIED</code>). The attribute value can also be defined within a 1544set:</p> 1545 1546<p><code><!ATTLIST list type (bullets|ordered|glossary) 1547"ordered"></code></p> 1548 1549<p>means <code>list</code> element have a <code>type</code> attribute with 3 1550allowed values "bullets", "ordered" or "glossary" and which default to 1551"ordered" if the attribute is not explicitely specified.</p> 1552 1553<p>The content type of an attribute can be text (<code>CDATA</code>), 1554anchor/reference/references 1555(<code>ID</code>/<code>IDREF</code>/<code>IDREFS</code>), entity(ies) 1556(<code>ENTITY</code>/<code>ENTITIES</code>) or name(s) 1557(<code>NMTOKEN</code>/<code>NMTOKENS</code>). The following defines that a 1558<code>chapter</code> element can have an optional <code>id</code> attribute 1559of type <code>ID</code>, usable for reference from attribute of type 1560IDREF:</p> 1561 1562<p><code><!ATTLIST chapter id ID #IMPLIED></code></p> 1563 1564<p>The last value of an attribute definition can be <code>#REQUIRED 1565</code>meaning that the attribute has to be given, <code>#IMPLIED</code> 1566meaning that it is optional, or the default value (possibly prefixed by 1567<code>#FIXED</code> if it is the only allowed).</p> 1568 1569<p>Notes:</p> 1570<ul> 1571 <li>usually the attributes pertaining to a given element are declared in a 1572 single expression, but it is just a convention adopted by a lot of DTD 1573 writers: 1574 <pre><!ATTLIST termdef 1575 id ID #REQUIRED 1576 name CDATA #IMPLIED></pre> 1577 <p>The previous construct defines both <code>id</code> and 1578 <code>name</code> attributes for the element <code>termdef</code></p> 1579 </li> 1580</ul> 1581 1582<h3><a name="Some1">Some examples</a></h3> 1583 1584<p>The directory <code>test/valid/dtds/</code> in the libxml distribution 1585contains some complex DTD examples. The <code>test/valid/dia.xml</code> 1586example shows an XML file where the simple DTD is directly included within 1587the document.</p> 1588 1589<h3><a name="validate1">How to validate</a></h3> 1590 1591<p>The simplest is to use the xmllint program comming with libxml. The 1592<code>--valid</code> option turn on validation of the files given as input, 1593for example the following validates a copy of the first revision of the XML 15941.0 specification:</p> 1595 1596<p><code>xmllint --valid --noout test/valid/REC-xml-19980210.xml</code></p> 1597 1598<p>the -- noout is used to not output the resulting tree.</p> 1599 1600<p>The <code>--dtdvalid dtd</code> allows to validate the document(s) against 1601a given DTD.</p> 1602 1603<p>Libxml exports an API to handle DTDs and validation, check the <a 1604href="http://xmlsoft.org/html/libxml-valid.html">associated 1605description</a>.</p> 1606 1607<h3><a name="Other1">Other resources</a></h3> 1608 1609<p>DTDs are as old as SGML. So there may be a number of examples on-line, I 1610will just list one for now, others pointers welcome:</p> 1611<ul> 1612 <li><a href="http://www.xml101.com:8081/dtd/">XML-101 DTD</a></li> 1613</ul> 1614 1615<p>I suggest looking at the examples found under test/valid/dtd and any of 1616the large number of books available on XML. The dia example in test/valid 1617should be both simple and complete enough to allow you to build your own.</p> 1618 1619<p></p> 1620 1621<h2><a name="Memory">Memory Management</a></h2> 1622 1623<p>Table of Content:</p> 1624<ol> 1625 <li><a href="#General3">General overview</a></li> 1626 <li><a href="#setting">Setting libxml set of memory routines</a></li> 1627 <li><a href="#cleanup">Cleaning up after parsing</a></li> 1628 <li><a href="#Debugging">Debugging routines</a></li> 1629 <li><a href="#General4">General memory requirements</a></li> 1630</ol> 1631 1632<h3><a name="General3">General overview</a></h3> 1633 1634<p>The module <code><a 1635href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlmemory.h</a></code> 1636provides the interfaces to the libxml memory system:</p> 1637<ul> 1638 <li>libxml does not use the libc memory allocator directly but xmlFree(), 1639 xmlMalloc() and xmlRealloc()</li> 1640 <li>those routines can be reallocated to a specific set of routine, by 1641 default the libc ones i.e. free(), malloc() and realloc()</li> 1642 <li>the xmlmemory.c module includes a set of debugging routine</li> 1643</ul> 1644 1645<h3><a name="setting">Setting libxml set of memory routines</a></h3> 1646 1647<p>It is sometimes useful to not use the default memory allocator, either for 1648debugging, analysis or to implement a specific behaviour on memory management 1649(like on embedded systems). Two function calls are available to do so:</p> 1650<ul> 1651 <li><a href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemGet ()</a> 1652 which return the current set of functions in use by the parser</li> 1653 <li><a 1654 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemSetup()</a> 1655 which allow to set up a new set of memory allocation functions</li> 1656</ul> 1657 1658<p>Of course a call to xmlMemSetup() should probably be done before calling 1659any other libxml routines (unless you are sure your allocations routines are 1660compatibles).</p> 1661 1662<h3><a name="cleanup">Cleaning up after parsing</a></h3> 1663 1664<p>Libxml is not stateless, there is a few set of memory structures needing 1665allocation before the parser is fully functionnal (some encoding structures 1666for example). This also mean that once parsing is finished there is a tiny 1667amount of memory (a few hundred bytes) which can be recollected if you don't 1668reuse the parser immediately:</p> 1669<ul> 1670 <li><a href="http://xmlsoft.org/html/libxml-parser.html">xmlCleanupParser 1671 ()</a> 1672 is a centralized routine to free the parsing states. Note that it won't 1673 deallocate any produced tree if any (use the xmlFreeDoc() and related 1674 routines for this).</li> 1675 <li><a href="http://xmlsoft.org/html/libxml-parser.html">xmlInitParser 1676 ()</a> 1677 is the dual routine allowing to preallocate the parsing state which can 1678 be useful for example to avoid initialization reentrancy problems when 1679 using libxml in multithreaded applications</li> 1680</ul> 1681 1682<p>Generally xmlCleanupParser() is safe, if needed the state will be rebuild 1683at the next invocation of parser routines, but be careful of the consequences 1684in multithreaded applications.</p> 1685 1686<h3><a name="Debugging">Debugging routines</a></h3> 1687 1688<p>When configured using --with-mem-debug flag (off by default), libxml uses 1689a set of memory allocation debugging routineskeeping track of all allocated 1690blocks and the location in the code where the routine was called. A couple of 1691other debugging routines allow to dump the memory allocated infos to a file 1692or call a specific routine when a given block number is allocated:</p> 1693<ul> 1694 <li><a 1695 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMallocLoc()</a> 1696 <a 1697 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlReallocLoc()</a> 1698 and <a 1699 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemStrdupLoc()</a> 1700 are the memory debugging replacement allocation routines</li> 1701 <li><a href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemoryDump 1702 ()</a> 1703 dumps all the informations about the allocated memory block lefts in the 1704 <code>.memdump</code> file</li> 1705</ul> 1706 1707<p>When developping libxml memory debug is enabled, the tests programs call 1708xmlMemoryDump () and the "make test" regression tests will check for any 1709memory leak during the full regression test sequence, this helps a lot 1710ensuring that libxml does not leak memory and bullet proof memory 1711allocations use (some libc implementations are known to be far too permissive 1712resulting in major portability problems!).</p> 1713 1714<p>If the .memdump reports a leak, it displays the allocation function and 1715also tries to give some informations about the content and structure of the 1716allocated blocks left. This is sufficient in most cases to find the culprit, 1717but not always. Assuming the allocation problem is reproductible, it is 1718possible to find more easilly:</p> 1719<ol> 1720 <li>write down the block number xxxx not allocated</li> 1721 <li>export the environement variable XML_MEM_BREAKPOINT=xxxx</li> 1722 <li>run the program under a debugger and set a breakpoint on 1723 xmlMallocBreakpoint() a specific function called when this precise block 1724 is allocated</li> 1725 <li>when the breakpoint is reached you can then do a fine analysis of the 1726 allocation an step to see the condition resulting in the missing 1727 deallocation.</li> 1728</ol> 1729 1730<p>I used to use a commercial tool to debug libxml memory problems but after 1731noticing that it was not detecting memory leaks that simple mechanism was 1732used and proved extremely efficient until now.</p> 1733 1734<h3><a name="General4">General memory requirements</a></h3> 1735 1736<p>How much libxml memory require ? It's hard to tell in average it depends 1737of a number of things:</p> 1738<ul> 1739 <li>the parser itself should work in a fixed amout of memory, except for 1740 information maintained about the stacks of names and entities locations. 1741 The I/O and encoding handlers will probably account for a few KBytes. 1742 This is true for both the XML and HTML parser (though the HTML parser 1743 need more state).</li> 1744 <li>If you are generating the DOM tree then memory requirements will grow 1745 nearly lineary with the size of the data. In general for a balanced 1746 textual document the internal memory requirement is about 4 times the 1747 size of the UTF8 serialization of this document (exmple the XML-1.0 1748 recommendation is a bit more of 150KBytes and takes 650KBytes of main 1749 memory when parsed). Validation will add a amount of memory required for 1750 maintaining the external Dtd state which should be linear with the 1751 complexity of the content model defined by the Dtd</li> 1752 <li>If you don't care about the advanced features of libxml like 1753 validation, DOM, XPath or XPointer, but really need to work fixed memory 1754 requirements, then the SAX interface should be used.</li> 1755</ul> 1756 1757<p></p> 1758 1759<h2><a name="Encodings">Encodings support</a></h2> 1760 1761<p>Table of Content:</p> 1762<ol> 1763 <li><a href="encoding.html#What">What does internationalization support 1764 mean ?</a></li> 1765 <li><a href="encoding.html#internal">The internal encoding, how and 1766 why</a></li> 1767 <li><a href="encoding.html#implemente">How is it implemented ?</a></li> 1768 <li><a href="encoding.html#Default">Default supported encodings</a></li> 1769 <li><a href="encoding.html#extend">How to extend the existing 1770 support</a></li> 1771</ol> 1772 1773<h3><a name="What">What does internationalization support mean ?</a></h3> 1774 1775<p>XML was designed from the start to allow the support of any character set 1776by using Unicode. Any conformant XML parser has to support the UTF-8 and 1777UTF-16 default encodings which can both express the full unicode ranges. UTF8 1778is a variable length encoding whose greatest point are to resuse the same 1779emcoding for ASCII and to save space for Western encodings, but it is a bit 1780more complex to handle in practice. UTF-16 use 2 bytes per characters (and 1781sometimes combines two pairs), it makes implementation easier, but looks a 1782bit overkill for Western languages encoding. Moreover the XML specification 1783allows document to be encoded in other encodings at the condition that they 1784are clearly labelled as such. For example the following is a wellformed XML 1785document encoded in ISO-8859 1 and using accentuated letter that we French 1786likes for both markup and content:</p> 1787<pre><?xml version="1.0" encoding="ISO-8859-1"?> 1788<tr�s>l�</tr�s></pre> 1789 1790<p>Having internationalization support in libxml means the foolowing:</p> 1791<ul> 1792 <li>the document is properly parsed</li> 1793 <li>informations about it's encoding are saved</li> 1794 <li>it can be modified</li> 1795 <li>it can be saved in its original encoding</li> 1796 <li>it can also be saved in another encoding supported by libxml (for 1797 example straight UTF8 or even an ASCII form)</li> 1798</ul> 1799 1800<p>Another very important point is that the whole libxml API, with the 1801exception of a few routines to read with a specific encoding or save to a 1802specific encoding, is completely agnostic about the original encoding of the 1803document.</p> 1804 1805<p>It should be noted too that the HTML parser embedded in libxml now obbey 1806the same rules too, the following document will be (as of 2.2.2) handled in 1807an internationalized fashion by libxml too:</p> 1808<pre><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" 1809 "http://www.w3.org/TR/REC-html40/loose.dtd"> 1810<html lang="fr"> 1811<head> 1812 <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=ISO-8859-1"> 1813</head> 1814<body> 1815<p>W3C cr�e des standards pour le Web.</body> 1816</html></pre> 1817 1818<h3><a name="internal">The internal encoding, how and why</a></h3> 1819 1820<p>One of the core decision was to force all documents to be converted to a 1821default internal encoding, and that encoding to be UTF-8, here are the 1822rationale for those choices:</p> 1823<ul> 1824 <li>keeping the native encoding in the internal form would force the libxml 1825 users (or the code associated) to be fully aware of the encoding of the 1826 original document, for examples when adding a text node to a document, 1827 the content would have to be provided in the document encoding, i.e. the 1828 client code would have to check it before hand, make sure it's conformant 1829 to the encoding, etc ... Very hard in practice, though in some specific 1830 cases this may make sense.</li> 1831 <li>the second decision was which encoding. From the XML spec only UTF8 and 1832 UTF16 really makes sense as being the two only encodings for which there 1833 is amndatory support. UCS-4 (32 bits fixed size encoding) could be 1834 considered an intelligent choice too since it's a direct Unicode mapping 1835 support. I selected UTF-8 on the basis of efficiency and compatibility 1836 with surrounding software: 1837 <ul> 1838 <li>UTF-8 while a bit more complex to convert from/to (i.e. slightly 1839 more costly to import and export CPU wise) is also far more compact 1840 than UTF-16 (and UCS-4) for a majority of the documents I see it used 1841 for right now (RPM RDF catalogs, advogato data, various configuration 1842 file formats, etc.) and the key point for today's computer 1843 architecture is efficient uses of caches. If one nearly double the 1844 memory requirement to store the same amount of data, this will trash 1845 caches (main memory/external caches/internal caches) and my take is 1846 that this harms the system far more than the CPU requirements needed 1847 for the conversion to UTF-8</li> 1848 <li>Most of libxml version 1 users were using it with straight ASCII 1849 most of the time, doing the conversion with an internal encoding 1850 requiring all their code to be rewritten was a serious show-stopper 1851 for using UTF-16 or UCS-4.</li> 1852 <li>UTF-8 is being used as the de-facto internal encoding standard for 1853 related code like the <a href="http://www.pango.org/">pango</a> 1854 upcoming Gnome text widget, and a lot of Unix code (yep another place 1855 where Unix programmer base takes a different approach from Microsoft 1856 - they are using UTF-16)</li> 1857 </ul> 1858 </li> 1859</ul> 1860 1861<p>What does this mean in practice for the libxml user:</p> 1862<ul> 1863 <li>xmlChar, the libxml data type is a byte, those bytes must be assembled 1864 as UTF-8 valid strings. The proper way to terminate an xmlChar * string 1865 is simply to append 0 byte, as usual.</li> 1866 <li>One just need to make sure that when using chars outside the ASCII set, 1867 the values has been properly converted to UTF-8</li> 1868</ul> 1869 1870<h3><a name="implemente">How is it implemented ?</a></h3> 1871 1872<p>Let's describe how all this works within libxml, basically the I18N 1873(internationalization) support get triggered only during I/O operation, i.e. 1874when reading a document or saving one. Let's look first at the reading 1875sequence:</p> 1876<ol> 1877 <li>when a document is processed, we usually don't know the encoding, a 1878 simple heuristic allows to detect UTF-18 and UCS-4 from whose where the 1879 ASCII range (0-0x7F) maps with ASCII</li> 1880 <li>the xml declaration if available is parsed, including the encoding 1881 declaration. At that point, if the autodetected encoding is different 1882 from the one declared a call to xmlSwitchEncoding() is issued.</li> 1883 <li>If there is no encoding declaration, then the input has to be in either 1884 UTF-8 or UTF-16, if it is not then at some point when processing the 1885 input, the converter/checker of UTF-8 form will raise an encoding error. 1886 You may end-up with a garbled document, or no document at all ! Example: 1887 <pre>~/XML -> /xmllint err.xml 1888err.xml:1: error: Input is not proper UTF-8, indicate encoding ! 1889<tr�s>l�</tr�s> 1890 ^ 1891err.xml:1: error: Bytes: 0xE8 0x73 0x3E 0x6C 1892<tr�s>l�</tr�s> 1893 ^</pre> 1894 </li> 1895 <li>xmlSwitchEncoding() does an encoding name lookup, canonalize it, and 1896 then search the default registered encoding converters for that encoding. 1897 If it's not within the default set and iconv() support has been compiled 1898 it, it will ask iconv for such an encoder. If this fails then the parser 1899 will report an error and stops processing: 1900 <pre>~/XML -> /xmllint err2.xml 1901err2.xml:1: error: Unsupported encoding UnsupportedEnc 1902<?xml version="1.0" encoding="UnsupportedEnc"?> 1903 ^</pre> 1904 </li> 1905 <li>From that point the encoder process progressingly the input (it is 1906 plugged as a front-end to the I/O module) for that entity. It captures 1907 and convert on-the-fly the document to be parsed to UTF-8. The parser 1908 itself just does UTF-8 checking of this input and process it 1909 transparently. The only difference is that the encoding information has 1910 been added to the parsing context (more precisely to the input 1911 corresponding to this entity).</li> 1912 <li>The result (when using DOM) is an internal form completely in UTF-8 1913 with just an encoding information on the document node.</li> 1914</ol> 1915 1916<p>Ok then what's happen when saving the document (assuming you 1917colllected/built an xmlDoc DOM like structure) ? It depends on the function 1918called, xmlSaveFile() will just try to save in the original encoding, while 1919xmlSaveFileTo() and xmlSaveFileEnc() can optionally save to a given 1920encoding:</p> 1921<ol> 1922 <li>if no encoding is given, libxml will look for an encoding value 1923 associated to the document and if it exists will try to save to that 1924 encoding, 1925 <p>otherwise everything is written in the internal form, i.e. UTF-8</p> 1926 </li> 1927 <li>so if an encoding was specified, either at the API level or on the 1928 document, libxml will again canonalize the encoding name, lookup for a 1929 converter in the registered set or through iconv. If not found the 1930 function will return an error code</li> 1931 <li>the converter is placed before the I/O buffer layer, as another kind of 1932 buffer, then libxml will simply push the UTF-8 serialization to through 1933 that buffer, which will then progressively be converted and pushed onto 1934 the I/O layer.</li> 1935 <li>It is possible that the converter code fails on some input, for example 1936 trying to push an UTF-8 encoded chinese character through the UTF-8 to 1937 ISO-8859-1 converter won't work. Since the encoders are progressive they 1938 will just report the error and the number of bytes converted, at that 1939 point libxml will decode the offending character, remove it from the 1940 buffer and replace it with the associated charRef encoding &#123; and 1941 resume the convertion. This guarante that any document will be saved 1942 without losses (except for markup names where this is not legal, this is 1943 a problem in the current version, in pactice avoid using non-ascci 1944 characters for tags or attributes names @@). A special "ascii" encoding 1945 name is used to save documents to a pure ascii form can be used when 1946 portability is really crucial</li> 1947</ol> 1948 1949<p>Here is a few examples based on the same test document:</p> 1950<pre>~/XML -> /xmllint isolat1 1951<?xml version="1.0" encoding="ISO-8859-1"?> 1952<tr�s>l�</tr�s> 1953~/XML -> /xmllint --encode UTF-8 isolat1 1954<?xml version="1.0" encoding="UTF-8"?> 1955<très>l� �</très> 1956~/XML -> </pre> 1957 1958<p>The same processing is applied (and reuse most of the code) for HTML I18N 1959processing. Looking up and modifying the content encoding is a bit more 1960difficult since it is located in a <meta> tag under the <head>, 1961so a couple of functions htmlGetMetaEncoding() and htmlSetMetaEncoding() have 1962been provided. The parser also attempts to switch encoding on the fly when 1963detecting such a tag on input. Except for that the processing is the same 1964(and again reuses the same code).</p> 1965 1966<h3><a name="Default">Default supported encodings</a></h3> 1967 1968<p>libxml has a set of default converters for the following encodings 1969(located in encoding.c):</p> 1970<ol> 1971 <li>UTF-8 is supported by default (null handlers)</li> 1972 <li>UTF-16, both little and big endian</li> 1973 <li>ISO-Latin-1 (ISO-8859-1) covering most western languages</li> 1974 <li>ASCII, useful mostly for saving</li> 1975 <li>HTML, a specific handler for the conversion of UTF-8 to ASCII with HTML 1976 predefined entities like &copy; for the Copyright sign.</li> 1977</ol> 1978 1979<p>More over when compiled on an Unix platfor with iconv support the full set 1980of encodings supported by iconv can be instantly be used by libxml. On a 1981linux machine with glibc-2.1 the list of supported encodings and aliases fill 19823 full pages, and include UCS-4, the full set of ISO-Latin encodings, and the 1983various Japanese ones.</p> 1984 1985<h4>Encoding aliases</h4> 1986 1987<p>From 2.2.3, libxml has support to register encoding names aliases. The 1988goal is to be able to parse document whose encoding is supported but where 1989the name differs (for example from the default set of names accepted by 1990iconv). The following functions allow to register and handle new aliases for 1991existing encodings. Once registered libxml will automatically lookup the 1992aliases when handling a document:</p> 1993<ul> 1994 <li>int xmlAddEncodingAlias(const char *name, const char *alias);</li> 1995 <li>int xmlDelEncodingAlias(const char *alias);</li> 1996 <li>const char * xmlGetEncodingAlias(const char *alias);</li> 1997 <li>void xmlCleanupEncodingAliases(void);</li> 1998</ul> 1999 2000<h3><a name="extend">How to extend the existing support</a></h3> 2001 2002<p>Well adding support for new encoding, or overriding one of the encoders 2003(assuming it is buggy) should not be hard, just write an input and output 2004conversion routines to/from UTF-8, and register them using 2005xmlNewCharEncodingHandler(name, xxxToUTF8, UTF8Toxxx), and they will be 2006called automatically if the parser(s) encounter such an encoding name 2007(register it uppercase, this will help). The description of the encoders, 2008their arguments and expected return values are described in the encoding.h 2009header.</p> 2010 2011<p>A quick note on the topic of subverting the parser to use a different 2012internal encoding than UTF-8, in some case people will absolutely want to 2013keep the internal encoding different, I think it's still possible (but the 2014encoding must be compliant with ASCII on the same subrange) though I didn't 2015tried it. The key is to override the default conversion routines (by 2016registering null encoders/decoders for your charsets), and bypass the UTF-8 2017checking of the parser by setting the parser context charset 2018(ctxt->charset) to something different than XML_CHAR_ENCODING_UTF8, but 2019there is no guarantee taht this will work. You may also have some troubles 2020saving back.</p> 2021 2022<p>Basically proper I18N support is important, this requires at least 2023libxml-2.0.0, but a lot of features and corrections are really available only 2024starting 2.2.</p> 2025 2026<h2><a name="IO">I/O Interfaces</a></h2> 2027 2028<p>Table of Content:</p> 2029<ol> 2030 <li><a href="#General1">General overview</a></li> 2031 <li><a href="#basic">The basic buffer type</a></li> 2032 <li><a href="#Input">Input I/O handlers</a></li> 2033 <li><a href="#Output">Output I/O handlers</a></li> 2034 <li><a href="#entities">The entities loader</a></li> 2035 <li><a href="#Example2">Example of customized I/O</a></li> 2036</ol> 2037 2038<h3><a name="General1">General overview</a></h3> 2039 2040<p>The module <code><a 2041href="http://xmlsoft.org/html/libxml-xmlio.html">xmlIO.h</a></code> provides 2042the interfaces to the libxml I/O system. This consists of 4 main parts:</p> 2043<ul> 2044 <li>Entities loader, this is a routine which tries to fetch the entities 2045 (files) based on their PUBLIC and SYSTEM identifiers. The default loader 2046 don't look at the public identifier since libxml do not maintain a 2047 catalog. You can redefine you own entity loader by using 2048 <code>xmlGetExternalEntityLoader()</code> and 2049 <code>xmlSetExternalEntityLoader()</code>. <a href="#entities">Check the 2050 example</a>.</li> 2051 <li>Input I/O buffers which are a commodity structure used by the parser(s) 2052 input layer to handle fetching the informations to feed the parser. This 2053 provides buffering and is also a placeholder where the encoding 2054 convertors to UTF8 are piggy-backed.</li> 2055 <li>Output I/O buffers are similar to the Input ones and fulfill similar 2056 task but when generating a serialization from a tree.</li> 2057 <li>A mechanism to register sets of I/O callbacks and associate them with 2058 specific naming schemes like the protocol part of the URIs. 2059 <p>This affect the default I/O operations and allows to use specific I/O 2060 handlers for certain names.</p> 2061 </li> 2062</ul> 2063 2064<p>The general mechanism used when loading http://rpmfind.net/xml.html for 2065example in the HTML parser is the following:</p> 2066<ol> 2067 <li>The default entity loader calls <code>xmlNewInputFromFile()</code> with 2068 the parsing context and the URI string.</li> 2069 <li>the URI string is checked against the existing registered handlers 2070 using their match() callback function, if the HTTP module was compiled 2071 in, it is registered and its match() function will succeeds</li> 2072 <li>the open() function of the handler is called and if successful will 2073 return an I/O Input buffer</li> 2074 <li>the parser will the start reading from this buffer and progressively 2075 fetch information from the resource, calling the read() function of the 2076 handler until the resource is exhausted</li> 2077 <li>if an encoding change is detected it will be installed on the input 2078 buffer, providing buffering and efficient use of the conversion 2079 routines</li> 2080 <li>once the parser has finished, the close() function of the handler is 2081 called once and the Input buffer and associed resources are 2082 deallocated.</li> 2083</ol> 2084 2085<p>The user defined callbacks are checked first to allow overriding of the 2086default libxml I/O routines.</p> 2087 2088<h3><a name="basic">The basic buffer type</a></h3> 2089 2090<p>All the buffer manipulation handling is done using the 2091<code>xmlBuffer</code> type define in <code><a 2092href="http://xmlsoft.org/html/libxml-tree.html">tree.h</a> </code>which is a 2093resizable memory buffer. The buffer allocation strategy can be selected to be 2094either best-fit or use an exponential doubling one (CPU vs. memory use 2095tradeoff). The values are <code>XML_BUFFER_ALLOC_EXACT</code> and 2096<code>XML_BUFFER_ALLOC_DOUBLEIT</code>, and can be set individually or on a 2097system wide basis using <code>xmlBufferSetAllocationScheme()</code>. A number 2098of functions allows to manipulate buffers with names starting with the 2099<code>xmlBuffer...</code> prefix.</p> 2100 2101<h3><a name="Input">Input I/O handlers</a></h3> 2102 2103<p>An Input I/O handler is a simple structure 2104<code>xmlParserInputBuffer</code> containing a context associated to the 2105resource (file descriptor, or pointer to a protocol handler), the read() and 2106close() callbacks to use and an xmlBuffer. And extra xmlBuffer and a charset 2107encoding handler are also present to support charset conversion when 2108needed.</p> 2109 2110<h3><a name="Output">Output I/O handlers</a></h3> 2111 2112<p>An Output handler <code>xmlOutputBuffer</code> is completely similar to an 2113Input one except the callbacks are write() and close().</p> 2114 2115<h3><a name="entities">The entities loader</a></h3> 2116 2117<p>The entity loader resolves requests for new entities and create inputs for 2118the parser. Creating an input from a filename or an URI string is done 2119through the xmlNewInputFromFile() routine. The default entity loader do not 2120handle the PUBLIC identifier associated with an entity (if any). So it just 2121calls xmlNewInputFromFile() with the SYSTEM identifier (which is mandatory in 2122XML).</p> 2123 2124<p>If you want to hook up a catalog mechanism then you simply need to 2125override the default entity loader, here is an example:</p> 2126<pre>#include <libxml/xmlIO.h> 2127 2128xmlExternalEntityLoader defaultLoader = NULL; 2129 2130xmlParserInputPtr 2131xmlMyExternalEntityLoader(const char *URL, const char *ID, 2132 xmlParserCtxtPtr ctxt) { 2133 xmlParserInputPtr ret; 2134 const char *fileID = NULL; 2135 /* lookup for the fileID depending on ID */ 2136 2137 ret = xmlNewInputFromFile(ctxt, fileID); 2138 if (ret != NULL) 2139 return(ret); 2140 if (defaultLoader != NULL) 2141 ret = defaultLoader(URL, ID, ctxt); 2142 return(ret); 2143} 2144 2145int main(..) { 2146 ... 2147 2148 /* 2149 * Install our own entity loader 2150 */ 2151 defaultLoader = xmlGetExternalEntityLoader(); 2152 xmlSetExternalEntityLoader(xmlMyExternalEntityLoader); 2153 2154 ... 2155}</pre> 2156 2157<h3><a name="Example2">Example of customized I/O</a></h3> 2158 2159<p>This example come from <a href="http://xmlsoft.org/messages/0708.html">a 2160real use case</a>, xmlDocDump() closes the FILE * passed by the application 2161and this was a problem. The <a 2162href="http://xmlsoft.org/messages/0711.html">solution</a> was to redefine a 2163new output handler with the closing call deactivated:</p> 2164<ol> 2165 <li>First define a new I/O ouput allocator where the output don't close the 2166 file: 2167 <pre>xmlOutputBufferPtr 2168xmlOutputBufferCreateOwn(FILE *file, xmlCharEncodingHandlerPtr encoder) { 2169����xmlOutputBufferPtr ret; 2170���� 2171����if (xmlOutputCallbackInitialized == 0) 2172��������xmlRegisterDefaultOutputCallbacks(); 2173 2174����if (file == NULL) return(NULL); 2175����ret = xmlAllocOutputBuffer(encoder); 2176����if (ret != NULL) { 2177��������ret->context = file; 2178��������ret->writecallback = xmlFileWrite; 2179��������ret->closecallback = NULL; /* No close callback */ 2180����} 2181����return(ret); <br> 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199} </pre> 2200 </li> 2201 <li>And then use it to save the document: 2202 <pre>FILE *f; 2203xmlOutputBufferPtr output; 2204xmlDocPtr doc; 2205int res; 2206 2207f = ... 2208doc = .... 2209 2210output = xmlOutputBufferCreateOwn(f, NULL); 2211res = xmlSaveFileTo(output, doc, NULL); 2212 </pre> 2213 </li> 2214</ol> 2215 2216<h2><a name="Catalog">Catalog support</a></h2> 2217 2218<p>Table of Content:</p> 2219<ol> 2220 <li><a href="General2">General overview</a></li> 2221 <li><a href="#definition">The definition</a></li> 2222 <li><a href="#Simple">Using catalogs</a></li> 2223 <li><a href="#Some">Some examples</a></li> 2224 <li><a href="#reference">How to tune catalog usage</a></li> 2225 <li><a href="#validate">How to debug catalog processing</a></li> 2226 <li><a href="#Declaring">How to create and maintain catalogs</a></li> 2227 <li><a href="#implemento">The implementor corner quick review of the 2228 API</a></li> 2229 <li><a href="#Other">Other resources</a></li> 2230</ol> 2231 2232<h3><a name="General2">General overview</a></h3> 2233 2234<p>What is a catalog? Basically it's a lookup mechanism used when an entity 2235(a file or a remote resource) references another entity. The catalog lookup 2236is inserted between the moment the reference is recognized by the software 2237(XML parser, stylesheet processing, or even images referenced for inclusion 2238in a rendering) and the time where loading that resource is actually 2239started.</p> 2240 2241<p>It is basically used for 3 things:</p> 2242<ul> 2243 <li>mapping from "logical" names, the public identifiers and a more 2244 concrete name usable for download (and URI). For example it can associate 2245 the logical name 2246 <p>"-//OASIS//DTD DocBook XML V4.1.2//EN"</p> 2247 <p>of the DocBook 4.1.2 XML DTD with the actual URL where it can be 2248 downloaded</p> 2249 <p>http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd</p> 2250 </li> 2251 <li>remapping from a given URL to another one, like an HTTP indirection 2252 saying that 2253 <p>"http://www.oasis-open.org/committes/tr.xsl"</p> 2254 <p>should really be looked at</p> 2255 <p>"http://www.oasis-open.org/committes/entity/stylesheets/base/tr.xsl"</p> 2256 </li> 2257 <li>providing a local cache mechanism allowing to load the entities 2258 associated to public identifiers or remote resources, this is a really 2259 important feature for any significant deployment of XML or SGML since it 2260 allows to avoid the aleas and delays associated to fetching remote 2261 resources.</li> 2262</ul> 2263 2264<h3><a name="definition">The definitions</a></h3> 2265 2266<p>Libxml, as of 2.4.3 implements 2 kind of catalogs:</p> 2267<ul> 2268 <li>the older SGML catalogs, the official spec is SGML Open Technical 2269 Resolution TR9401:1997, but is better understood by reading <a 2270 href="http://www.jclark.com/sp/catalog.htm">the SP Catalog page</a> from 2271 James Clark. This is relatively old and not the preferred mode of 2272 operation of libxml.</li> 2273 <li><a href="http://www.oasis-open.org/committees/entity/spec.html">XML 2274 Catalogs</a> 2275 is far more flexible, more recent, uses an XML syntax and should scale 2276 quite better. This is the default option of libxml.</li> 2277</ul> 2278 2279<p></p> 2280 2281<h3><a name="Simple">Using catalog</a></h3> 2282 2283<p>In a normal environment libxml will by default check the presence of a 2284catalog in /etc/xml/catalog, and assuming it has been correctly populated, 2285the processing is completely transparent to the document user. To take a 2286concrete example, suppose you are authoring a DocBook document, this one 2287starts with the following DOCTYPE definition:</p> 2288<pre><?xml version='1.0'?> 2289<!DOCTYPE book PUBLIC "-//Norman Walsh//DTD DocBk XML V3.1.4//EN" 2290 "http://nwalsh.com/docbook/xml/3.1.4/db3xml.dtd"></pre> 2291 2292<p>When validating the document with libxml, the catalog will be 2293automatically consulted to lookup the public identifier "-//Norman Walsh//DTD 2294DocBk XML V3.1.4//EN" and the system identifier 2295"http://nwalsh.com/docbook/xml/3.1.4/db3xml.dtd", and if these entities have 2296been installed on your system and the catalogs actually point to them, libxml 2297will fetch them from the local disk.</p> 2298 2299<p style="font-size: 10pt"><strong>Note</strong>: Really don't use this 2300DOCTYPE example it's a really old version, but is fine as an example.</p> 2301 2302<p>Libxml will check the catalog each time that it is requested to load an 2303entity, this includes DTD, external parsed entities, stylesheets, etc ... If 2304your system is correctly configured all the authoring phase and processing 2305should use only local files, even if your document stays portable because it 2306uses the canonical public and system ID, referencing the remote document.</p> 2307 2308<h3><a name="Some">Some examples:</a></h3> 2309 2310<p>Here is a couple of fragments from XML Catalogs used in libxml early 2311regression tests in <code>test/catalogs</code> :</p> 2312<pre><?xml version="1.0"?> 2313<!DOCTYPE catalog PUBLIC 2314 "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" 2315 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 2316<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"> 2317 <public publicId="-//OASIS//DTD DocBook XML V4.1.2//EN" 2318 uri="http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"/> 2319...</pre> 2320 2321<p>This is the beginning of a catalog for DocBook 4.1.2, XML Catalogs are 2322written in XML, there is a specific namespace for catalog elements 2323"urn:oasis:names:tc:entity:xmlns:xml:catalog". The first entry in this 2324catalog is a <code>public</code> mapping it allows to associate a Public 2325Identifier with an URI.</p> 2326<pre>... 2327 <rewriteSystem systemIdStartString="http://www.oasis-open.org/docbook/" 2328 rewritePrefix="file:///usr/share/xml/docbook/"/> 2329...</pre> 2330 2331<p>A <code>rewriteSystem</code> is a very powerful instruction, it says that 2332any URI starting with a given prefix should be looked at another URI 2333constructed by replacing the prefix with an new one. In effect this acts like 2334a cache system for a full area of the Web. In practice it is extremely useful 2335with a file prefix if you have installed a copy of those resources on your 2336local system.</p> 2337<pre>... 2338<delegatePublic publicIdStartString="-//OASIS//DTD XML Catalog //" 2339 catalog="file:///usr/share/xml/docbook.xml"/> 2340<delegatePublic publicIdStartString="-//OASIS//ENTITIES DocBook XML" 2341 catalog="file:///usr/share/xml/docbook.xml"/> 2342<delegatePublic publicIdStartString="-//OASIS//DTD DocBook XML" 2343 catalog="file:///usr/share/xml/docbook.xml"/> 2344<delegateSystem systemIdStartString="http://www.oasis-open.org/docbook/" 2345 catalog="file:///usr/share/xml/docbook.xml"/> 2346<delegateURI uriStartString="http://www.oasis-open.org/docbook/" 2347 catalog="file:///usr/share/xml/docbook.xml"/> 2348...</pre> 2349 2350<p>Delegation is the core features which allows to build a tree of catalogs, 2351easier to maintain than a single catalog, based on Public Identifier, System 2352Identifier or URI prefixes it instructs the catalog software to look up 2353entries in another resource. This feature allow to build hierarchies of 2354catalogs, the set of entries presented should be sufficient to redirect the 2355resolution of all DocBook references to the specific catalog in 2356<code>/usr/share/xml/docbook.xml</code> this one in turn could delegate all 2357references for DocBook 4.2.1 to a specific catalog installed at the same time 2358as the DocBook resources on the local machine.</p> 2359 2360<h3><a name="reference">How to tune catalog usage:</a></h3> 2361 2362<p>The user can change the default catalog behaviour by redirecting queries 2363to its own set of catalogs, this can be done by setting the 2364<code>XML_CATALOG_FILES</code> environment variable to a list of catalogs, an 2365empty one should deactivate loading the default <code>/etc/xml/catalog</code> 2366default catalog</p> 2367 2368<h3><a name="validate">How to debug catalog processing:</a></h3> 2369 2370<p>Setting up the <code>XML_DEBUG_CATALOG</code> environment variable will 2371make libxml output debugging informations for each catalog operations, for 2372example:</p> 2373<pre>orchis:~/XML -> xmllint --memory --noout test/ent2 2374warning: failed to load external entity "title.xml" 2375orchis:~/XML -> export XML_DEBUG_CATALOG= 2376orchis:~/XML -> xmllint --memory --noout test/ent2 2377Failed to parse catalog /etc/xml/catalog 2378Failed to parse catalog /etc/xml/catalog 2379warning: failed to load external entity "title.xml" 2380Catalogs cleanup 2381orchis:~/XML -> </pre> 2382 2383<p>The test/ent2 references an entity, running the parser from memory makes 2384the base URI unavailable and the the "title.xml" entity cannot be loaded. 2385Setting up the debug environment variable allows to detect that an attempt is 2386made to load the <code>/etc/xml/catalog</code> but since it's not present the 2387resolution fails.</p> 2388 2389<p>But the most advanced way to debug XML catalog processing is to use the 2390<strong>xmlcatalog</strong> command shipped with libxml2, it allows to load 2391catalogs and make resolution queries to see what is going on. This is also 2392used for the regression tests:</p> 2393<pre>orchis:~/XML -> /xmlcatalog test/catalogs/docbook.xml \ 2394 "-//OASIS//DTD DocBook XML V4.1.2//EN" 2395http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd 2396orchis:~/XML -> </pre> 2397 2398<p>For debugging what is going on, adding one -v flags increase the verbosity 2399level to indicate the processing done (adding a second flag also indicate 2400what elements are recognized at parsing):</p> 2401<pre>orchis:~/XML -> /xmlcatalog -v test/catalogs/docbook.xml \ 2402 "-//OASIS//DTD DocBook XML V4.1.2//EN" 2403Parsing catalog test/catalogs/docbook.xml's content 2404Found public match -//OASIS//DTD DocBook XML V4.1.2//EN 2405http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd 2406Catalogs cleanup 2407orchis:~/XML -> </pre> 2408 2409<p>A shell interface is also available to debug and process multiple queries 2410(and for regression tests):</p> 2411<pre>orchis:~/XML -> /xmlcatalog -shell test/catalogs/docbook.xml \ 2412 "-//OASIS//DTD DocBook XML V4.1.2//EN" 2413> help 2414Commands available: 2415public PublicID: make a PUBLIC identifier lookup 2416system SystemID: make a SYSTEM identifier lookup 2417resolve PublicID SystemID: do a full resolver lookup 2418add 'type' 'orig' 'replace' : add an entry 2419del 'values' : remove values 2420dump: print the current catalog state 2421debug: increase the verbosity level 2422quiet: decrease the verbosity level 2423exit: quit the shell 2424> public "-//OASIS//DTD DocBook XML V4.1.2//EN" 2425http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd 2426> quit 2427orchis:~/XML -> </pre> 2428 2429<p>This should be sufficient for most debugging purpose, this was actually 2430used heavily to debug the XML Catalog implementation itself.</p> 2431 2432<h3><a name="Declaring">How to create and maintain</a> catalogs:</h3> 2433 2434<p>Basically XML Catalogs are XML files, you can either use XML tools to 2435manage them or use <strong>xmlcatalog</strong> for this. The basic step is 2436to create a catalog the -create option provide this facility:</p> 2437<pre>orchis:~/XML -> /xmlcatalog --create tst.xml 2438<?xml version="1.0"?> 2439<!DOCTYPE catalog PUBLIC "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" 2440 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 2441<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"/> 2442orchis:~/XML -> </pre> 2443 2444<p>By default xmlcatalog does not overwrite the original catalog and save the 2445result on the standard output, this can be overridden using the -noout 2446option. The <code>-add</code> command allows to add entries in the 2447catalog:</p> 2448<pre>orchis:~/XML -> /xmlcatalog --noout --create --add "public" \ 2449 "-//OASIS//DTD DocBook XML V4.1.2//EN" \ 2450 http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd tst.xml 2451orchis:~/XML -> cat tst.xml 2452<?xml version="1.0"?> 2453<!DOCTYPE catalog PUBLIC "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" \ 2454 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 2455<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"> 2456<public publicId="-//OASIS//DTD DocBook XML V4.1.2//EN" 2457 uri="http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"/> 2458</catalog> 2459orchis:~/XML -> </pre> 2460 2461<p>The <code>-add</code> option will always take 3 parameters even if some of 2462the XML Catalog constructs (like nextCatalog) will have only a single 2463argument, just pass a third empty string, it will be ignored.</p> 2464 2465<p>Similarly the <code>-del</code> option remove matching entries from the 2466catalog:</p> 2467<pre>orchis:~/XML -> /xmlcatalog --del \ 2468 "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" tst.xml 2469<?xml version="1.0"?> 2470<!DOCTYPE catalog PUBLIC "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" 2471 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 2472<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"/> 2473orchis:~/XML -> </pre> 2474 2475<p>The catalog is now empty. Note that the matching of <code>-del</code> is 2476exact and would have worked in a similar fashion with the Public ID 2477string.</p> 2478 2479<p>This is rudimentary but should be sufficient to manage a not too complex 2480catalog tree of resources.</p> 2481 2482<h3><a name="implemento">The implementor corner quick review of the 2483API:</a></h3> 2484 2485<p>First, and like for every other module of libxml, there is an 2486automatically generated <a href="html/libxml-catalog.html">API page for 2487catalog support</a>.</p> 2488 2489<p>The header for the catalog interfaces should be included as:</p> 2490<pre>#include <libxml/catalog.h></pre> 2491 2492<p>The API is voluntarily kept very simple. First it is not obvious that 2493applications really need access to it since it is the default behaviour of 2494libxml (Note: it is possible to completely override libxml default catalog by 2495using <a href="html/libxml-parser.html">xmlSetExternalEntityLoader</a> to 2496plug an application specific resolver).</p> 2497 2498<p>Basically libxml support 2 catalog lists:</p> 2499<ul> 2500 <li>the default one, global shared by all the application</li> 2501 <li>a per-document catalog, this one is built if the document uses the 2502 <code>oasis-xml-catalog</code> PIs to specify its own catalog list, it is 2503 associated to the parser context and destroyed when the parsing context 2504 is destroyed.</li> 2505</ul> 2506 2507<p>the document one will be used first if it exists.</p> 2508 2509<h4>Initialization routines:</h4> 2510 2511<p>xmlInitializeCatalog(), xmlLoadCatalog() and xmlLoadCatalogs() should be 2512used at startup to initialize the catalog, if the catalog should be 2513initialized with specific values xmlLoadCatalog() or xmlLoadCatalogs() 2514should be called before xmlInitializeCatalog() which would otherwise do a 2515default initialization first.</p> 2516 2517<p>The xmlCatalogAddLocal() call is used by the parser to grow the document 2518own catalog list if needed.</p> 2519 2520<h4>Preferences setup:</h4> 2521 2522<p>The XML Catalog spec requires the possibility to select default 2523preferences between public and system delegation, 2524xmlCatalogSetDefaultPrefer() allows this, xmlCatalogSetDefaults() and 2525xmlCatalogGetDefaults() allow to control if XML Catalogs resolution should 2526be forbidden, allowed for global catalog, for document catalog or both, the 2527default is to allow both.</p> 2528 2529<p>And of course xmlCatalogSetDebug() allows to generate debug messages 2530(through the xmlGenericError() mechanism).</p> 2531 2532<h4>Querying routines:</h4> 2533 2534<p>xmlCatalogResolve(), xmlCatalogResolveSystem(), xmlCatalogResolvePublic() 2535and xmlCatalogResolveURI() are relatively explicit if you read the XML 2536Catalog specification they correspond to section 7 algorithms, they should 2537also work if you have loaded an SGML catalog with a simplified semantic.</p> 2538 2539<p>xmlCatalogLocalResolve() and xmlCatalogLocalResolveURI() are the same but 2540operate on the document catalog list</p> 2541 2542<h4>Cleanup and Miscellaneous:</h4> 2543 2544<p>xmlCatalogCleanup() free-up the global catalog, xmlCatalogFreeLocal() is 2545the per-document equivalent.</p> 2546 2547<p>xmlCatalogAdd() and xmlCatalogRemove() are used to dynamically modify the 2548first catalog in the global list, and xmlCatalogDump() allows to dump a 2549catalog state, those routines are primarily designed for xmlcatalog, I'm not 2550sure that exposing more complex interfaces (like navigation ones) would be 2551really useful.</p> 2552 2553<p>The xmlParseCatalogFile() is a function used to load XML Catalog files, 2554it's similar as xmlParseFile() except it bypass all catalog lookups, it's 2555provided because this functionality may be useful for client tools.</p> 2556 2557<h4>threaded environments:</h4> 2558 2559<p>Since the catalog tree is built progressively, some care has been taken to 2560try to avoid troubles in multithreaded environments. The code is now thread 2561safe assuming that the libxml library has been compiled with threads 2562support.</p> 2563 2564<p></p> 2565 2566<h3><a name="Other">Other resources</a></h3> 2567 2568<p>The XML Catalog specification is relatively recent so there isn't much 2569literature to point at:</p> 2570<ul> 2571 <li>You can find an good rant from Norm Walsh about <a 2572 href="http://www.arbortext.com/Think_Tank/XML_Resources/Issue_Three/issue_three.html">the 2573 need for catalogs</a>, it provides a lot of context informations even if 2574 I don't agree with everything presented.</li> 2575 <li>An <a href="http://home.ccil.org/~cowan/XML/XCatalog.html">old XML 2576 catalog proposal</a> from John Cowan</li> 2577 <li>The <a href="http://www.rddl.org/">Resource Directory Description 2578 Language</a> (RDDL) another catalog system but more oriented toward 2579 providing metadata for XML namespaces.</li> 2580 <li>the page from the OASIS Technical <a 2581 href="http://www.oasis-open.org/committees/entity/">Committee on Entity 2582 Resolution</a> who maintains XML Catalog, you will find pointers to the 2583 specification update, some background and pointers to others tools 2584 providing XML Catalog support</li> 2585 <li>Here is a <a href="buildDocBookCatalog">shell script</a> to generate 2586 XML Catalogs for DocBook 4.1.2 . If it can write to the /etc/xml/ 2587 directory, it will set-up /etc/xml/catalog and /etc/xml/docbook based on 2588 the resources found on the system. Otherwise it will just create 2589 ~/xmlcatalog and ~/dbkxmlcatalog and doing: 2590 <p><code> export XMLCATALOG=$HOME/xmlcatalog</code></p> 2591 <p>should allow to process DocBook documentations without requiring 2592 network accesses for the DTd or stylesheets</p> 2593 </li> 2594 <li>I have uploaded <a href="ftp://xmlsoft.org/test/dbk412catalog.tar.gz">a 2595 small tarball</a> containing XML Catalogs for DocBook 4.1.2 which seems 2596 to work fine for me too</li> 2597 <li>The <a href="http://www.xmlsoft.org/xmlcatalog_man.html">xmlcatalog 2598 manual page</a></li> 2599</ul> 2600 2601<p>If you have suggestions for corrections or additions, simply contact 2602me:</p> 2603 2604<h2><a name="library">The parser interfaces</a></h2> 2605 2606<p>This section is directly intended to help programmers getting bootstrapped 2607using the XML library from the C language. It is not intended to be 2608extensive. I hope the automatically generated documents will provide the 2609completeness required, but as a separate set of documents. The interfaces of 2610the XML library are by principle low level, there is nearly zero abstraction. 2611Those interested in a higher level API should <a href="#DOM">look at 2612DOM</a>.</p> 2613 2614<p>The <a href="html/libxml-parser.html">parser interfaces for XML</a> are 2615separated from the <a href="html/libxml-htmlparser.html">HTML parser 2616interfaces</a>. Let's have a look at how the XML parser can be called:</p> 2617 2618<h3><a name="Invoking">Invoking the parser : the pull method</a></h3> 2619 2620<p>Usually, the first thing to do is to read an XML input. The parser accepts 2621documents either from in-memory strings or from files. The functions are 2622defined in "parser.h":</p> 2623<dl> 2624 <dt><code>xmlDocPtr xmlParseMemory(char *buffer, int size);</code></dt> 2625 <dd><p>Parse a null-terminated string containing the document.</p> 2626 </dd> 2627</dl> 2628<dl> 2629 <dt><code>xmlDocPtr xmlParseFile(const char *filename);</code></dt> 2630 <dd><p>Parse an XML document contained in a (possibly compressed) 2631 file.</p> 2632 </dd> 2633</dl> 2634 2635<p>The parser returns a pointer to the document structure (or NULL in case of 2636failure).</p> 2637 2638<h3 id="Invoking1">Invoking the parser: the push method</h3> 2639 2640<p>In order for the application to keep the control when the document is 2641being fetched (which is common for GUI based programs) libxml provides a push 2642interface, too, as of version 1.8.3. Here are the interface functions:</p> 2643<pre>xmlParserCtxtPtr xmlCreatePushParserCtxt(xmlSAXHandlerPtr sax, 2644 void *user_data, 2645 const char *chunk, 2646 int size, 2647 const char *filename); 2648int xmlParseChunk (xmlParserCtxtPtr ctxt, 2649 const char *chunk, 2650 int size, 2651 int terminate);</pre> 2652 2653<p>and here is a simple example showing how to use the interface:</p> 2654<pre> FILE *f; 2655 2656 f = fopen(filename, "r"); 2657 if (f != NULL) { 2658 int res, size = 1024; 2659 char chars[1024]; 2660 xmlParserCtxtPtr ctxt; 2661 2662 res = fread(chars, 1, 4, f); 2663 if (res > 0) { 2664 ctxt = xmlCreatePushParserCtxt(NULL, NULL, 2665 chars, res, filename); 2666 while ((res = fread(chars, 1, size, f)) > 0) { 2667 xmlParseChunk(ctxt, chars, res, 0); 2668 } 2669 xmlParseChunk(ctxt, chars, 0, 1); 2670 doc = ctxt->myDoc; 2671 xmlFreeParserCtxt(ctxt); 2672 } 2673 }</pre> 2674 2675<p>The HTML parser embedded into libxml also has a push interface; the 2676functions are just prefixed by "html" rather than "xml".</p> 2677 2678<h3 id="Invoking2">Invoking the parser: the SAX interface</h3> 2679 2680<p>The tree-building interface makes the parser memory-hungry, first loading 2681the document in memory and then building the tree itself. Reading a document 2682without building the tree is possible using the SAX interfaces (see SAX.h and 2683<a href="http://www.daa.com.au/~james/gnome/xml-sax/xml-sax.html">James 2684Henstridge's documentation</a>). Note also that the push interface can be 2685limited to SAX: just use the two first arguments of 2686<code>xmlCreatePushParserCtxt()</code>.</p> 2687 2688<h3><a name="Building">Building a tree from scratch</a></h3> 2689 2690<p>The other way to get an XML tree in memory is by building it. Basically 2691there is a set of functions dedicated to building new elements. (These are 2692also described in <libxml/tree.h>.) For example, here is a piece of 2693code that produces the XML document used in the previous examples:</p> 2694<pre> #include <libxml/tree.h> 2695 xmlDocPtr doc; 2696 xmlNodePtr tree, subtree; 2697 2698 doc = xmlNewDoc("1.0"); 2699 doc->children = xmlNewDocNode(doc, NULL, "EXAMPLE", NULL); 2700 xmlSetProp(doc->children, "prop1", "gnome is great"); 2701 xmlSetProp(doc->children, "prop2", "& linux too"); 2702 tree = xmlNewChild(doc->children, NULL, "head", NULL); 2703 subtree = xmlNewChild(tree, NULL, "title", "Welcome to Gnome"); 2704 tree = xmlNewChild(doc->children, NULL, "chapter", NULL); 2705 subtree = xmlNewChild(tree, NULL, "title", "The Linux adventure"); 2706 subtree = xmlNewChild(tree, NULL, "p", "bla bla bla ..."); 2707 subtree = xmlNewChild(tree, NULL, "image", NULL); 2708 xmlSetProp(subtree, "href", "linus.gif");</pre> 2709 2710<p>Not really rocket science ...</p> 2711 2712<h3><a name="Traversing">Traversing the tree</a></h3> 2713 2714<p>Basically by <a href="html/libxml-tree.html">including "tree.h"</a> your 2715code has access to the internal structure of all the elements of the tree. 2716The names should be somewhat simple like <strong>parent</strong>, 2717<strong>children</strong>, <strong>next</strong>, <strong>prev</strong>, 2718<strong>properties</strong>, etc... For example, still with the previous 2719example:</p> 2720<pre><code>doc->children->children->children</code></pre> 2721 2722<p>points to the title element,</p> 2723<pre>doc->children->children->next->children->children</pre> 2724 2725<p>points to the text node containing the chapter title "The Linux 2726adventure".</p> 2727 2728<p><strong>NOTE</strong>: XML allows <em>PI</em>s and <em>comments</em> to be 2729present before the document root, so <code>doc->children</code> may point 2730to an element which is not the document Root Element; a function 2731<code>xmlDocGetRootElement()</code> was added for this purpose.</p> 2732 2733<h3><a name="Modifying">Modifying the tree</a></h3> 2734 2735<p>Functions are provided for reading and writing the document content. Here 2736is an excerpt from the <a href="html/libxml-tree.html">tree API</a>:</p> 2737<dl> 2738 <dt><code>xmlAttrPtr xmlSetProp(xmlNodePtr node, const xmlChar *name, const 2739 xmlChar *value);</code></dt> 2740 <dd><p>This sets (or changes) an attribute carried by an ELEMENT node. 2741 The value can be NULL.</p> 2742 </dd> 2743</dl> 2744<dl> 2745 <dt><code>const xmlChar *xmlGetProp(xmlNodePtr node, const xmlChar 2746 *name);</code></dt> 2747 <dd><p>This function returns a pointer to new copy of the property 2748 content. Note that the user must deallocate the result.</p> 2749 </dd> 2750</dl> 2751 2752<p>Two functions are provided for reading and writing the text associated 2753with elements:</p> 2754<dl> 2755 <dt><code>xmlNodePtr xmlStringGetNodeList(xmlDocPtr doc, const xmlChar 2756 *value);</code></dt> 2757 <dd><p>This function takes an "external" string and converts it to one 2758 text node or possibly to a list of entity and text nodes. All 2759 non-predefined entity references like &Gnome; will be stored 2760 internally as entity nodes, hence the result of the function may not be 2761 a single node.</p> 2762 </dd> 2763</dl> 2764<dl> 2765 <dt><code>xmlChar *xmlNodeListGetString(xmlDocPtr doc, xmlNodePtr list, int 2766 inLine);</code></dt> 2767 <dd><p>This function is the inverse of 2768 <code>xmlStringGetNodeList()</code>. It generates a new string 2769 containing the content of the text and entity nodes. Note the extra 2770 argument inLine. If this argument is set to 1, the function will expand 2771 entity references. For example, instead of returning the &Gnome; 2772 XML encoding in the string, it will substitute it with its value (say, 2773 "GNU Network Object Model Environment").</p> 2774 </dd> 2775</dl> 2776 2777<h3><a name="Saving">Saving a tree</a></h3> 2778 2779<p>Basically 3 options are possible:</p> 2780<dl> 2781 <dt><code>void xmlDocDumpMemory(xmlDocPtr cur, xmlChar**mem, int 2782 *size);</code></dt> 2783 <dd><p>Returns a buffer into which the document has been saved.</p> 2784 </dd> 2785</dl> 2786<dl> 2787 <dt><code>extern void xmlDocDump(FILE *f, xmlDocPtr doc);</code></dt> 2788 <dd><p>Dumps a document to an open file descriptor.</p> 2789 </dd> 2790</dl> 2791<dl> 2792 <dt><code>int xmlSaveFile(const char *filename, xmlDocPtr cur);</code></dt> 2793 <dd><p>Saves the document to a file. In this case, the compression 2794 interface is triggered if it has been turned on.</p> 2795 </dd> 2796</dl> 2797 2798<h3><a name="Compressio">Compression</a></h3> 2799 2800<p>The library transparently handles compression when doing file-based 2801accesses. The level of compression on saves can be turned on either globally 2802or individually for one file:</p> 2803<dl> 2804 <dt><code>int xmlGetDocCompressMode (xmlDocPtr doc);</code></dt> 2805 <dd><p>Gets the document compression ratio (0-9).</p> 2806 </dd> 2807</dl> 2808<dl> 2809 <dt><code>void xmlSetDocCompressMode (xmlDocPtr doc, int mode);</code></dt> 2810 <dd><p>Sets the document compression ratio.</p> 2811 </dd> 2812</dl> 2813<dl> 2814 <dt><code>int xmlGetCompressMode(void);</code></dt> 2815 <dd><p>Gets the default compression ratio.</p> 2816 </dd> 2817</dl> 2818<dl> 2819 <dt><code>void xmlSetCompressMode(int mode);</code></dt> 2820 <dd><p>Sets the default compression ratio.</p> 2821 </dd> 2822</dl> 2823 2824<h2><a name="Entities">Entities or no entities</a></h2> 2825 2826<p>Entities in principle are similar to simple C macros. An entity defines an 2827abbreviation for a given string that you can reuse many times throughout the 2828content of your document. Entities are especially useful when a given string 2829may occur frequently within a document, or to confine the change needed to a 2830document to a restricted area in the internal subset of the document (at the 2831beginning). Example:</p> 2832<pre>1 <?xml version="1.0"?> 28332 <!DOCTYPE EXAMPLE SYSTEM "example.dtd" [ 28343 <!ENTITY xml "Extensible Markup Language"> 28354 ]> 28365 <EXAMPLE> 28376 &xml; 28387 </EXAMPLE></pre> 2839 2840<p>Line 3 declares the xml entity. Line 6 uses the xml entity, by prefixing 2841its name with '&' and following it by ';' without any spaces added. There 2842are 5 predefined entities in libxml allowing you to escape charaters with 2843predefined meaning in some parts of the xml document content: 2844<strong>&lt;</strong> for the character '<', <strong>&gt;</strong> 2845for the character '>', <strong>&apos;</strong> for the character ''', 2846<strong>&quot;</strong> for the character '"', and 2847<strong>&amp;</strong> for the character '&'.</p> 2848 2849<p>One of the problems related to entities is that you may want the parser to 2850substitute an entity's content so that you can see the replacement text in 2851your application. Or you may prefer to keep entity references as such in the 2852content to be able to save the document back without losing this usually 2853precious information (if the user went through the pain of explicitly 2854defining entities, he may have a a rather negative attitude if you blindly 2855susbtitute them as saving time). The <a 2856href="html/libxml-parser.html#XMLSUBSTITUTEENTITIESDEFAULT">xmlSubstituteEntitiesDefault()</a> 2857function allows you to check and change the behaviour, which is to not 2858substitute entities by default.</p> 2859 2860<p>Here is the DOM tree built by libxml for the previous document in the 2861default case:</p> 2862<pre>/gnome/src/gnome-xml -> /xmllint --debug test/ent1 2863DOCUMENT 2864version=1.0 2865 ELEMENT EXAMPLE 2866 TEXT 2867 content= 2868 ENTITY_REF 2869 INTERNAL_GENERAL_ENTITY xml 2870 content=Extensible Markup Language 2871 TEXT 2872 content=</pre> 2873 2874<p>And here is the result when substituting entities:</p> 2875<pre>/gnome/src/gnome-xml -> /tester --debug --noent test/ent1 2876DOCUMENT 2877version=1.0 2878 ELEMENT EXAMPLE 2879 TEXT 2880 content= Extensible Markup Language</pre> 2881 2882<p>So, entities or no entities? Basically, it depends on your use case. I 2883suggest that you keep the non-substituting default behaviour and avoid using 2884entities in your XML document or data if you are not willing to handle the 2885entity references elements in the DOM tree.</p> 2886 2887<p>Note that at save time libxml enforces the conversion of the predefined 2888entities where necessary to prevent well-formedness problems, and will also 2889transparently replace those with chars (i.e. it will not generate entity 2890reference elements in the DOM tree or call the reference() SAX callback when 2891finding them in the input).</p> 2892 2893<p><span style="background-color: #FF0000">WARNING</span>: handling entities 2894on top of the libxml SAX interface is difficult!!! If you plan to use 2895non-predefined entities in your documents, then the learning cuvre to handle 2896then using the SAX API may be long. If you plan to use complex documents, I 2897strongly suggest you consider using the DOM interface instead and let libxml 2898deal with the complexity rather than trying to do it yourself.</p> 2899 2900<h2><a name="Namespaces">Namespaces</a></h2> 2901 2902<p>The libxml library implements <a 2903href="http://www.w3.org/TR/REC-xml-names/">XML namespaces</a> support by 2904recognizing namespace contructs in the input, and does namespace lookup 2905automatically when building the DOM tree. A namespace declaration is 2906associated with an in-memory structure and all elements or attributes within 2907that namespace point to it. Hence testing the namespace is a simple and fast 2908equality operation at the user level.</p> 2909 2910<p>I suggest that people using libxml use a namespace, and declare it in the 2911root element of their document as the default namespace. Then they don't need 2912to use the prefix in the content but we will have a basis for future semantic 2913refinement and merging of data from different sources. This doesn't increase 2914the size of the XML output significantly, but significantly increases its 2915value in the long-term. Example:</p> 2916<pre><mydoc xmlns="http://mydoc.example.org/schemas/"> 2917 <elem1>...</elem1> 2918 <elem2>...</elem2> 2919</mydoc></pre> 2920 2921<p>The namespace value has to be an absolute URL, but the URL doesn't have to 2922point to any existing resource on the Web. It will bind all the element and 2923atributes with that URL. I suggest to use an URL within a domain you control, 2924and that the URL should contain some kind of version information if possible. 2925For example, <code>"http://www.gnome.org/gnumeric/1.0/"</code> is a good 2926namespace scheme.</p> 2927 2928<p>Then when you load a file, make sure that a namespace carrying the 2929version-independent prefix is installed on the root element of your document, 2930and if the version information don't match something you know, warn the user 2931and be liberal in what you accept as the input. Also do *not* try to base 2932namespace checking on the prefix value. <foo:text> may be exactly the 2933same as <bar:text> in another document. What really matters is the URI 2934associated with the element or the attribute, not the prefix string (which is 2935just a shortcut for the full URI). In libxml, element and attributes have an 2936<code>ns</code> field pointing to an xmlNs structure detailing the namespace 2937prefix and its URI.</p> 2938 2939<p>@@Interfaces@@</p> 2940 2941<p>@@Examples@@</p> 2942 2943<p>Usually people object to using namespaces together with validity checking. 2944I will try to make sure that using namespaces won't break validity checking, 2945so even if you plan to use or currently are using validation I strongly 2946suggest adding namespaces to your document. A default namespace scheme 2947<code>xmlns="http://...."</code> should not break validity even on less 2948flexible parsers. Using namespaces to mix and differentiate content coming 2949from multiple DTDs will certainly break current validation schemes. I will 2950try to provide ways to do this, but this may not be portable or 2951standardized.</p> 2952 2953<h2><a name="Upgrading">Upgrading 1.x code</a></h2> 2954 2955<p>Incompatible changes:</p> 2956 2957<p>Version 2 of libxml is the first version introducing serious backward 2958incompatible changes. The main goals were:</p> 2959<ul> 2960 <li>a general cleanup. A number of mistakes inherited from the very early 2961 versions couldn't be changed due to compatibility constraints. Example 2962 the "childs" element in the nodes.</li> 2963 <li>Uniformization of the various nodes, at least for their header and link 2964 parts (doc, parent, children, prev, next), the goal is a simpler 2965 programming model and simplifying the task of the DOM implementors.</li> 2966 <li>better conformances to the XML specification, for example version 1.x 2967 had an heuristic to try to detect ignorable white spaces. As a result the 2968 SAX event generated were ignorableWhitespace() while the spec requires 2969 character() in that case. This also mean that a number of DOM node 2970 containing blank text may populate the DOM tree which were not present 2971 before.</li> 2972</ul> 2973 2974<h3>How to fix libxml-1.x code:</h3> 2975 2976<p>So client code of libxml designed to run with version 1.x may have to be 2977changed to compile against version 2.x of libxml. Here is a list of changes 2978that I have collected, they may not be sufficient, so in case you find other 2979change which are required, <a href="mailto:Daniel.�eillardw3.org">drop me a 2980mail</a>:</p> 2981<ol> 2982 <li>The package name have changed from libxml to libxml2, the library name 2983 is now -lxml2 . There is a new xml2-config script which should be used to 2984 select the right parameters libxml2</li> 2985 <li>Node <strong>childs</strong> field has been renamed 2986 <strong>children</strong> so s/childs/children/g should be applied 2987 (probablility of having "childs" anywere else is close to 0+</li> 2988 <li>The document don't have anymore a <strong>root</strong> element it has 2989 been replaced by <strong>children</strong> and usually you will get a 2990 list of element here. For example a Dtd element for the internal subset 2991 and it's declaration may be found in that list, as well as processing 2992 instructions or comments found before or after the document root element. 2993 Use <strong>xmlDocGetRootElement(doc)</strong> to get the root element of 2994 a document. Alternatively if you are sure to not reference Dtds nor have 2995 PIs or comments before or after the root element 2996 s/->root/->children/g will probably do it.</li> 2997 <li>The white space issue, this one is more complex, unless special case of 2998 validating parsing, the line breaks and spaces usually used for indenting 2999 and formatting the document content becomes significant. So they are 3000 reported by SAX and if your using the DOM tree, corresponding nodes are 3001 generated. Too approach can be taken: 3002 <ol> 3003 <li>lazy one, use the compatibility call 3004 <strong>xmlKeepBlanksDefault(0)</strong> but be aware that you are 3005 relying on a special (and possibly broken) set of heuristics of 3006 libxml to detect ignorable blanks. Don't complain if it breaks or 3007 make your application not 100% clean w.r.t. to it's input.</li> 3008 <li>the Right Way: change you code to accept possibly unsignificant 3009 blanks characters, or have your tree populated with weird blank text 3010 nodes. You can spot them using the comodity function 3011 <strong>xmlIsBlankNode(node)</strong> returning 1 for such blank 3012 nodes.</li> 3013 </ol> 3014 <p>Note also that with the new default the output functions don't add any 3015 extra indentation when saving a tree in order to be able to round trip 3016 (read and save) without inflating the document with extra formatting 3017 chars.</p> 3018 </li> 3019 <li>The include path has changed to $prefix/libxml/ and the includes 3020 themselves uses this new prefix in includes instructions... If you are 3021 using (as expected) the 3022 <pre>xml2-config --cflags</pre> 3023 <p>output to generate you compile commands this will probably work out of 3024 the box</p> 3025 </li> 3026 <li>xmlDetectCharEncoding takes an extra argument indicating the lenght in 3027 byte of the head of the document available for character detection.</li> 3028</ol> 3029 3030<h3>Ensuring both libxml-1.x and libxml-2.x compatibility</h3> 3031 3032<p>Two new version of libxml (1.8.11) and libxml2 (2.3.4) have been released 3033to allow smoth upgrade of existing libxml v1code while retaining 3034compatibility. They offers the following:</p> 3035<ol> 3036 <li>similar include naming, one should use 3037 <strong>#include<libxml/...></strong> in both cases.</li> 3038 <li>similar identifiers defined via macros for the child and root fields: 3039 respectively <strong>xmlChildrenNode</strong> and 3040 <strong>xmlRootNode</strong></li> 3041 <li>a new macro <strong>LIBXML_TEST_VERSION</strong> which should be 3042 inserted once in the client code</li> 3043</ol> 3044 3045<p>So the roadmap to upgrade your existing libxml applications is the 3046following:</p> 3047<ol> 3048 <li>install the libxml-1.8.8 (and libxml-devel-1.8.8) packages</li> 3049 <li>find all occurences where the xmlDoc <strong>root</strong> field is 3050 used and change it to <strong>xmlRootNode</strong></li> 3051 <li>similary find all occurences where the xmlNode <strong>childs</strong> 3052 field is used and change it to <strong>xmlChildrenNode</strong></li> 3053 <li>add a <strong>LIBXML_TEST_VERSION</strong> macro somewhere in your 3054 <strong>main()</strong> or in the library init entry point</li> 3055 <li>Recompile, check compatibility, it should still work</li> 3056 <li>Change your configure script to look first for xml2-config and fallback 3057 using xml-config . Use the --cflags and --libs ouptut of the command as 3058 the Include and Linking parameters needed to use libxml.</li> 3059 <li>install libxml2-2.3.x and libxml2-devel-2.3.x (libxml-1.8.y and 3060 libxml-devel-1.8.y can be kept simultaneously)</li> 3061 <li>remove your config.cache, relaunch your configuration mechanism, and 3062 recompile, if steps 2 and 3 were done right it should compile as-is</li> 3063 <li>Test that your application is still running correctly, if not this may 3064 be due to extra empty nodes due to formating spaces being kept in libxml2 3065 contrary to libxml1, in that case insert xmlKeepBlanksDefault(1) in your 3066 code before calling the parser (next to 3067 <strong>LIBXML_TEST_VERSION</strong> is a fine place).</li> 3068</ol> 3069 3070<p>Following those steps should work. It worked for some of my own code.</p> 3071 3072<p>Let me put some emphasis on the fact that there is far more changes from 3073libxml 1.x to 2.x than the ones you may have to patch for. The overall code 3074has been considerably cleaned up and the conformance to the XML specification 3075has been drastically improved too. Don't take those changes as an excuse to 3076not upgrade, it may cost a lot on the long term ...</p> 3077 3078<h2><a name="Thread">Thread safety</a></h2> 3079 3080<p>Starting with 2.4.7, libxml makes provisions to ensure that concurent 3081threads can safely work in parallel parsing different documents. There is 3082however a couple of things to do to ensure it:</p> 3083<ul> 3084 <li>configure the library accordingly using the --with-threads options</li> 3085 <li>call xmlInitParser() in the "main" thread before using any of the 3086 libxml API (except possibly selecting a different memory allocator)</li> 3087</ul> 3088 3089<p>Note that the thread safety cannot be ensured for multiple threads sharing 3090the same document, the locking must be done at the application level, libxml 3091exports a basic mutex and reentrant mutexes API in <libxml/threads.h>. 3092The parts of the library checked for thread safety are:</p> 3093<ul> 3094 <li>concurrent loading</li> 3095 <li>file access resolution</li> 3096 <li>catalog access</li> 3097 <li>catalog building</li> 3098 <li>entities lookup/accesses</li> 3099 <li>validation</li> 3100 <li>global variables per-thread override</li> 3101 <li>memory handling</li> 3102</ul> 3103 3104<p>XPath is supposed to be thread safe now, but this wasn't tested 3105seriously.</p> 3106 3107<h2><a name="DOM"></a><a name="Principles">DOM Principles</a></h2> 3108 3109<p><a href="http://www.w3.org/DOM/">DOM</a> stands for the <em>Document 3110Object Model</em>; this is an API for accessing XML or HTML structured 3111documents. Native support for DOM in Gnome is on the way (module gnome-dom), 3112and will be based on gnome-xml. This will be a far cleaner interface to 3113manipulate XML files within Gnome since it won't expose the internal 3114structure.</p> 3115 3116<p>The current DOM implementation on top of libxml is the <a 3117href="http://cvs.gnome.org/lxr/source/gdome2/">gdome2 Gnome module</a>, this 3118is a full DOM interface, thanks to Paolo Casarini, check the <a 3119href="http://www.cs.unibo.it/~casarini/gdome2/">Gdome2 homepage</a> for more 3120informations.</p> 3121 3122<h2><a name="Example"></a><a name="real">A real example</a></h2> 3123 3124<p>Here is a real size example, where the actual content of the application 3125data is not kept in the DOM tree but uses internal structures. It is based on 3126a proposal to keep a database of jobs related to Gnome, with an XML based 3127storage structure. Here is an <a href="gjobs.xml">XML encoded jobs 3128base</a>:</p> 3129<pre><?xml version="1.0"?> 3130<gjob:Helping xmlns:gjob="http://www.gnome.org/some-location"> 3131 <gjob:Jobs> 3132 3133 <gjob:Job> 3134 <gjob:Project ID="3"/> 3135 <gjob:Application>GBackup</gjob:Application> 3136 <gjob:Category>Development</gjob:Category> 3137 3138 <gjob:Update> 3139 <gjob:Status>Open</gjob:Status> 3140 <gjob:Modified>Mon, 07 Jun 1999 20:27:45 -0400 MET DST</gjob:Modified> 3141 <gjob:Salary>USD 0.00</gjob:Salary> 3142 </gjob:Update> 3143 3144 <gjob:Developers> 3145 <gjob:Developer> 3146 </gjob:Developer> 3147 </gjob:Developers> 3148 3149 <gjob:Contact> 3150 <gjob:Person>Nathan Clemons</gjob:Person> 3151 <gjob:Email>nathan@windsofstorm.net</gjob:Email> 3152 <gjob:Company> 3153 </gjob:Company> 3154 <gjob:Organisation> 3155 </gjob:Organisation> 3156 <gjob:Webpage> 3157 </gjob:Webpage> 3158 <gjob:Snailmail> 3159 </gjob:Snailmail> 3160 <gjob:Phone> 3161 </gjob:Phone> 3162 </gjob:Contact> 3163 3164 <gjob:Requirements> 3165 The program should be released as free software, under the GPL. 3166 </gjob:Requirements> 3167 3168 <gjob:Skills> 3169 </gjob:Skills> 3170 3171 <gjob:Details> 3172 A GNOME based system that will allow a superuser to configure 3173 compressed and uncompressed files and/or file systems to be backed 3174 up with a supported media in the system. This should be able to 3175 perform via find commands generating a list of files that are passed 3176 to tar, dd, cpio, cp, gzip, etc., to be directed to the tape machine 3177 or via operations performed on the filesystem itself. Email 3178 notification and GUI status display very important. 3179 </gjob:Details> 3180 3181 </gjob:Job> 3182 3183 </gjob:Jobs> 3184</gjob:Helping></pre> 3185 3186<p>While loading the XML file into an internal DOM tree is a matter of 3187calling only a couple of functions, browsing the tree to gather the ata and 3188generate the internal structures is harder, and more error prone.</p> 3189 3190<p>The suggested principle is to be tolerant with respect to the input 3191structure. For example, the ordering of the attributes is not significant, 3192the XML specification is clear about it. It's also usually a good idea not to 3193depend on the order of the children of a given node, unless it really makes 3194things harder. Here is some code to parse the information for a person:</p> 3195<pre>/* 3196 * A person record 3197 */ 3198typedef struct person { 3199 char *name; 3200 char *email; 3201 char *company; 3202 char *organisation; 3203 char *smail; 3204 char *webPage; 3205 char *phone; 3206} person, *personPtr; 3207 3208/* 3209 * And the code needed to parse it 3210 */ 3211personPtr parsePerson(xmlDocPtr doc, xmlNsPtr ns, xmlNodePtr cur) { 3212 personPtr ret = NULL; 3213 3214DEBUG("parsePerson\n"); 3215 /* 3216 * allocate the struct 3217 */ 3218 ret = (personPtr) malloc(sizeof(person)); 3219 if (ret == NULL) { 3220 fprintf(stderr,"out of memory\n"); 3221 return(NULL); 3222 } 3223 memset(ret, 0, sizeof(person)); 3224 3225 /* We don't care what the top level element name is */ 3226 cur = cur->xmlChildrenNode; 3227 while (cur != NULL) { 3228 if ((!strcmp(cur->name, "Person")) && (cur->ns == ns)) 3229 ret->name = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 3230 if ((!strcmp(cur->name, "Email")) && (cur->ns == ns)) 3231 ret->email = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 3232 cur = cur->next; 3233 } 3234 3235 return(ret); 3236}</pre> 3237 3238<p>Here are a couple of things to notice:</p> 3239<ul> 3240 <li>Usually a recursive parsing style is the more convenient one: XML data 3241 is by nature subject to repetitive constructs and usually exibits highly 3242 stuctured patterns.</li> 3243 <li>The two arguments of type <em>xmlDocPtr</em> and <em>xmlNsPtr</em>, 3244 i.e. the pointer to the global XML document and the namespace reserved to 3245 the application. Document wide information are needed for example to 3246 decode entities and it's a good coding practice to define a namespace for 3247 your application set of data and test that the element and attributes 3248 you're analyzing actually pertains to your application space. This is 3249 done by a simple equality test (cur->ns == ns).</li> 3250 <li>To retrieve text and attributes value, you can use the function 3251 <em>xmlNodeListGetString</em> to gather all the text and entity reference 3252 nodes generated by the DOM output and produce an single text string.</li> 3253</ul> 3254 3255<p>Here is another piece of code used to parse another level of the 3256structure:</p> 3257<pre>#include <libxml/tree.h> 3258/* 3259 * a Description for a Job 3260 */ 3261typedef struct job { 3262 char *projectID; 3263 char *application; 3264 char *category; 3265 personPtr contact; 3266 int nbDevelopers; 3267 personPtr developers[100]; /* using dynamic alloc is left as an exercise */ 3268} job, *jobPtr; 3269 3270/* 3271 * And the code needed to parse it 3272 */ 3273jobPtr parseJob(xmlDocPtr doc, xmlNsPtr ns, xmlNodePtr cur) { 3274 jobPtr ret = NULL; 3275 3276DEBUG("parseJob\n"); 3277 /* 3278 * allocate the struct 3279 */ 3280 ret = (jobPtr) malloc(sizeof(job)); 3281 if (ret == NULL) { 3282 fprintf(stderr,"out of memory\n"); 3283 return(NULL); 3284 } 3285 memset(ret, 0, sizeof(job)); 3286 3287 /* We don't care what the top level element name is */ 3288 cur = cur->xmlChildrenNode; 3289 while (cur != NULL) { 3290 3291 if ((!strcmp(cur->name, "Project")) && (cur->ns == ns)) { 3292 ret->projectID = xmlGetProp(cur, "ID"); 3293 if (ret->projectID == NULL) { 3294 fprintf(stderr, "Project has no ID\n"); 3295 } 3296 } 3297 if ((!strcmp(cur->name, "Application")) && (cur->ns == ns)) 3298 ret->application = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 3299 if ((!strcmp(cur->name, "Category")) && (cur->ns == ns)) 3300 ret->category = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 3301 if ((!strcmp(cur->name, "Contact")) && (cur->ns == ns)) 3302 ret->contact = parsePerson(doc, ns, cur); 3303 cur = cur->next; 3304 } 3305 3306 return(ret); 3307}</pre> 3308 3309<p>Once you are used to it, writing this kind of code is quite simple, but 3310boring. Ultimately, it could be possble to write stubbers taking either C 3311data structure definitions, a set of XML examples or an XML DTD and produce 3312the code needed to import and export the content between C data and XML 3313storage. This is left as an exercise to the reader :-)</p> 3314 3315<p>Feel free to use <a href="example/gjobread.c">the code for the full C 3316parsing example</a> as a template, it is also available with Makefile in the 3317Gnome CVS base under gnome-xml/example</p> 3318 3319<h2><a name="Contributi">Contributions</a></h2> 3320<ul> 3321 <li>Bjorn Reese, William Brack and Thomas Broyer have provided a number of 3322 patches, Gary Pennington worked on the validation API, threading support 3323 and Solaris port.</li> 3324 <li>John Fleck helps maintaining the documentation and man pages.</li> 3325 <li><a href="mailto:ari@lusis.org">Ari Johnson</a> 3326 provides a C++ wrapper for libxml:<br> 3327 Website: <a 3328 href="http://lusis.org/~ari/xml++/">http://lusis.org/~ari/xml++/</a><br> 3329 Download: <a 3330 href="http://lusis.org/~ari/xml++/libxml++.tar.gz">http://lusis.org/~ari/xml++/libxml++.tar.gz</a></li> 3331 <li><a href="mailto:izlatkovic@daenet.de">Igor Zlatkovic</a> 3332 is now the maintainer of the Windows port, <a 3333 href="http://www.fh-frankfurt.de/~igor/projects/libxml/index.html">he 3334 provides binaries</a></li> 3335 <li><a href="mailto:Gary.Pennington@sun.com">Gary Pennington</a> 3336 provides <a href="http://garypennington.net/libxml2/">Solaris 3337 binaries</a></li> 3338 <li><a 3339 href="http://mail.gnome.org/archives/xml/2001-March/msg00014.html">Matt 3340 Sergeant</a> 3341 developped <a href="http://axkit.org/download/">XML::LibXSLT</a>, a perl 3342 wrapper for libxml2/libxslt as part of the <a 3343 href="http://axkit.com/">AxKit XML application server</a></li> 3344 <li><a href="mailto:fnatter@gmx.net">Felix Natter</a> 3345 and <a href="mailto:geertk@ai.rug.nl">Geert Kloosterman</a> provide <a 3346 href="libxml-doc.el">an emacs module</a> to lookup libxml(2) functions 3347 documentation</li> 3348 <li><a href="mailto:sherwin@nlm.nih.gov">Ziying Sherwin</a> 3349 provided <a href="http://xmlsoft.org/messages/0488.html">man 3350 pages</a></li> 3351 <li>there is a module for <a 3352 href="http://acs-misc.sourceforge.net/nsxml.html">libxml/libxslt support 3353 in OpenNSD/AOLServer</a></li> 3354 <li><a href="mailto:dkuhlman@cutter.rexx.com">Dave Kuhlman</a> 3355 provides libxml/libxslt <a href="http://www.rexx.com/~dkuhlman">wrappers 3356 for Python</a></li> 3357 <li>Petr Kozelka provides <a 3358 href="http://sourceforge.net/projects/libxml2-pas">Pascal units to glue 3359 libxml2</a> with Kylix and Delphi and other Pascal compilers</li> 3360</ul> 3361 3362<p></p> 3363</body> 3364</html> 3365
