xml.html revision 7ebac02df6a57a5b9598a1d7e701cd0fac8d9b6e
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 parser and toolkit of Gnome</title> 6 <meta name="GENERATOR" content="amaya 5.1"> 7 <meta http-equiv="Content-Type" content="text/html"> 8</head> 9 10<body bgcolor="#ffffff"> 11<h1 align="center">The XML C parser and toolkit of 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 21style="text-align: right; font-style: italic; font-size: 10pt">"Programming 22with libxml2 is like the thrilling embrace of an exotic stranger." <a 23href="http://diveintomark.org/archives/2004/02/18/libxml2">Mark 24Pilgrim</a></p> 25 26<p>Libxml2 is the XML C parser and toolkit developed for the Gnome project 27(but usable outside of the Gnome platform), it is free software available 28under the <a href="http://www.opensource.org/licenses/mit-license.html">MIT 29License</a>. XML itself is a metalanguage to design markup languages, i.e. 30text language where semantic and structure are added to the content using 31extra "markup" information enclosed between angle brackets. HTML is the most 32well-known markup language. Though the library is written in C <a 33href="python.html">a variety of language bindings</a> make it available in 34other environments.</p> 35 36<p>Libxml2 is known to be very portable, the library should build and work 37without serious troubles on a variety of systems (Linux, Unix, Windows, 38CygWin, MacOS, MacOS X, RISC Os, OS/2, VMS, QNX, MVS, ...)</p> 39 40<p>Libxml2 implements a number of existing standards related to markup 41languages:</p> 42<ul> 43 <li>the XML standard: <a 44 href="http://www.w3.org/TR/REC-xml">http://www.w3.org/TR/REC-xml</a></li> 45 <li>Namespaces in XML: <a 46 href="http://www.w3.org/TR/REC-xml-names/">http://www.w3.org/TR/REC-xml-names/</a></li> 47 <li>XML Base: <a 48 href="http://www.w3.org/TR/xmlbase/">http://www.w3.org/TR/xmlbase/</a></li> 49 <li><a href="http://www.cis.ohio-state.edu/rfc/rfc2396.txt">RFC 2396</a> : 50 Uniform Resource Identifiers <a 51 href="http://www.ietf.org/rfc/rfc2396.txt">http://www.ietf.org/rfc/rfc2396.txt</a></li> 52 <li>XML Path Language (XPath) 1.0: <a 53 href="http://www.w3.org/TR/xpath">http://www.w3.org/TR/xpath</a></li> 54 <li>HTML4 parser: <a 55 href="http://www.w3.org/TR/html401/">http://www.w3.org/TR/html401/</a></li> 56 <li>XML Pointer Language (XPointer) Version 1.0: <a 57 href="http://www.w3.org/TR/xptr">http://www.w3.org/TR/xptr</a></li> 58 <li>XML Inclusions (XInclude) Version 1.0: <a 59 href="http://www.w3.org/TR/xinclude/">http://www.w3.org/TR/xinclude/</a></li> 60 <li>ISO-8859-x encodings, as well as <a 61 href="http://www.cis.ohio-state.edu/rfc/rfc2044.txt">rfc2044</a> [UTF-8] 62 and <a href="http://www.cis.ohio-state.edu/rfc/rfc2781.txt">rfc2781</a> 63 [UTF-16] Unicode encodings, and more if using iconv support</li> 64 <li>part of SGML Open Technical Resolution TR9401:1997</li> 65 <li>XML Catalogs Working Draft 06 August 2001: <a 66 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> 67 <li>Canonical XML Version 1.0: <a 68 href="http://www.w3.org/TR/xml-c14n">http://www.w3.org/TR/xml-c14n</a> 69 and the Exclusive XML Canonicalization CR draft <a 70 href="http://www.w3.org/TR/xml-exc-c14n">http://www.w3.org/TR/xml-exc-c14n</a></li> 71 <li>Relax NG, ISO/IEC 19757-2:2003, <a 72 href="http://www.oasis-open.org/committees/relax-ng/spec-20011203.html">http://www.oasis-open.org/committees/relax-ng/spec-20011203.html</a></li> 73 <li>W3C XML Schemas Part 2: Datatypes <a 74 href="http://www.w3.org/TR/2001/REC-xmlschema-2-20010502/">REC 02 May 75 2001</a></li> 76</ul> 77 78<p>In most cases libxml2 tries to implement the specifications in a 79relatively strictly compliant way. As of release 2.4.16, libxml2 passes all 801800+ tests from the <a 81href="http://www.oasis-open.org/committees/xml-conformance/">OASIS XML Tests 82Suite</a>.</p> 83 84<p>To some extent libxml2 provides support for the following additional 85specifications but doesn't claim to implement them completely:</p> 86<ul> 87 <li>Document Object Model (DOM) <a 88 href="http://www.w3.org/TR/DOM-Level-2-Core/">http://www.w3.org/TR/DOM-Level-2-Core/</a> 89 it doesn't implement the API itself, gdome2 does this on top of 90 libxml2</li> 91 <li><a href="http://www.cis.ohio-state.edu/rfc/rfc959.txt">RFC 959</a> : 92 libxml2 implements a basic FTP client code</li> 93 <li><a href="http://www.cis.ohio-state.edu/rfc/rfc1945.txt">RFC 1945</a> : 94 HTTP/1.0, again a basic HTTP client code</li> 95 <li>SAX: a minimal SAX implementation compatible with early expat 96 versions</li> 97 <li>DocBook SGML v4: libxml2 includes a hackish parser to transition to 98 XML</li> 99</ul> 100 101<p>A partial implementation of <a 102href="http://www.w3.org/TR/2001/REC-xmlschema-1-20010502/">XML Schemas Part 1031: Structure</a> is being worked on but it would be far too early to make any 104conformance statement about it at the moment.</p> 105 106<p>Separate documents:</p> 107<ul> 108 <li><a href="http://xmlsoft.org/XSLT/">the libxslt page</a> providing an 109 implementation of XSLT 1.0 and common extensions like EXSLT for 110 libxml2</li> 111 <li><a href="http://www.cs.unibo.it/~casarini/gdome2/">the gdome2 page</a> 112 : a standard DOM2 implementation for libxml2</li> 113 <li><a href="http://www.aleksey.com/xmlsec/">the XMLSec page</a>: an 114 implementation of <a href="http://www.w3.org/TR/xmldsig-core/">W3C XML 115 Digital Signature</a> for libxml2</li> 116 <li>also check the related links section below for more related and active 117 projects.</li> 118</ul> 119 120<p>Results of the <a 121href="http://xmlbench.sourceforge.net/results/benchmark/index.html">xmlbench 122benchmark</a> on sourceforge February 2004 (smaller is better):</p> 123 124<p align="center"><img src="benchmark.png" 125alt="benchmark results for Expat Xerces libxml2 Oracle and Sun toolkits"></p> 126 127<p>Logo designed by <a href="mailto:liyanage@access.ch">Marc Liyanage</a>.</p> 128 129<h2><a name="Introducti">Introduction</a></h2> 130 131<p>This document describes libxml, the <a 132href="http://www.w3.org/XML/">XML</a> C parser and toolkit developed for the 133<a href="http://www.gnome.org/">Gnome</a> project. <a 134href="http://www.w3.org/XML/">XML is a standard</a> for building tag-based 135structured documents/data.</p> 136 137<p>Here are some key points about libxml:</p> 138<ul> 139 <li>Libxml2 exports Push (progressive) and Pull (blocking) type parser 140 interfaces for both XML and HTML.</li> 141 <li>Libxml2 can do DTD validation at parse time, using a parsed document 142 instance, or with an arbitrary DTD.</li> 143 <li>Libxml2 includes complete <a 144 href="http://www.w3.org/TR/xpath">XPath</a>, <a 145 href="http://www.w3.org/TR/xptr">XPointer</a> and <a 146 href="http://www.w3.org/TR/xinclude">XInclude</a> implementations.</li> 147 <li>It is written in plain C, making as few assumptions as possible, and 148 sticking closely to ANSI C/POSIX for easy embedding. Works on 149 Linux/Unix/Windows, ported to a number of other platforms.</li> 150 <li>Basic support for HTTP and FTP client allowing applications to fetch 151 remote resources.</li> 152 <li>The design is modular, most of the extensions can be compiled out.</li> 153 <li>The internal document representation is as close as possible to the <a 154 href="http://www.w3.org/DOM/">DOM</a> interfaces.</li> 155 <li>Libxml2 also has a <a 156 href="http://www.megginson.com/SAX/index.html">SAX like interface</a>; 157 the interface is designed to be compatible with <a 158 href="http://www.jclark.com/xml/expat.html">Expat</a>.</li> 159 <li>This library is released under the <a 160 href="http://www.opensource.org/licenses/mit-license.html">MIT 161 License</a>. See the Copyright file in the distribution for the precise 162 wording.</li> 163</ul> 164 165<p>Warning: unless you are forced to because your application links with a 166Gnome-1.X library requiring it, <strong><span 167style="background-color: #FF0000">Do Not Use libxml1</span></strong>, use 168libxml2</p> 169 170<h2><a name="FAQ">FAQ</a></h2> 171 172<p>Table of Contents:</p> 173<ul> 174 <li><a href="FAQ.html#License">License(s)</a></li> 175 <li><a href="FAQ.html#Installati">Installation</a></li> 176 <li><a href="FAQ.html#Compilatio">Compilation</a></li> 177 <li><a href="FAQ.html#Developer">Developer corner</a></li> 178</ul> 179 180<h3><a name="License">License</a>(s)</h3> 181<ol> 182 <li><em>Licensing Terms for libxml</em> 183 <p>libxml2 is released under the <a 184 href="http://www.opensource.org/licenses/mit-license.html">MIT 185 License</a>; see the file Copyright in the distribution for the precise 186 wording</p> 187 </li> 188 <li><em>Can I embed libxml2 in a proprietary application ?</em> 189 <p>Yes. The MIT License allows you to keep proprietary the changes you 190 made to libxml, but it would be graceful to send-back bug fixes and 191 improvements as patches for possible incorporation in the main 192 development tree.</p> 193 </li> 194</ol> 195 196<h3><a name="Installati">Installation</a></h3> 197<ol> 198 <li><strong><span style="background-color: #FF0000">Do Not Use 199 libxml1</span></strong>, use libxml2</li> 200 <li><em>Where can I get libxml</em> ? 201 <p>The original distribution comes from <a 202 href="ftp://rpmfind.net/pub/libxml/">rpmfind.net</a> or <a 203 href="ftp://ftp.gnome.org/pub/GNOME/sources/libxml2/2.6/">gnome.org</a></p> 204 <p>Most Linux and BSD distributions include libxml, this is probably the 205 safer way for end-users to use libxml.</p> 206 <p>David Doolin provides precompiled Windows versions at <a 207 href="http://www.ce.berkeley.edu/~doolin/code/libxmlwin32/ ">http://www.ce.berkeley.edu/~doolin/code/libxmlwin32/</a></p> 208 </li> 209 <li><em>I see libxml and libxml2 releases, which one should I install ?</em> 210 <ul> 211 <li>If you are not constrained by backward compatibility issues with 212 existing applications, install libxml2 only</li> 213 <li>If you are not doing development, you can safely install both. 214 Usually the packages <a 215 href="http://rpmfind.net/linux/RPM/libxml.html">libxml</a> and <a 216 href="http://rpmfind.net/linux/RPM/libxml2.html">libxml2</a> are 217 compatible (this is not the case for development packages).</li> 218 <li>If you are a developer and your system provides separate packaging 219 for shared libraries and the development components, it is possible 220 to install libxml and libxml2, and also <a 221 href="http://rpmfind.net/linux/RPM/libxml-devel.html">libxml-devel</a> 222 and <a 223 href="http://rpmfind.net/linux/RPM/libxml2-devel.html">libxml2-devel</a> 224 too for libxml2 >= 2.3.0</li> 225 <li>If you are developing a new application, please develop against 226 libxml2(-devel)</li> 227 </ul> 228 </li> 229 <li><em>I can't install the libxml package, it conflicts with libxml0</em> 230 <p>You probably have an old libxml0 package used to provide the shared 231 library for libxml.so.0, you can probably safely remove it. The libxml 232 packages provided on <a 233 href="ftp://rpmfind.net/pub/libxml/">rpmfind.net</a> provide 234 libxml.so.0</p> 235 </li> 236 <li><em>I can't install the libxml(2) RPM package due to failed 237 dependencies</em> 238 <p>The most generic solution is to re-fetch the latest src.rpm , and 239 rebuild it locally with</p> 240 <p><code>rpm --rebuild libxml(2)-xxx.src.rpm</code>.</p> 241 <p>If everything goes well it will generate two binary rpm packages (one 242 providing the shared libs and xmllint, and the other one, the -devel 243 package, providing includes, static libraries and scripts needed to build 244 applications with libxml(2)) that you can install locally.</p> 245 </li> 246</ol> 247 248<h3><a name="Compilatio">Compilation</a></h3> 249<ol> 250 <li><em>What is the process to compile libxml2 ?</em> 251 <p>As most UNIX libraries libxml2 follows the "standard":</p> 252 <p><code>gunzip -c xxx.tar.gz | tar xvf -</code></p> 253 <p><code>cd libxml-xxxx</code></p> 254 <p><code>/configure --help</code></p> 255 <p>to see the options, then the compilation/installation proper</p> 256 <p><code>/configure [possible options]</code></p> 257 <p><code>make</code></p> 258 <p><code>make install</code></p> 259 <p>At that point you may have to rerun ldconfig or a similar utility to 260 update your list of installed shared libs.</p> 261 </li> 262 <li><em>What other libraries are needed to compile/install libxml2 ?</em> 263 <p>Libxml2 does not require any other library, the normal C ANSI API 264 should be sufficient (please report any violation to this rule you may 265 find).</p> 266 <p>However if found at configuration time libxml2 will detect and use the 267 following libs:</p> 268 <ul> 269 <li><a href="http://www.info-zip.org/pub/infozip/zlib/">libz</a> : a 270 highly portable and available widely compression library.</li> 271 <li>iconv: a powerful character encoding conversion library. It is 272 included by default in recent glibc libraries, so it doesn't need to 273 be installed specifically on Linux. It now seems a <a 274 href="http://www.opennc.org/onlinepubs/7908799/xsh/iconv.html">part 275 of the official UNIX</a> specification. Here is one <a 276 href="http://www.gnu.org/software/libiconv/">implementation of the 277 library</a> which source can be found <a 278 href="ftp://ftp.ilog.fr/pub/Users/haible/gnu/">here</a>.</li> 279 </ul> 280 </li> 281 <li><em>Make check fails on some platforms</em> 282 <p>Sometimes the regression tests' results don't completely match the 283 value produced by the parser, and the makefile uses diff to print the 284 delta. On some platforms the diff return breaks the compilation process; 285 if the diff is small this is probably not a serious problem.</p> 286 <p>Sometimes (especially on Solaris) make checks fail due to limitations 287 in make. Try using GNU-make instead.</p> 288 </li> 289 <li><em>I use the CVS version and there is no configure script</em> 290 <p>The configure script (and other Makefiles) are generated. Use the 291 autogen.sh script to regenerate the configure script and Makefiles, 292 like:</p> 293 <p><code>/autogen.sh --prefix=/usr --disable-shared</code></p> 294 </li> 295 <li><em>I have troubles when running make tests with gcc-3.0</em> 296 <p>It seems the initial release of gcc-3.0 has a problem with the 297 optimizer which miscompiles the URI module. Please use another 298 compiler.</p> 299 </li> 300</ol> 301 302<h3><a name="Developer">Developer</a> corner</h3> 303<ol> 304 <li><em>Troubles compiling or linking programs using libxml2</em> 305 <p>Usually the problem comes from the fact that the compiler doesn't get 306 the right compilation or linking flags. There is a small shell script 307 <code>xml2-config</code> which is installed as part of libxml2 usual 308 install process which provides those flags. Use</p> 309 <p><code>xml2-config --cflags</code></p> 310 <p>to get the compilation flags and</p> 311 <p><code>xml2-config --libs</code></p> 312 <p>to get the linker flags. Usually this is done directly from the 313 Makefile as:</p> 314 <p><code>CFLAGS=`xml2-config --cflags`</code></p> 315 <p><code>LIBS=`xml2-config --libs`</code></p> 316 </li> 317 <li><em>xmlDocDump() generates output on one line.</em> 318 <p>Libxml2 will not <strong>invent</strong> spaces in the content of a 319 document since <strong>all spaces in the content of a document are 320 significant</strong>. If you build a tree from the API and want 321 indentation:</p> 322 <ol> 323 <li>the correct way is to generate those yourself too.</li> 324 <li>the dangerous way is to ask libxml2 to add those blanks to your 325 content <strong>modifying the content of your document in the 326 process</strong>. The result may not be what you expect. There is 327 <strong>NO</strong> way to guarantee that such a modification won't 328 affect other parts of the content of your document. See <a 329 href="http://xmlsoft.org/html/libxml-parser.html#xmlKeepBlanksDefault">xmlKeepBlanksDefault 330 ()</a> and <a 331 href="http://xmlsoft.org/html/libxml-tree.html#xmlSaveFormatFile">xmlSaveFormatFile 332 ()</a></li> 333 </ol> 334 </li> 335 <li>Extra nodes in the document: 336 <p><em>For a XML file as below:</em></p> 337 <pre><?xml version="1.0"?> 338<PLAN xmlns="http://www.argus.ca/autotest/1.0/"> 339<NODE CommFlag="0"/> 340<NODE CommFlag="1"/> 341</PLAN></pre> 342 <p><em>after parsing it with the function 343 pxmlDoc=xmlParseFile(...);</em></p> 344 <p><em>I want to the get the content of the first node (node with the 345 CommFlag="0")</em></p> 346 <p><em>so I did it as following;</em></p> 347 <pre>xmlNodePtr pnode; 348pnode=pxmlDoc->children->children;</pre> 349 <p><em>but it does not work. If I change it to</em></p> 350 <pre>pnode=pxmlDoc->children->children->next;</pre> 351 <p><em>then it works. Can someone explain it to me.</em></p> 352 <p></p> 353 <p>In XML all characters in the content of the document are significant 354 <strong>including blanks and formatting line breaks</strong>.</p> 355 <p>The extra nodes you are wondering about are just that, text nodes with 356 the formatting spaces which are part of the document but that people tend 357 to forget. There is a function <a 358 href="http://xmlsoft.org/html/libxml-parser.html">xmlKeepBlanksDefault 359 ()</a> to remove those at parse time, but that's an heuristic, and its 360 use should be limited to cases where you are certain there is no 361 mixed-content in the document.</p> 362 </li> 363 <li><em>I get compilation errors of existing code like when accessing 364 <strong>root</strong> or <strong>child fields</strong> of nodes.</em> 365 <p>You are compiling code developed for libxml version 1 and using a 366 libxml2 development environment. Either switch back to libxml v1 devel or 367 even better fix the code to compile with libxml2 (or both) by <a 368 href="upgrade.html">following the instructions</a>.</p> 369 </li> 370 <li><em>I get compilation errors about non existing 371 <strong>xmlRootNode</strong> or <strong>xmlChildrenNode</strong> 372 fields.</em> 373 <p>The source code you are using has been <a 374 href="upgrade.html">upgraded</a> to be able to compile with both libxml 375 and libxml2, but you need to install a more recent version: 376 libxml(-devel) >= 1.8.8 or libxml2(-devel) >= 2.1.0</p> 377 </li> 378 <li><em>XPath implementation looks seriously broken</em> 379 <p>XPath implementation prior to 2.3.0 was really incomplete. Upgrade to 380 a recent version, there are no known bugs in the current version.</p> 381 </li> 382 <li><em>The example provided in the web page does not compile.</em> 383 <p>It's hard to maintain the documentation in sync with the code 384 <grin/> ...</p> 385 <p>Check the previous points 1/ and 2/ raised before, and please send 386 patches.</p> 387 </li> 388 <li><em>Where can I get more examples and information than provided on the 389 web page?</em> 390 <p>Ideally a libxml2 book would be nice. I have no such plan ... But you 391 can:</p> 392 <ul> 393 <li>check more deeply the <a href="html/libxml-lib.html">existing 394 generated doc</a></li> 395 <li>have a look at <a href="examples/index.html">the set of 396 examples</a>.</li> 397 <li>look for examples of use for libxml2 function using the Gnome code. 398 For example the following will query the full Gnome CVS base for the 399 use of the <strong>xmlAddChild()</strong> function: 400 <p><a 401 href="http://cvs.gnome.org/lxr/search?string=xmlAddChild">http://cvs.gnome.org/lxr/search?string=xmlAddChild</a></p> 402 <p>This may be slow, a large hardware donation to the gnome project 403 could cure this :-)</p> 404 </li> 405 <li><a 406 href="http://cvs.gnome.org/bonsai/rview.cgi?cvsroot=/cvs/gnome&dir=gnome-xml">Browse 407 the libxml2 source</a> , I try to write code as clean and documented 408 as possible, so looking at it may be helpful. In particular the code 409 of xmllint.c and of the various testXXX.c test programs should 410 provide good examples of how to do things with the library.</li> 411 </ul> 412 </li> 413 <li>What about C++ ? 414 <p>libxml2 is written in pure C in order to allow easy reuse on a number 415 of platforms, including embedded systems. I don't intend to convert to 416 C++.</p> 417 <p>There is however a C++ wrapper which may fulfill your needs:</p> 418 <ul> 419 <li>by Ari Johnson <ari@btigate.com>: 420 <p>Website: <a 421 href="http://libxmlplusplus.sourceforge.net/">http://libxmlplusplus.sourceforge.net/</a></p> 422 <p>Download: <a 423 href="http://sourceforge.net/project/showfiles.php?group_id=12999">http://sourceforge.net/project/showfiles.php?group_id=12999</a></p> 424 </li> 425 <!-- Website is currently unavailable as of 2003-08-02 426 <li>by Peter Jones <pjones@pmade.org> 427 <p>Website: <a 428 href="http://pmade.org/pjones/software/xmlwrapp/">http://pmade.org/pjones/software/xmlwrapp/</a></p> 429 </li> 430 --> 431 </ul> 432 </li> 433 <li>How to validate a document a posteriori ? 434 <p>It is possible to validate documents which had not been validated at 435 initial parsing time or documents which have been built from scratch 436 using the API. Use the <a 437 href="http://xmlsoft.org/html/libxml-valid.html#xmlValidateDtd">xmlValidateDtd()</a> 438 function. It is also possible to simply add a DTD to an existing 439 document:</p> 440 <pre>xmlDocPtr doc; /* your existing document */ 441xmlDtdPtr dtd = xmlParseDTD(NULL, filename_of_dtd); /* parse the DTD */ 442 443 dtd->name = xmlStrDup((xmlChar*)"root_name"); /* use the given root */ 444 445 doc->intSubset = dtd; 446 if (doc->children == NULL) xmlAddChild((xmlNodePtr)doc, (xmlNodePtr)dtd); 447 else xmlAddPrevSibling(doc->children, (xmlNodePtr)dtd); 448 </pre> 449 </li> 450 <li>So what is this funky "xmlChar" used all the time? 451 <p>It is a null terminated sequence of utf-8 characters. And only utf-8! 452 You need to convert strings encoded in different ways to utf-8 before 453 passing them to the API. This can be accomplished with the iconv library 454 for instance.</p> 455 </li> 456 <li>etc ...</li> 457</ol> 458 459<p></p> 460 461<h2><a name="Documentat">Developer Menu</a></h2> 462 463<p>There are several on-line resources related to using libxml:</p> 464<ol> 465 <li>Use the <a href="search.php">search engine</a> to look up 466 information.</li> 467 <li>Check the <a href="FAQ.html">FAQ.</a></li> 468 <li>Check the <a href="http://xmlsoft.org/html/libxml-lib.html">extensive 469 documentation</a> automatically extracted from code comments.</li> 470 <li>Look at the documentation about <a href="encoding.html">libxml 471 internationalization support</a>.</li> 472 <li>This page provides a global overview and <a href="example.html">some 473 examples</a> on how to use libxml.</li> 474 <li><a href="examples/index.html">Code examples</a></li> 475 <li>John Fleck's libxml2 tutorial: <a href="tutorial/index.html">html</a> 476 or <a href="tutorial/xmltutorial.pdf">pdf</a>.</li> 477 <li>If you need to parse large files, check the <a 478 href="xmlreader.html">xmlReader</a> API tutorial</li> 479 <li><a href="mailto:james@daa.com.au">James Henstridge</a> wrote <a 480 href="http://www.daa.com.au/~james/gnome/xml-sax/xml-sax.html">some nice 481 documentation</a> explaining how to use the libxml SAX interface.</li> 482 <li>George Lebl wrote <a 483 href="http://www-106.ibm.com/developerworks/library/l-gnome3/">an article 484 for IBM developerWorks</a> about using libxml.</li> 485 <li>Check <a href="http://cvs.gnome.org/lxr/source/gnome-xml/TODO">the TODO 486 file</a>.</li> 487 <li>Read the <a href="upgrade.html">1.x to 2.x upgrade path</a> 488 description. If you are starting a new project using libxml you should 489 really use the 2.x version.</li> 490 <li>And don't forget to look at the <a 491 href="http://mail.gnome.org/archives/xml/">mailing-list archive</a>.</li> 492</ol> 493 494<h2><a name="Reporting">Reporting bugs and getting help</a></h2> 495 496<p>Well, bugs or missing features are always possible, and I will make a 497point of fixing them in a timely fashion. The best way to report a bug is to 498use the <a href="http://bugzilla.gnome.org/buglist.cgi?product=libxml2">Gnome 499bug tracking database</a> (make sure to use the "libxml2" module name). I 500look at reports there regularly and it's good to have a reminder when a bug 501is still open. Be sure to specify that the bug is for the package libxml2.</p> 502 503<p>For small problems you can try to get help on IRC, the #xml channel on 504irc.gnome.org (port 6667) usually have a few person subscribed which may help 505(but there is no garantee and if a real issue is raised it should go on the 506mailing-list for archival).</p> 507 508<p>There is also a mailing-list <a 509href="mailto:xml@gnome.org">xml@gnome.org</a> for libxml, with an <a 510href="http://mail.gnome.org/archives/xml/">on-line archive</a> (<a 511href="http://xmlsoft.org/messages">old</a>). To subscribe to this list, 512please visit the <a 513href="http://mail.gnome.org/mailman/listinfo/xml">associated Web</a> page and 514follow the instructions. <strong>Do not send code, I won't debug it</strong> 515(but patches are really appreciated!).</p> 516 517<p>Check the following <strong><span style="color: #FF0000">before 518posting</span></strong>:</p> 519<ul> 520 <li>Read the <a href="FAQ.html">FAQ</a> and <a href="search.php">use the 521 search engine</a> to get information related to your problem.</li> 522 <li>Make sure you are <a href="ftp://xmlsoft.org/">using a recent 523 version</a>, and that the problem still shows up in a recent version.</li> 524 <li>Check the <a href="http://mail.gnome.org/archives/xml/">list 525 archives</a> to see if the problem was reported already. In this case 526 there is probably a fix available, similarly check the <a 527 href="http://bugzilla.gnome.org/buglist.cgi?product=libxml2">registered 528 open bugs</a>.</li> 529 <li>Make sure you can reproduce the bug with xmllint or one of the test 530 programs found in source in the distribution.</li> 531 <li>Please send the command showing the error as well as the input (as an 532 attachment)</li> 533</ul> 534 535<p>Then send the bug with associated information to reproduce it to the <a 536href="mailto:xml@gnome.org">xml@gnome.org</a> list; if it's really libxml 537related I will approve it. Please do not send mail to me directly, it makes 538things really hard to track and in some cases I am not the best person to 539answer a given question, ask on the list.</p> 540 541<p>To <span style="color: #E50000">be really clear about support</span>:</p> 542<ul> 543 <li>Support or help <span style="color: #E50000">requests MUST be sent to 544 the list or on bugzilla</span> in case of problems, so that the Question 545 and Answers can be shared publicly. Failing to do so carries the implicit 546 message "I want free support but I don't want to share the benefits with 547 others" and is not welcome. I will automatically Carbon-Copy the 548 xml@gnome.org mailing list for any technical reply made about libxml2 or 549 libxslt.</li> 550 <li>There is <span style="color: #E50000">no garantee of support</span>, if 551 your question remains unanswered after a week, repost it, making sure you 552 gave all the detail needed and the information requested.</li> 553 <li>Failing to provide information as requested or double checking first 554 for prior feedback also carries the implicit message "the time of the 555 library maintainers is less valuable than my time" and might not be 556 welcome.</li> 557</ul> 558 559<p>Of course, bugs reported with a suggested patch for fixing them will 560probably be processed faster than those without.</p> 561 562<p>If you're looking for help, a quick look at <a 563href="http://mail.gnome.org/archives/xml/">the list archive</a> may actually 564provide the answer. I usually send source samples when answering libxml2 565usage questions. The <a 566href="http://xmlsoft.org/html/book1.html">auto-generated documentation</a> is 567not as polished as I would like (i need to learn more about DocBook), but 568it's a good starting point.</p> 569 570<h2><a name="help">How to help</a></h2> 571 572<p>You can help the project in various ways, the best thing to do first is to 573subscribe to the mailing-list as explained before, check the <a 574href="http://mail.gnome.org/archives/xml/">archives </a>and the <a 575href="http://bugzilla.gnome.org/buglist.cgi?product=libxml2">Gnome bug 576database</a>:</p> 577<ol> 578 <li>Provide patches when you find problems.</li> 579 <li>Provide the diffs when you port libxml2 to a new platform. They may not 580 be integrated in all cases but help pinpointing portability problems 581 and</li> 582 <li>Provide documentation fixes (either as patches to the code comments or 583 as HTML diffs).</li> 584 <li>Provide new documentations pieces (translations, examples, etc 585 ...).</li> 586 <li>Check the TODO file and try to close one of the items.</li> 587 <li>Take one of the points raised in the archive or the bug database and 588 provide a fix. <a href="mailto:daniel@veillard.com">Get in touch with me 589 </a>before to avoid synchronization problems and check that the suggested 590 fix will fit in nicely :-)</li> 591</ol> 592 593<h2><a name="Downloads">Downloads</a></h2> 594 595<p>The latest versions of libxml2 can be found on <a 596href="ftp://xmlsoft.org/">xmlsoft.org</a> (<a 597href="ftp://speakeasy.rpmfind.net/pub/libxml/">Seattle</a>, <a 598href="ftp://fr.rpmfind.net/pub/libxml/">France</a>) or on the <a 599href="ftp://ftp.gnome.org/pub/GNOME/MIRRORS.html">Gnome FTP server</a> either 600as a <a href="ftp://ftp.gnome.org/pub/GNOME/sources/libxml2/2.6/">source 601archive</a><!-- commenting this out because they seem to have disappeared or <a 602href="ftp://ftp.gnome.org/pub/GNOME/stable/redhat/i386/libxml/">RPM 603packages</a> --> 604 , Antonin Sprinzl also provide <a href="ftp://gd.tuwien.ac.at/pub/libxml/">a 605mirror in Austria</a>. (NOTE that you need both the <a 606href="http://rpmfind.net/linux/RPM/libxml2.html">libxml(2)</a> and <a 607href="http://rpmfind.net/linux/RPM/libxml2-devel.html">libxml(2)-devel</a> 608packages installed to compile applications using libxml.)</p> 609 610<p>Binary ports:</p> 611<ul> 612 <li>Red Hat RPMs for i386 are available directly on <a 613 href="ftp://xmlsoft.org/">xmlsoft.org</a>, the source RPM will compile on 614 any architecture supported by Red Hat.</li> 615 <li><p><a href="mailto:igor@zlatkovic.com">Igor Zlatkovic</a></p> 616 is now the maintainer of the Windows port, <a 617 href="http://www.zlatkovic.com/projects/libxml/index.html">he provides 618 binaries</a>.</li> 619 <li><a href="mailto:Gary.Pennington@sun.com">Gary Pennington</a> provides 620 <a href="http://garypennington.net/libxml2/">Solaris binaries</a>.</li> 621 <li><a href="mailto:Steve.Ball@zveno.com">Steve Ball</a> provides <a 622 href="http://www.zveno.com/open_source/libxml2xslt.html">Mac Os X 623 binaries</a>.</li> 624 <li>The HP-UX porting center provides <a 625 href="http://hpux.connect.org.uk/hppd/hpux/Gnome/">HP-UX binaries</a></li> 626</ul> 627 628<p>If you know other supported binary ports, please <a 629href="http://veillard.com/">contact me</a>.</p> 630 631<p><a name="Snapshot">Snapshot:</a></p> 632<ul> 633 <li>Code from the W3C cvs base gnome-xml <a 634 href="ftp://xmlsoft.org/cvs-snapshot.tar.gz">cvs-snapshot.tar.gz</a>.</li> 635 <li>Docs, content of the web site, the list archive included <a 636 href="ftp://xmlsoft.org/libxml-docs.tar.gz">libxml-docs.tar.gz</a>.</li> 637</ul> 638 639<p><a name="Contribs">Contributions:</a></p> 640 641<p>I do accept external contributions, especially if compiling on another 642platform, get in touch with me to upload the package, wrappers for various 643languages have been provided, and can be found in the <a 644href="contribs.html">contrib section</a></p> 645 646<p>Libxml2 is also available from CVS:</p> 647<ul> 648 <li><p>The <a 649 href="http://cvs.gnome.org/bonsai/rview.cgi?cvsroot=/cvs/gnome&dir=gnome-xml">Gnome 650 CVS base</a>. Check the <a 651 href="http://developer.gnome.org/tools/cvs.html">Gnome CVS Tools</a> 652 page; the CVS module is <b>gnome-xml</b>.</p> 653 </li> 654 <li>The <strong>libxslt</strong> module is also present there</li> 655</ul> 656 657<h2><a name="News">News</a></h2> 658 659<h3>CVS only : check the <a 660href="http://cvs.gnome.org/lxr/source/gnome-xml/ChangeLog">Changelog</a> file 661for a really accurate description</h3> 662 663<p>Items not finished and worked on, get in touch with the list if you want 664to test those</p> 665<ul> 666 <li>More testing on RelaxNG</li> 667 <li>Finishing up <a href="http://www.w3.org/TR/xmlschema-1/">XML 668 Schemas</a></li> 669</ul> 670 671<h3>2.6.7: Feb 23 2004</h3> 672<ul> 673 <li>documentation: tutorial updates (John Fleck), benchmark results</li> 674 <li>xmlWriter: updates and fixes (Alfred Mickautsch, Lucas Brasilino)</li> 675 <li>XPath optimization (Petr Pajas)</li> 676 <li>DTD ID handling optimization</li> 677 <li>bugfixes: xpath number with > 19 fractional (William Brack), push 678 mode with unescaped '>' characters, fix xmllint --stream --timing, fix 679 xmllint --memory --stream memory usage, xmlAttrSerializeTxtContent 680 handling NULL, trying to fix Relax-NG/Perl interface.</li> 681 <li>python: 2.3 compatibility, whitespace fixes (Malcolm Tredinnick)</li> 682 <li>Added relaxng option to xmllint --shell</li> 683</ul> 684 685<h3>2.6.6: Feb 12 2004</h3> 686<ul> 687 <li>nanohttp and nanoftp: buffer overflow error on URI parsing (Igor and 688 William) reported by Yuuichi Teranishi</li> 689 <li>bugfixes: make test and path issues, xmlWriter attribute serialization 690 (William Brack), xmlWriter indentation (William), schemas validation 691 (Eric Haszlakiewicz), XInclude dictionnaries issues (William and Oleg 692 Paraschenko), XInclude empty fallback (William), HTML warnings (William), 693 XPointer in XInclude (William), Python namespace serialization, 694 isolat1ToUTF8 bound error (Alfred Mickautsch), output of parameter 695 entities in internal subset (William), internal subset bug in push mode, 696 <xs:all> fix (Alexey Sarytchev)</li> 697 <li>Build: fix for automake-1.8 (Alexander Winston), warnings removal 698 (Philip Ludlam), SOCKLEN_T detection fixes (Daniel Richard), fix 699 --with-minimum configuration.</li> 700 <li>XInclude: allow the 2001 namespace without warning.</li> 701 <li>Documentation: missing example/index.html (John Fleck), version 702 dependancies (John Fleck)</li> 703 <li>reader API: structured error reporting (Steve Ball)</li> 704 <li>Windows compilation: mingw, msys (Mikhail Grushinskiy), function 705 prototype (Cameron Johnson), MSVC6 compiler warnings, _WINSOCKAPI_ 706 patch</li> 707 <li>Parsers: added xmlByteConsumed(ctxt) API to get the byte offest in 708 input.</li> 709</ul> 710 711<h3>2.6.5: Jan 25 2004</h3> 712<ul> 713 <li>Bugfixes: dictionnaries for schemas (William Brack), regexp segfault 714 (William), xs:all problem (William), a number of XPointer bugfixes 715 (William), xmllint error go to stderr, DTD validation problem with 716 namespace, memory leak (William), SAX1 cleanup and minimal options fixes 717 (Mark Vadoc), parser context reset on error (Shaun McCance), XPath union 718 evaluation problem (William) , xmlReallocLoc with NULL (Aleksey Sanin), 719 XML Schemas double free (Steve Ball), XInclude with no href, argument 720 callbacks order for XPath callbacks (Frederic Peters)</li> 721 <li>Documentation: python scripts (William Brack), xslt stylesheets (John 722 Fleck), doc (Sven Zimmerman), I/O example.</li> 723 <li>Python bindings: fixes (William), enum support (St�phane Bidoul), 724 structured error reporting (St�phane Bidoul)</li> 725 <li>XInclude: various fixes for conformance, problem related to dictionnary 726 references (William & me), recursion (William)</li> 727 <li>xmlWriter: indentation (Lucas Brasilino), memory leaks (Alfred 728 Mickautsch),</li> 729 <li>xmlSchemas: normalizedString datatype (John Belmonte)</li> 730 <li>code cleanup for strings functions (William)</li> 731 <li>Windows: compiler patches (Mark Vakoc)</li> 732 <li>Parser optimizations, a few new XPath and dictionnary APIs for future 733 XSLT optimizations.</li> 734</ul> 735 736<h3>2.6.4: Dec 24 2003</h3> 737<ul> 738 <li>Windows build fixes (Igor Zlatkovic)</li> 739 <li>Some serious XInclude problems reported by Oleg Paraschenko and</li> 740 <li>Unix and Makefile packaging fixes (me, William Brack,</li> 741 <li>Documentation improvements (John Fleck, William Brack), example fix 742 (Lucas Brasilino)</li> 743 <li>bugfixes: xmlTextReaderExpand() with xmlReaderWalker, XPath handling of 744 NULL strings (William Brack) , API building reader or parser from 745 filedescriptor should not close it, changed XPath sorting to be stable 746 again (William Brack), xmlGetNodePath() generating '(null)' (William 747 Brack), DTD validation and namespace bug (William Brack), XML Schemas 748 double inclusion behaviour</li> 749</ul> 750 751<h3>2.6.3: Dec 10 2003</h3> 752<ul> 753 <li>documentation updates and cleanup (DV, William Brack, John Fleck)</li> 754 <li>added a repository of examples, examples from Aleksey Sanin, Dodji 755 Seketeli, Alfred Mickautsch</li> 756 <li>Windows updates: Mark Vakoc, Igor Zlatkovic, Eric Zurcher, Mingw 757 (Kenneth Haley)</li> 758 <li>Unicode range checking (William Brack)</li> 759 <li>code cleanup (William Brack)</li> 760 <li>Python bindings: doc (John Fleck), bug fixes</li> 761 <li>UTF-16 cleanup and BOM issues (William Brack)</li> 762 <li>bug fixes: ID and xmlReader validation, XPath (William Brack), 763 xmlWriter (Alfred Mickautsch), hash.h inclusion problem, HTML parser 764 (James Bursa), attribute defaulting and validation, some serialization 765 cleanups, XML_GET_LINE macro, memory debug when using threads (William 766 Brack), serialization of attributes and entities content, xmlWriter 767 (Daniel Schulman)</li> 768 <li>XInclude bugfix, new APIs and update to the last version including the 769 namespace change.</li> 770 <li>XML Schemas improvements: include (Robert Stepanek), import and 771 namespace handling, fixed the regression tests troubles, added examples 772 based on Eric van der Vlist book, regexp fixes</li> 773 <li>preliminary pattern support for streaming (needed for schemas 774 constraints), added xmlTextReaderPreservePattern() to collect subdocument 775 when streaming.</li> 776 <li>various fixes in the structured error handling</li> 777</ul> 778 779<h3>2.6.2: Nov 4 2003</h3> 780<ul> 781 <li>XPath context unregistration fixes</li> 782 <li>text node coalescing fixes (Mark Lilback)</li> 783 <li>API to screate a W3C Schemas from an existing document (Steve Ball)</li> 784 <li>BeOS patches (Marcin 'Shard' Konicki)</li> 785 <li>xmlStrVPrintf function added (Aleksey Sanin)</li> 786 <li>compilation fixes (Mark Vakoc)</li> 787 <li>stdin parsing fix (William Brack)</li> 788 <li>a posteriori DTD validation fixes</li> 789 <li>xmlReader bug fixes: Walker fixes, python bindings</li> 790 <li>fixed xmlStopParser() to really stop the parser and errors</li> 791 <li>always generate line numbers when using the new xmlReadxxx 792 functions</li> 793 <li>added XInclude support to the xmlReader interface</li> 794 <li>implemented XML_PARSE_NONET parser option</li> 795 <li>DocBook XSLT processing bug fixed</li> 796 <li>HTML serialization for <p> elements (William Brack and me)</li> 797 <li>XPointer failure in XInclude are now handled as resource errors</li> 798 <li>fixed xmllint --html to use the HTML serializer on output (added 799 --xmlout to implement the previous behaviour of saving it using the XML 800 serializer)</li> 801</ul> 802 803<h3>2.6.1: Oct 28 2003</h3> 804<ul> 805 <li>Mostly bugfixes after the big 2.6.0 changes</li> 806 <li>Unix compilation patches: libxml.m4 (Patrick Welche), warnings cleanup 807 (William Brack)</li> 808 <li>Windows compilation patches (Joachim Bauch, Stephane Bidoul, Igor 809 Zlatkovic)</li> 810 <li>xmlWriter bugfix (Alfred Mickautsch)</li> 811 <li>chvalid.[ch]: couple of fixes from Stephane Bidoul</li> 812 <li>context reset: error state reset, push parser reset (Graham 813 Bennett)</li> 814 <li>context reuse: generate errors if file is not readable</li> 815 <li>defaulted attributes for element coming from internal entities 816 (Stephane Bidoul)</li> 817 <li>Python: tab and spaces mix (William Brack)</li> 818 <li>Error handler could crash in DTD validation in 2.6.0</li> 819 <li>xmlReader: do not use the document or element _private field</li> 820 <li>testSAX.c: avoid a problem with some PIs (Massimo Morara)</li> 821 <li>general bug fixes: mandatory encoding in text decl, serializing 822 Document Fragment nodes, xmlSearchNs 2.6.0 problem (Kasimier Buchcik), 823 XPath errors not reported, slow HTML parsing of large documents.</li> 824</ul> 825 826<h3>2.6.0: Oct 20 2003</h3> 827<ul> 828 <li>Major revision release: should be API and ABI compatible but got a lot 829 of change</li> 830 <li>Increased the library modularity, far more options can be stripped out, 831 a --with-minimum configuration will weight around 160KBytes</li> 832 <li>Use per parser and per document dictionnary, allocate names and small 833 text nodes from the dictionnary</li> 834 <li>Switch to a SAX2 like parser rewrote most of the XML parser core, 835 provides namespace resolution and defaulted attributes, minimize memory 836 allocations and copies, namespace checking and specific error handling, 837 immutable buffers, make predefined entities static structures, etc...</li> 838 <li>rewrote all the error handling in the library, all errors can be 839 intercepted at a structured level, with precise information 840 available.</li> 841 <li>New simpler and more generic XML and HTML parser APIs, allowing to 842 easilly modify the parsing options and reuse parser context for multiple 843 consecutive documents.</li> 844 <li>Similar new APIs for the xmlReader, for options and reuse, provided new 845 functions to access content as const strings, use them for Python 846 bindings</li> 847 <li>a lot of other smaller API improvements: xmlStrPrintf (Aleksey Sanin), 848 Walker i.e. reader on a document tree based on Alfred Mickautsch code, 849 make room in nodes for line numbers, reference counting and future PSVI 850 extensions, generation of character ranges to be checked with faster 851 algorithm (William), xmlParserMaxDepth (Crutcher Dunnavant), buffer 852 access</li> 853 <li>New xmlWriter API provided by Alfred Mickautsch</li> 854 <li>Schemas: base64 support by Anthony Carrico</li> 855 <li>Parser<->HTTP integration fix, proper processing of the Mime-Type 856 and charset informations if available.</li> 857 <li>Relax-NG: bug fixes including the one reported by Martijn Faassen and 858 zeroOrMore, better error reporting.</li> 859 <li>Python bindings (St�phane Bidoul), never use stdout for errors 860 output</li> 861 <li>Portability: all the headers have macros for export and calling 862 convention definitions (Igor Zlatkovic), VMS update (Craig A. Berry), 863 Windows: threads (Jesse Pelton), Borland compiler (Eric Zurcher, Igor), 864 Mingw (Igor), typos (Mark Vakoc), beta version (Stephane Bidoul), 865 warning cleanups on AIX and MIPS compilers (William Brack), BeOS (Marcin 866 'Shard' Konicki)</li> 867 <li>Documentation fixes and README (William Brack), search fix (William), 868 tutorial updates (John Fleck), namespace docs (Stefan Kost)</li> 869 <li>Bug fixes: xmlCleanupParser (Dave Beckett), threading uninitialized 870 mutexes, HTML doctype lowercase, SAX/IO (William), compression detection 871 and restore (William), attribute declaration in DTDs (William), namespace 872 on attribute in HTML output (William), input filename (Rob Richards), 873 namespace DTD validation, xmlReplaceNode (Chris Ryland), I/O callbacks 874 (Markus Keim), CDATA serialization (Shaun McCance), xmlReader (Peter 875 Derr), high codepoint charref like &#x10FFFF;, buffer access in push 876 mode (Justin Fletcher), TLS threads on Windows (Jesse Pelton), XPath bug 877 (William), xmlCleanupParser (Marc Liyanage), CDATA output (William), HTTP 878 error handling.</li> 879 <li>xmllint options: --dtdvalidfpi for Tobias Reif, --sax1 for compat 880 testing, --nodict for building without tree dictionnary, --nocdata to 881 replace CDATA by text, --nsclean to remove surperfluous namespace 882 declarations</li> 883 <li>added xml2-config --libtool-libs option from Kevin P. Fleming</li> 884 <li>a lot of profiling and tuning of the code, speedup patch for 885 xmlSearchNs() by Luca Padovani. The xmlReader should do far less 886 allocation and it speed should get closer to SAX. Chris Anderson worked 887 on speeding and cleaning up repetitive checking code.</li> 888 <li>cleanup of "make tests"</li> 889 <li>libxml-2.0-uninstalled.pc from Malcolm Tredinnick</li> 890 <li>deactivated the broken docBook SGML parser code and plugged the XML 891 parser instead.</li> 892</ul> 893 894<h3>2.5.11: Sep 9 2003</h3> 895 896<p>A bugfix only release:</p> 897<ul> 898 <li>risk of crash in Relax-NG</li> 899 <li>risk of crash when using multithreaded programs</li> 900</ul> 901 902<h3>2.5.10: Aug 15 2003</h3> 903 904<p>A bugfixes only release</p> 905<ul> 906 <li>Windows Makefiles (William Brack)</li> 907 <li>UTF-16 support fixes (Mark Itzcovitz)</li> 908 <li>Makefile and portability (William Brack) automake, Linux alpha, Mingw 909 on Windows (Mikhail Grushinskiy)</li> 910 <li>HTML parser (Oliver Stoeneberg)</li> 911 <li>XInclude performance problem reported by Kevin Ruscoe</li> 912 <li>XML parser performance problem reported by Grant Goodale</li> 913 <li>xmlSAXParseDTD() bug fix from Malcolm Tredinnick</li> 914 <li>and a couple other cleanup</li> 915</ul> 916 917<h3>2.5.9: Aug 9 2003</h3> 918<ul> 919 <li>bugfixes: IPv6 portability, xmlHasNsProp (Markus Keim), Windows build 920 (Wiliam Brake, Jesse Pelton, Igor), Schemas (Peter Sobisch), threading 921 (Rob Richards), hexBinary type (), UTF-16 BOM (Dodji Seketeli), 922 xmlReader, Relax-NG schemas compilation, namespace handling, EXSLT (Sean 923 Griffin), HTML parsing problem (William Brack), DTD validation for mixed 924 content + namespaces, HTML serialization, library initialization, 925 progressive HTML parser</li> 926 <li>better interfaces for Relax-NG error handling (Joachim Bauch, )</li> 927 <li>adding xmlXIncludeProcessTree() for XInclud'ing in a subtree</li> 928 <li>doc fixes and improvements (John Fleck)</li> 929 <li>configure flag for -with-fexceptions when embedding in C++</li> 930 <li>couple of new UTF-8 helper functions (William Brack)</li> 931 <li>general encoding cleanup + ISO-8859-x without iconv (Peter Jacobi)</li> 932 <li>xmlTextReader cleanup + enum for node types (Bjorn Reese)</li> 933 <li>general compilation/warning cleanup Solaris/HP-UX/... (William 934 Brack)</li> 935</ul> 936 937<h3>2.5.8: Jul 6 2003</h3> 938<ul> 939 <li>bugfixes: XPath, XInclude, file/URI mapping, UTF-16 save (Mark 940 Itzcovitz), UTF-8 checking, URI saving, error printing (William Brack), 941 PI related memleak, compilation without schemas or without xpath (Joerg 942 Schmitz-Linneweber/Garry Pennington), xmlUnlinkNode problem with DTDs, 943 rpm problem on , i86_64, removed a few compilation problems from 2.5.7, 944 xmlIOParseDTD, and xmlSAXParseDTD (Malcolm Tredinnick)</li> 945 <li>portability: DJGPP (MsDos) , OpenVMS (Craig A. Berry)</li> 946 <li>William Brack fixed multithreading lock problems</li> 947 <li>IPv6 patch for FTP and HTTP accesses (Archana Shah/Wipro)</li> 948 <li>Windows fixes (Igor Zlatkovic, Eric Zurcher), threading (St�phane 949 Bidoul)</li> 950 <li>A few W3C Schemas Structure improvements</li> 951 <li>W3C Schemas Datatype improvements (Charlie Bozeman)</li> 952 <li>Python bindings for thread globals (St�phane Bidoul), and method/class 953 generator</li> 954 <li>added --nonet option to xmllint</li> 955 <li>documentation improvements (John Fleck)</li> 956</ul> 957 958<h3>2.5.7: Apr 25 2003</h3> 959<ul> 960 <li>Relax-NG: Compiling to regexp and streaming validation on top of the 961 xmlReader interface, added to xmllint --stream</li> 962 <li>xmlReader: Expand(), Next() and DOM access glue, bug fixes</li> 963 <li>Support for large files: RGN validated a 4.5GB instance</li> 964 <li>Thread support is now configured in by default</li> 965 <li>Fixes: update of the Trio code (Bjorn), WXS Date and Duration fixes 966 (Charles Bozeman), DTD and namespaces (Brent Hendricks), HTML push parser 967 and zero bytes handling, some missing Windows file path conversions, 968 behaviour of the parser and validator in the presence of "out of memory" 969 error conditions</li> 970 <li>extended the API to be able to plug a garbage collecting memory 971 allocator, added xmlMallocAtomic() and modified the allocations 972 accordingly.</li> 973 <li>Performances: removed excessive malloc() calls, speedup of the push and 974 xmlReader interfaces, removed excessive thread locking</li> 975 <li>Documentation: man page (John Fleck), xmlReader documentation</li> 976 <li>Python: adding binding for xmlCatalogAddLocal (Brent M Hendricks)</li> 977</ul> 978 979<h3>2.5.6: Apr 1 2003</h3> 980<ul> 981 <li>Fixed W3C XML Schemas datatype, should be compliant now except for 982 binHex and base64 which are not supported yet.</li> 983 <li>bug fixes: non-ASCII IDs, HTML output, XInclude on large docs and 984 XInclude entities handling, encoding detection on external subsets, XML 985 Schemas bugs and memory leaks, HTML parser (James Bursa)</li> 986 <li>portability: python/trio (Albert Chin), Sun compiler warnings</li> 987 <li>documentation: added --relaxng option to xmllint man page (John)</li> 988 <li>improved error reporting: xml:space, start/end tag mismatches, Relax NG 989 errors</li> 990</ul> 991 992<h3>2.5.5: Mar 24 2003</h3> 993<ul> 994 <li>Lot of fixes on the Relax NG implementation. More testing including 995 DocBook and TEI examples.</li> 996 <li>Increased the support for W3C XML Schemas datatype</li> 997 <li>Several bug fixes in the URI handling layer</li> 998 <li>Bug fixes: HTML parser, xmlReader, DTD validation, XPath, encoding 999 conversion, line counting in the parser.</li> 1000 <li>Added support for $XMLLINT_INDENT environment variable, FTP delete</li> 1001 <li>Fixed the RPM spec file name</li> 1002</ul> 1003 1004<h3>2.5.4: Feb 20 2003</h3> 1005<ul> 1006 <li>Conformance testing and lot of fixes on Relax NG and XInclude 1007 implementation</li> 1008 <li>Implementation of XPointer element() scheme</li> 1009 <li>Bug fixes: XML parser, XInclude entities merge, validity checking on 1010 namespaces, 1011 <p>2 serialization bugs, node info generation problems, a DTD regexp 1012 generation problem.</p> 1013 </li> 1014 <li>Portability: windows updates and path canonicalization (Igor)</li> 1015 <li>A few typo fixes (Kjartan Maraas)</li> 1016 <li>Python bindings generator fixes (Stephane Bidoul)</li> 1017</ul> 1018 1019<h3>2.5.3: Feb 10 2003</h3> 1020<ul> 1021 <li>RelaxNG and XML Schemas datatypes improvements, and added a first 1022 version of RelaxNG Python bindings</li> 1023 <li>Fixes: XLink (Sean Chittenden), XInclude (Sean Chittenden), API fix for 1024 serializing namespace nodes, encoding conversion bug, XHTML1 1025 serialization</li> 1026 <li>Portability fixes: Windows (Igor), AMD 64bits RPM spec file</li> 1027</ul> 1028 1029<h3>2.5.2: Feb 5 2003</h3> 1030<ul> 1031 <li>First implementation of RelaxNG, added --relaxng flag to xmllint</li> 1032 <li>Schemas support now compiled in by default.</li> 1033 <li>Bug fixes: DTD validation, namespace checking, XInclude and entities, 1034 delegateURI in XML Catalogs, HTML parser, XML reader (St�phane Bidoul), 1035 XPath parser and evaluation, UTF8ToUTF8 serialization, XML reader memory 1036 consumption, HTML parser, HTML serialization in the presence of 1037 namespaces</li> 1038 <li>added an HTML API to check elements and attributes.</li> 1039 <li>Documentation improvement, PDF for the tutorial (John Fleck), doc 1040 patches (Stefan Kost)</li> 1041 <li>Portability fixes: NetBSD (Julio Merino), Windows (Igor Zlatkovic)</li> 1042 <li>Added python bindings for XPointer, contextual error reporting 1043 (St�phane Bidoul)</li> 1044 <li>URI/file escaping problems (Stefano Zacchiroli)</li> 1045</ul> 1046 1047<h3>2.5.1: Jan 8 2003</h3> 1048<ul> 1049 <li>Fixes a memory leak and configuration/compilation problems in 2.5.0</li> 1050 <li>documentation updates (John)</li> 1051 <li>a couple of XmlTextReader fixes</li> 1052</ul> 1053 1054<h3>2.5.0: Jan 6 2003</h3> 1055<ul> 1056 <li>New <a href="xmlreader.html">XmltextReader interface</a> based on C# 1057 API (with help of St�phane Bidoul)</li> 1058 <li>Windows: more exports, including the new API (Igor)</li> 1059 <li>XInclude fallback fix</li> 1060 <li>Python: bindings for the new API, packaging (St�phane Bidoul), 1061 drv_libxml2.py Python xml.sax driver (St�phane Bidoul), fixes, speedup 1062 and iterators for Python-2.2 (Hannu Krosing)</li> 1063 <li>Tutorial fixes (john Fleck and Niraj Tolia) xmllint man update 1064 (John)</li> 1065 <li>Fix an XML parser bug raised by Vyacheslav Pindyura</li> 1066 <li>Fix for VMS serialization (Nigel Hall) and config (Craig A. Berry)</li> 1067 <li>Entities handling fixes</li> 1068 <li>new API to optionally track node creation and deletion (Lukas 1069 Schroeder)</li> 1070 <li>Added documentation for the XmltextReader interface and some <a 1071 href="guidelines.html">XML guidelines</a></li> 1072</ul> 1073 1074<h3>2.4.30: Dec 12 2002</h3> 1075<ul> 1076 <li>2.4.29 broke the python bindings, rereleasing</li> 1077 <li>Improvement/fixes of the XML API generator, and couple of minor code 1078 fixes.</li> 1079</ul> 1080 1081<h3>2.4.29: Dec 11 2002</h3> 1082<ul> 1083 <li>Windows fixes (Igor): Windows CE port, pthread linking, python bindings 1084 (St�phane Bidoul), Mingw (Magnus Henoch), and export list updates</li> 1085 <li>Fix for prev in python bindings (ERDI Gergo)</li> 1086 <li>Fix for entities handling (Marcus Clarke)</li> 1087 <li>Refactored the XML and HTML dumps to a single code path, fixed XHTML1 1088 dump</li> 1089 <li>Fix for URI parsing when handling URNs with fragment identifiers</li> 1090 <li>Fix for HTTP URL escaping problem</li> 1091 <li>added an TextXmlReader (C#) like API (work in progress)</li> 1092 <li>Rewrote the API in XML generation script, includes a C parser and saves 1093 more informations needed for C# bindings</li> 1094</ul> 1095 1096<h3>2.4.28: Nov 22 2002</h3> 1097<ul> 1098 <li>a couple of python binding fixes</li> 1099 <li>2 bug fixes in the XML push parser</li> 1100 <li>potential memory leak removed (Martin Stoilov)</li> 1101 <li>fix to the configure script for Unix (Dimitri Papadopoulos)</li> 1102 <li>added encoding support for XInclude parse="text"</li> 1103 <li>autodetection of XHTML1 and specific serialization rules added</li> 1104 <li>nasty threading bug fixed (William Brack)</li> 1105</ul> 1106 1107<h3>2.4.27: Nov 17 2002</h3> 1108<ul> 1109 <li>fixes for the Python bindings</li> 1110 <li>a number of bug fixes: SGML catalogs, xmlParseBalancedChunkMemory(), 1111 HTML parser, Schemas (Charles Bozeman), document fragment support 1112 (Christian Glahn), xmlReconciliateNs (Brian Stafford), XPointer, 1113 xmlFreeNode(), xmlSAXParseMemory (Peter Jones), xmlGetNodePath (Petr 1114 Pajas), entities processing</li> 1115 <li>added grep to xmllint --shell</li> 1116 <li>VMS update patch from Craig A. Berry</li> 1117 <li>cleanup of the Windows build with support for more compilers (Igor), 1118 better thread support on Windows</li> 1119 <li>cleanup of Unix Makefiles and spec file</li> 1120 <li>Improvements to the documentation (John Fleck)</li> 1121</ul> 1122 1123<h3>2.4.26: Oct 18 2002</h3> 1124<ul> 1125 <li>Patches for Windows CE port, improvements on Windows paths handling</li> 1126 <li>Fixes to the validation code (DTD and Schemas), xmlNodeGetPath() , 1127 HTML serialization, Namespace compliance, and a number of small 1128 problems</li> 1129</ul> 1130 1131<h3>2.4.25: Sep 26 2002</h3> 1132<ul> 1133 <li>A number of bug fixes: XPath, validation, Python bindings, DOM and 1134 tree, xmlI/O, Html</li> 1135 <li>Serious rewrite of XInclude</li> 1136 <li>Made XML Schemas regexp part of the default build and APIs, small fix 1137 and improvement of the regexp core</li> 1138 <li>Changed the validation code to reuse XML Schemas regexp APIs</li> 1139 <li>Better handling of Windows file paths, improvement of Makefiles (Igor, 1140 Daniel Gehriger, Mark Vakoc)</li> 1141 <li>Improved the python I/O bindings, the tests, added resolver and regexp 1142 APIs</li> 1143 <li>New logos from Marc Liyanage</li> 1144 <li>Tutorial improvements: John Fleck, Christopher Harris</li> 1145 <li>Makefile: Fixes for AMD x86_64 (Mandrake), DESTDIR (Christophe 1146 Merlet)</li> 1147 <li>removal of all stderr/perror use for error reporting</li> 1148 <li>Better error reporting: XPath and DTD validation</li> 1149 <li>update of the trio portability layer (Bjorn Reese)</li> 1150</ul> 1151 1152<p><strong>2.4.24: Aug 22 2002</strong></p> 1153<ul> 1154 <li>XPath fixes (William), xf:escape-uri() (Wesley Terpstra)</li> 1155 <li>Python binding fixes: makefiles (William), generator, rpm build, x86-64 1156 (fcrozat)</li> 1157 <li>HTML <style> and boolean attributes serializer fixes</li> 1158 <li>C14N improvements by Aleksey</li> 1159 <li>doc cleanups: Rick Jones</li> 1160 <li>Windows compiler makefile updates: Igor and Elizabeth Barham</li> 1161 <li>XInclude: implementation of fallback and xml:base fixup added</li> 1162</ul> 1163 1164<h3>2.4.23: July 6 2002</h3> 1165<ul> 1166 <li>performances patches: Peter Jacobi</li> 1167 <li>c14n fixes, testsuite and performances: Aleksey Sanin</li> 1168 <li>added xmlDocFormatDump: Chema Celorio</li> 1169 <li>new tutorial: John Fleck</li> 1170 <li>new hash functions and performances: Sander Vesik, portability fix from 1171 Peter Jacobi</li> 1172 <li>a number of bug fixes: XPath (William Brack, Richard Jinks), XML and 1173 HTML parsers, ID lookup function</li> 1174 <li>removal of all remaining sprintf: Aleksey Sanin</li> 1175</ul> 1176 1177<h3>2.4.22: May 27 2002</h3> 1178<ul> 1179 <li>a number of bug fixes: configure scripts, base handling, parser, memory 1180 usage, HTML parser, XPath, documentation (Christian Cornelssen), 1181 indentation, URI parsing</li> 1182 <li>Optimizations for XMLSec, fixing and making public some of the network 1183 protocol handlers (Aleksey)</li> 1184 <li>performance patch from Gary Pennington</li> 1185 <li>Charles Bozeman provided date and time support for XML Schemas 1186 datatypes</li> 1187</ul> 1188 1189<h3>2.4.21: Apr 29 2002</h3> 1190 1191<p>This release is both a bug fix release and also contains the early XML 1192Schemas <a href="http://www.w3.org/TR/xmlschema-1/">structures</a> and <a 1193href="http://www.w3.org/TR/xmlschema-2/">datatypes</a> code, beware, all 1194interfaces are likely to change, there is huge holes, it is clearly a work in 1195progress and don't even think of putting this code in a production system, 1196it's actually not compiled in by default. The real fixes are:</p> 1197<ul> 1198 <li>a couple of bugs or limitations introduced in 2.4.20</li> 1199 <li>patches for Borland C++ and MSC by Igor</li> 1200 <li>some fixes on XPath strings and conformance patches by Richard 1201 Jinks</li> 1202 <li>patch from Aleksey for the ExcC14N specification</li> 1203 <li>OSF/1 bug fix by Bjorn</li> 1204</ul> 1205 1206<h3>2.4.20: Apr 15 2002</h3> 1207<ul> 1208 <li>bug fixes: file descriptor leak, XPath, HTML output, DTD validation</li> 1209 <li>XPath conformance testing by Richard Jinks</li> 1210 <li>Portability fixes: Solaris, MPE/iX, Windows, OSF/1, python bindings, 1211 libxml.m4</li> 1212</ul> 1213 1214<h3>2.4.19: Mar 25 2002</h3> 1215<ul> 1216 <li>bug fixes: half a dozen XPath bugs, Validation, ISO-Latin to UTF8 1217 encoder</li> 1218 <li>portability fixes in the HTTP code</li> 1219 <li>memory allocation checks using valgrind, and profiling tests</li> 1220 <li>revamp of the Windows build and Makefiles</li> 1221</ul> 1222 1223<h3>2.4.18: Mar 18 2002</h3> 1224<ul> 1225 <li>bug fixes: tree, SAX, canonicalization, validation, portability, 1226 XPath</li> 1227 <li>removed the --with-buffer option it was becoming unmaintainable</li> 1228 <li>serious cleanup of the Python makefiles</li> 1229 <li>speedup patch to XPath very effective for DocBook stylesheets</li> 1230 <li>Fixes for Windows build, cleanup of the documentation</li> 1231</ul> 1232 1233<h3>2.4.17: Mar 8 2002</h3> 1234<ul> 1235 <li>a lot of bug fixes, including "namespace nodes have no parents in 1236 XPath"</li> 1237 <li>fixed/improved the Python wrappers, added more examples and more 1238 regression tests, XPath extension functions can now return node-sets</li> 1239 <li>added the XML Canonicalization support from Aleksey Sanin</li> 1240</ul> 1241 1242<h3>2.4.16: Feb 20 2002</h3> 1243<ul> 1244 <li>a lot of bug fixes, most of them were triggered by the XML Testsuite 1245 from OASIS and W3C. Compliance has been significantly improved.</li> 1246 <li>a couple of portability fixes too.</li> 1247</ul> 1248 1249<h3>2.4.15: Feb 11 2002</h3> 1250<ul> 1251 <li>Fixed the Makefiles, especially the python module ones</li> 1252 <li>A few bug fixes and cleanup</li> 1253 <li>Includes cleanup</li> 1254</ul> 1255 1256<h3>2.4.14: Feb 8 2002</h3> 1257<ul> 1258 <li>Change of License to the <a 1259 href="http://www.opensource.org/licenses/mit-license.html">MIT 1260 License</a> basically for integration in XFree86 codebase, and removing 1261 confusion around the previous dual-licensing</li> 1262 <li>added Python bindings, beta software but should already be quite 1263 complete</li> 1264 <li>a large number of fixes and cleanups, especially for all tree 1265 manipulations</li> 1266 <li>cleanup of the headers, generation of a reference API definition in 1267 XML</li> 1268</ul> 1269 1270<h3>2.4.13: Jan 14 2002</h3> 1271<ul> 1272 <li>update of the documentation: John Fleck and Charlie Bozeman</li> 1273 <li>cleanup of timing code from Justin Fletcher</li> 1274 <li>fixes for Windows and initial thread support on Win32: Igor and Serguei 1275 Narojnyi</li> 1276 <li>Cygwin patch from Robert Collins</li> 1277 <li>added xmlSetEntityReferenceFunc() for Keith Isdale work on xsldbg</li> 1278</ul> 1279 1280<h3>2.4.12: Dec 7 2001</h3> 1281<ul> 1282 <li>a few bug fixes: thread (Gary Pennington), xmllint (Geert Kloosterman), 1283 XML parser (Robin Berjon), XPointer (Danny Jamshy), I/O cleanups 1284 (robert)</li> 1285 <li>Eric Lavigne contributed project files for MacOS</li> 1286 <li>some makefiles cleanups</li> 1287</ul> 1288 1289<h3>2.4.11: Nov 26 2001</h3> 1290<ul> 1291 <li>fixed a couple of errors in the includes, fixed a few bugs, some code 1292 cleanups</li> 1293 <li>xmllint man pages improvement by Heiko Rupp</li> 1294 <li>updated VMS build instructions from John A Fotheringham</li> 1295 <li>Windows Makefiles updates from Igor</li> 1296</ul> 1297 1298<h3>2.4.10: Nov 10 2001</h3> 1299<ul> 1300 <li>URI escaping fix (Joel Young)</li> 1301 <li>added xmlGetNodePath() (for paths or XPointers generation)</li> 1302 <li>Fixes namespace handling problems when using DTD and validation</li> 1303 <li>improvements on xmllint: Morus Walter patches for --format and 1304 --encode, Stefan Kost and Heiko Rupp improvements on the --shell</li> 1305 <li>fixes for xmlcatalog linking pointed by Weiqi Gao</li> 1306 <li>fixes to the HTML parser</li> 1307</ul> 1308 1309<h3>2.4.9: Nov 6 2001</h3> 1310<ul> 1311 <li>fixes more catalog bugs</li> 1312 <li>avoid a compilation problem, improve xmlGetLineNo()</li> 1313</ul> 1314 1315<h3>2.4.8: Nov 4 2001</h3> 1316<ul> 1317 <li>fixed SGML catalogs broken in previous release, updated xmlcatalog 1318 tool</li> 1319 <li>fixed a compile errors and some includes troubles.</li> 1320</ul> 1321 1322<h3>2.4.7: Oct 30 2001</h3> 1323<ul> 1324 <li>exported some debugging interfaces</li> 1325 <li>serious rewrite of the catalog code</li> 1326 <li>integrated Gary Pennington thread safety patch, added configure option 1327 and regression tests</li> 1328 <li>removed an HTML parser bug</li> 1329 <li>fixed a couple of potentially serious validation bugs</li> 1330 <li>integrated the SGML DocBook support in xmllint</li> 1331 <li>changed the nanoftp anonymous login passwd</li> 1332 <li>some I/O cleanup and a couple of interfaces for Perl wrapper</li> 1333 <li>general bug fixes</li> 1334 <li>updated xmllint man page by John Fleck</li> 1335 <li>some VMS and Windows updates</li> 1336</ul> 1337 1338<h3>2.4.6: Oct 10 2001</h3> 1339<ul> 1340 <li>added an updated man pages by John Fleck</li> 1341 <li>portability and configure fixes</li> 1342 <li>an infinite loop on the HTML parser was removed (William)</li> 1343 <li>Windows makefile patches from Igor</li> 1344 <li>fixed half a dozen bugs reported for libxml or libxslt</li> 1345 <li>updated xmlcatalog to be able to modify SGML super catalogs</li> 1346</ul> 1347 1348<h3>2.4.5: Sep 14 2001</h3> 1349<ul> 1350 <li>Remove a few annoying bugs in 2.4.4</li> 1351 <li>forces the HTML serializer to output decimal charrefs since some 1352 version of Netscape can't handle hexadecimal ones</li> 1353</ul> 1354 1355<h3>1.8.16: Sep 14 2001</h3> 1356<ul> 1357 <li>maintenance release of the old libxml1 branch, couple of bug and 1358 portability fixes</li> 1359</ul> 1360 1361<h3>2.4.4: Sep 12 2001</h3> 1362<ul> 1363 <li>added --convert to xmlcatalog, bug fixes and cleanups of XML 1364 Catalog</li> 1365 <li>a few bug fixes and some portability changes</li> 1366 <li>some documentation cleanups</li> 1367</ul> 1368 1369<h3>2.4.3: Aug 23 2001</h3> 1370<ul> 1371 <li>XML Catalog support see the doc</li> 1372 <li>New NaN/Infinity floating point code</li> 1373 <li>A few bug fixes</li> 1374</ul> 1375 1376<h3>2.4.2: Aug 15 2001</h3> 1377<ul> 1378 <li>adds xmlLineNumbersDefault() to control line number generation</li> 1379 <li>lot of bug fixes</li> 1380 <li>the Microsoft MSC projects files should now be up to date</li> 1381 <li>inheritance of namespaces from DTD defaulted attributes</li> 1382 <li>fixes a serious potential security bug</li> 1383 <li>added a --format option to xmllint</li> 1384</ul> 1385 1386<h3>2.4.1: July 24 2001</h3> 1387<ul> 1388 <li>possibility to keep line numbers in the tree</li> 1389 <li>some computation NaN fixes</li> 1390 <li>extension of the XPath API</li> 1391 <li>cleanup for alpha and ia64 targets</li> 1392 <li>patch to allow saving through HTTP PUT or POST</li> 1393</ul> 1394 1395<h3>2.4.0: July 10 2001</h3> 1396<ul> 1397 <li>Fixed a few bugs in XPath, validation, and tree handling.</li> 1398 <li>Fixed XML Base implementation, added a couple of examples to the 1399 regression tests</li> 1400 <li>A bit of cleanup</li> 1401</ul> 1402 1403<h3>2.3.14: July 5 2001</h3> 1404<ul> 1405 <li>fixed some entities problems and reduce memory requirement when 1406 substituting them</li> 1407 <li>lots of improvements in the XPath queries interpreter can be 1408 substantially faster</li> 1409 <li>Makefiles and configure cleanups</li> 1410 <li>Fixes to XPath variable eval, and compare on empty node set</li> 1411 <li>HTML tag closing bug fixed</li> 1412 <li>Fixed an URI reference computation problem when validating</li> 1413</ul> 1414 1415<h3>2.3.13: June 28 2001</h3> 1416<ul> 1417 <li>2.3.12 configure.in was broken as well as the push mode XML parser</li> 1418 <li>a few more fixes for compilation on Windows MSC by Yon Derek</li> 1419</ul> 1420 1421<h3>1.8.14: June 28 2001</h3> 1422<ul> 1423 <li>Zbigniew Chyla gave a patch to use the old XML parser in push mode</li> 1424 <li>Small Makefile fix</li> 1425</ul> 1426 1427<h3>2.3.12: June 26 2001</h3> 1428<ul> 1429 <li>lots of cleanup</li> 1430 <li>a couple of validation fix</li> 1431 <li>fixed line number counting</li> 1432 <li>fixed serious problems in the XInclude processing</li> 1433 <li>added support for UTF8 BOM at beginning of entities</li> 1434 <li>fixed a strange gcc optimizer bugs in xpath handling of float, gcc-3.0 1435 miscompile uri.c (William), Thomas Leitner provided a fix for the 1436 optimizer on Tru64</li> 1437 <li>incorporated Yon Derek and Igor Zlatkovic fixes and improvements for 1438 compilation on Windows MSC</li> 1439 <li>update of libxml-doc.el (Felix Natter)</li> 1440 <li>fixed 2 bugs in URI normalization code</li> 1441</ul> 1442 1443<h3>2.3.11: June 17 2001</h3> 1444<ul> 1445 <li>updates to trio, Makefiles and configure should fix some portability 1446 problems (alpha)</li> 1447 <li>fixed some HTML serialization problems (pre, script, and block/inline 1448 handling), added encoding aware APIs, cleanup of this code</li> 1449 <li>added xmlHasNsProp()</li> 1450 <li>implemented a specific PI for encoding support in the DocBook SGML 1451 parser</li> 1452 <li>some XPath fixes (-Infinity, / as a function parameter and namespaces 1453 node selection)</li> 1454 <li>fixed a performance problem and an error in the validation code</li> 1455 <li>fixed XInclude routine to implement the recursive behaviour</li> 1456 <li>fixed xmlFreeNode problem when libxml is included statically twice</li> 1457 <li>added --version to xmllint for bug reports</li> 1458</ul> 1459 1460<h3>2.3.10: June 1 2001</h3> 1461<ul> 1462 <li>fixed the SGML catalog support</li> 1463 <li>a number of reported bugs got fixed, in XPath, iconv detection, 1464 XInclude processing</li> 1465 <li>XPath string function should now handle unicode correctly</li> 1466</ul> 1467 1468<h3>2.3.9: May 19 2001</h3> 1469 1470<p>Lots of bugfixes, and added a basic SGML catalog support:</p> 1471<ul> 1472 <li>HTML push bugfix #54891 and another patch from Jonas Borgstr�m</li> 1473 <li>some serious speed optimization again</li> 1474 <li>some documentation cleanups</li> 1475 <li>trying to get better linking on Solaris (-R)</li> 1476 <li>XPath API cleanup from Thomas Broyer</li> 1477 <li>Validation bug fixed #54631, added a patch from Gary Pennington, fixed 1478 xmlValidGetValidElements()</li> 1479 <li>Added an INSTALL file</li> 1480 <li>Attribute removal added to API: #54433</li> 1481 <li>added a basic support for SGML catalogs</li> 1482 <li>fixed xmlKeepBlanksDefault(0) API</li> 1483 <li>bugfix in xmlNodeGetLang()</li> 1484 <li>fixed a small configure portability problem</li> 1485 <li>fixed an inversion of SYSTEM and PUBLIC identifier in HTML document</li> 1486</ul> 1487 1488<h3>1.8.13: May 14 2001</h3> 1489<ul> 1490 <li>bugfixes release of the old libxml1 branch used by Gnome</li> 1491</ul> 1492 1493<h3>2.3.8: May 3 2001</h3> 1494<ul> 1495 <li>Integrated an SGML DocBook parser for the Gnome project</li> 1496 <li>Fixed a few things in the HTML parser</li> 1497 <li>Fixed some XPath bugs raised by XSLT use, tried to fix the floating 1498 point portability issue</li> 1499 <li>Speed improvement (8M/s for SAX, 3M/s for DOM, 1.5M/s for 1500 DOM+validation using the XML REC as input and a 700MHz celeron).</li> 1501 <li>incorporated more Windows cleanup</li> 1502 <li>added xmlSaveFormatFile()</li> 1503 <li>fixed problems in copying nodes with entities references (gdome)</li> 1504 <li>removed some troubles surrounding the new validation module</li> 1505</ul> 1506 1507<h3>2.3.7: April 22 2001</h3> 1508<ul> 1509 <li>lots of small bug fixes, corrected XPointer</li> 1510 <li>Non deterministic content model validation support</li> 1511 <li>added xmlDocCopyNode for gdome2</li> 1512 <li>revamped the way the HTML parser handles end of tags</li> 1513 <li>XPath: corrections of namespaces support and number formatting</li> 1514 <li>Windows: Igor Zlatkovic patches for MSC compilation</li> 1515 <li>HTML output fixes from P C Chow and William M. Brack</li> 1516 <li>Improved validation speed sensible for DocBook</li> 1517 <li>fixed a big bug with ID declared in external parsed entities</li> 1518 <li>portability fixes, update of Trio from Bjorn Reese</li> 1519</ul> 1520 1521<h3>2.3.6: April 8 2001</h3> 1522<ul> 1523 <li>Code cleanup using extreme gcc compiler warning options, found and 1524 cleared half a dozen potential problem</li> 1525 <li>the Eazel team found an XML parser bug</li> 1526 <li>cleaned up the user of some of the string formatting function. used the 1527 trio library code to provide the one needed when the platform is missing 1528 them</li> 1529 <li>xpath: removed a memory leak and fixed the predicate evaluation 1530 problem, extended the testsuite and cleaned up the result. XPointer seems 1531 broken ...</li> 1532</ul> 1533 1534<h3>2.3.5: Mar 23 2001</h3> 1535<ul> 1536 <li>Biggest change is separate parsing and evaluation of XPath expressions, 1537 there is some new APIs for this too</li> 1538 <li>included a number of bug fixes(XML push parser, 51876, notations, 1539 52299)</li> 1540 <li>Fixed some portability issues</li> 1541</ul> 1542 1543<h3>2.3.4: Mar 10 2001</h3> 1544<ul> 1545 <li>Fixed bugs #51860 and #51861</li> 1546 <li>Added a global variable xmlDefaultBufferSize to allow default buffer 1547 size to be application tunable.</li> 1548 <li>Some cleanup in the validation code, still a bug left and this part 1549 should probably be rewritten to support ambiguous content model :-\</li> 1550 <li>Fix a couple of serious bugs introduced or raised by changes in 2.3.3 1551 parser</li> 1552 <li>Fixed another bug in xmlNodeGetContent()</li> 1553 <li>Bjorn fixed XPath node collection and Number formatting</li> 1554 <li>Fixed a loop reported in the HTML parsing</li> 1555 <li>blank space are reported even if the Dtd content model proves that they 1556 are formatting spaces, this is for XML conformance</li> 1557</ul> 1558 1559<h3>2.3.3: Mar 1 2001</h3> 1560<ul> 1561 <li>small change in XPath for XSLT</li> 1562 <li>documentation cleanups</li> 1563 <li>fix in validation by Gary Pennington</li> 1564 <li>serious parsing performances improvements</li> 1565</ul> 1566 1567<h3>2.3.2: Feb 24 2001</h3> 1568<ul> 1569 <li>chasing XPath bugs, found a bunch, completed some TODO</li> 1570 <li>fixed a Dtd parsing bug</li> 1571 <li>fixed a bug in xmlNodeGetContent</li> 1572 <li>ID/IDREF support partly rewritten by Gary Pennington</li> 1573</ul> 1574 1575<h3>2.3.1: Feb 15 2001</h3> 1576<ul> 1577 <li>some XPath and HTML bug fixes for XSLT</li> 1578 <li>small extension of the hash table interfaces for DOM gdome2 1579 implementation</li> 1580 <li>A few bug fixes</li> 1581</ul> 1582 1583<h3>2.3.0: Feb 8 2001 (2.2.12 was on 25 Jan but I didn't kept track)</h3> 1584<ul> 1585 <li>Lots of XPath bug fixes</li> 1586 <li>Add a mode with Dtd lookup but without validation error reporting for 1587 XSLT</li> 1588 <li>Add support for text node without escaping (XSLT)</li> 1589 <li>bug fixes for xmlCheckFilename</li> 1590 <li>validation code bug fixes from Gary Pennington</li> 1591 <li>Patch from Paul D. Smith correcting URI path normalization</li> 1592 <li>Patch to allow simultaneous install of libxml-devel and 1593 libxml2-devel</li> 1594 <li>the example Makefile is now fixed</li> 1595 <li>added HTML to the RPM packages</li> 1596 <li>tree copying bugfixes</li> 1597 <li>updates to Windows makefiles</li> 1598 <li>optimization patch from Bjorn Reese</li> 1599</ul> 1600 1601<h3>2.2.11: Jan 4 2001</h3> 1602<ul> 1603 <li>bunch of bug fixes (memory I/O, xpath, ftp/http, ...)</li> 1604 <li>added htmlHandleOmittedElem()</li> 1605 <li>Applied Bjorn Reese's IPV6 first patch</li> 1606 <li>Applied Paul D. Smith patches for validation of XInclude results</li> 1607 <li>added XPointer xmlns() new scheme support</li> 1608</ul> 1609 1610<h3>2.2.10: Nov 25 2000</h3> 1611<ul> 1612 <li>Fix the Windows problems of 2.2.8</li> 1613 <li>integrate OpenVMS patches</li> 1614 <li>better handling of some nasty HTML input</li> 1615 <li>Improved the XPointer implementation</li> 1616 <li>integrate a number of provided patches</li> 1617</ul> 1618 1619<h3>2.2.9: Nov 25 2000</h3> 1620<ul> 1621 <li>erroneous release :-(</li> 1622</ul> 1623 1624<h3>2.2.8: Nov 13 2000</h3> 1625<ul> 1626 <li>First version of <a href="http://www.w3.org/TR/xinclude">XInclude</a> 1627 support</li> 1628 <li>Patch in conditional section handling</li> 1629 <li>updated MS compiler project</li> 1630 <li>fixed some XPath problems</li> 1631 <li>added an URI escaping function</li> 1632 <li>some other bug fixes</li> 1633</ul> 1634 1635<h3>2.2.7: Oct 31 2000</h3> 1636<ul> 1637 <li>added message redirection</li> 1638 <li>XPath improvements (thanks TOM !)</li> 1639 <li>xmlIOParseDTD() added</li> 1640 <li>various small fixes in the HTML, URI, HTTP and XPointer support</li> 1641 <li>some cleanup of the Makefile, autoconf and the distribution content</li> 1642</ul> 1643 1644<h3>2.2.6: Oct 25 2000:</h3> 1645<ul> 1646 <li>Added an hash table module, migrated a number of internal structure to 1647 those</li> 1648 <li>Fixed a posteriori validation problems</li> 1649 <li>HTTP module cleanups</li> 1650 <li>HTML parser improvements (tag errors, script/style handling, attribute 1651 normalization)</li> 1652 <li>coalescing of adjacent text nodes</li> 1653 <li>couple of XPath bug fixes, exported the internal API</li> 1654</ul> 1655 1656<h3>2.2.5: Oct 15 2000:</h3> 1657<ul> 1658 <li>XPointer implementation and testsuite</li> 1659 <li>Lot of XPath fixes, added variable and functions registration, more 1660 tests</li> 1661 <li>Portability fixes, lots of enhancements toward an easy Windows build 1662 and release</li> 1663 <li>Late validation fixes</li> 1664 <li>Integrated a lot of contributed patches</li> 1665 <li>added memory management docs</li> 1666 <li>a performance problem when using large buffer seems fixed</li> 1667</ul> 1668 1669<h3>2.2.4: Oct 1 2000:</h3> 1670<ul> 1671 <li>main XPath problem fixed</li> 1672 <li>Integrated portability patches for Windows</li> 1673 <li>Serious bug fixes on the URI and HTML code</li> 1674</ul> 1675 1676<h3>2.2.3: Sep 17 2000</h3> 1677<ul> 1678 <li>bug fixes</li> 1679 <li>cleanup of entity handling code</li> 1680 <li>overall review of all loops in the parsers, all sprintf usage has been 1681 checked too</li> 1682 <li>Far better handling of larges Dtd. Validating against DocBook XML Dtd 1683 works smoothly now.</li> 1684</ul> 1685 1686<h3>1.8.10: Sep 6 2000</h3> 1687<ul> 1688 <li>bug fix release for some Gnome projects</li> 1689</ul> 1690 1691<h3>2.2.2: August 12 2000</h3> 1692<ul> 1693 <li>mostly bug fixes</li> 1694 <li>started adding routines to access xml parser context options</li> 1695</ul> 1696 1697<h3>2.2.1: July 21 2000</h3> 1698<ul> 1699 <li>a purely bug fixes release</li> 1700 <li>fixed an encoding support problem when parsing from a memory block</li> 1701 <li>fixed a DOCTYPE parsing problem</li> 1702 <li>removed a bug in the function allowing to override the memory 1703 allocation routines</li> 1704</ul> 1705 1706<h3>2.2.0: July 14 2000</h3> 1707<ul> 1708 <li>applied a lot of portability fixes</li> 1709 <li>better encoding support/cleanup and saving (content is now always 1710 encoded in UTF-8)</li> 1711 <li>the HTML parser now correctly handles encodings</li> 1712 <li>added xmlHasProp()</li> 1713 <li>fixed a serious problem with &#38;</li> 1714 <li>propagated the fix to FTP client</li> 1715 <li>cleanup, bugfixes, etc ...</li> 1716 <li>Added a page about <a href="encoding.html">libxml Internationalization 1717 support</a></li> 1718</ul> 1719 1720<h3>1.8.9: July 9 2000</h3> 1721<ul> 1722 <li>fixed the spec the RPMs should be better</li> 1723 <li>fixed a serious bug in the FTP implementation, released 1.8.9 to solve 1724 rpmfind users problem</li> 1725</ul> 1726 1727<h3>2.1.1: July 1 2000</h3> 1728<ul> 1729 <li>fixes a couple of bugs in the 2.1.0 packaging</li> 1730 <li>improvements on the HTML parser</li> 1731</ul> 1732 1733<h3>2.1.0 and 1.8.8: June 29 2000</h3> 1734<ul> 1735 <li>1.8.8 is mostly a commodity package for upgrading to libxml2 according 1736 to <a href="upgrade.html">new instructions</a>. It fixes a nasty problem 1737 about &#38; charref parsing</li> 1738 <li>2.1.0 also ease the upgrade from libxml v1 to the recent version. it 1739 also contains numerous fixes and enhancements: 1740 <ul> 1741 <li>added xmlStopParser() to stop parsing</li> 1742 <li>improved a lot parsing speed when there is large CDATA blocs</li> 1743 <li>includes XPath patches provided by Picdar Technology</li> 1744 <li>tried to fix as much as possible DTD validation and namespace 1745 related problems</li> 1746 <li>output to a given encoding has been added/tested</li> 1747 <li>lot of various fixes</li> 1748 </ul> 1749 </li> 1750</ul> 1751 1752<h3>2.0.0: Apr 12 2000</h3> 1753<ul> 1754 <li>First public release of libxml2. If you are using libxml, it's a good 1755 idea to check the 1.x to 2.x upgrade instructions. NOTE: while initially 1756 scheduled for Apr 3 the release occurred only on Apr 12 due to massive 1757 workload.</li> 1758 <li>The include are now located under $prefix/include/libxml (instead of 1759 $prefix/include/gnome-xml), they also are referenced by 1760 <pre>#include <libxml/xxx.h></pre> 1761 <p>instead of</p> 1762 <pre>#include "xxx.h"</pre> 1763 </li> 1764 <li>a new URI module for parsing URIs and following strictly RFC 2396</li> 1765 <li>the memory allocation routines used by libxml can now be overloaded 1766 dynamically by using xmlMemSetup()</li> 1767 <li>The previously CVS only tool tester has been renamed 1768 <strong>xmllint</strong> and is now installed as part of the libxml2 1769 package</li> 1770 <li>The I/O interface has been revamped. There is now ways to plug in 1771 specific I/O modules, either at the URI scheme detection level using 1772 xmlRegisterInputCallbacks() or by passing I/O functions when creating a 1773 parser context using xmlCreateIOParserCtxt()</li> 1774 <li>there is a C preprocessor macro LIBXML_VERSION providing the version 1775 number of the libxml module in use</li> 1776 <li>a number of optional features of libxml can now be excluded at 1777 configure time (FTP/HTTP/HTML/XPath/Debug)</li> 1778</ul> 1779 1780<h3>2.0.0beta: Mar 14 2000</h3> 1781<ul> 1782 <li>This is a first Beta release of libxml version 2</li> 1783 <li>It's available only from<a href="ftp://xmlsoft.org/">xmlsoft.org 1784 FTP</a>, it's packaged as libxml2-2.0.0beta and available as tar and 1785 RPMs</li> 1786 <li>This version is now the head in the Gnome CVS base, the old one is 1787 available under the tag LIB_XML_1_X</li> 1788 <li>This includes a very large set of changes. From a programmatic point 1789 of view applications should not have to be modified too much, check the 1790 <a href="upgrade.html">upgrade page</a></li> 1791 <li>Some interfaces may changes (especially a bit about encoding).</li> 1792 <li>the updates includes: 1793 <ul> 1794 <li>fix I18N support. ISO-Latin-x/UTF-8/UTF-16 (nearly) seems correctly 1795 handled now</li> 1796 <li>Better handling of entities, especially well-formedness checking 1797 and proper PEref extensions in external subsets</li> 1798 <li>DTD conditional sections</li> 1799 <li>Validation now correctly handle entities content</li> 1800 <li><a href="http://rpmfind.net/tools/gdome/messages/0039.html">change 1801 structures to accommodate DOM</a></li> 1802 </ul> 1803 </li> 1804 <li>Serious progress were made toward compliance, <a 1805 href="conf/result.html">here are the result of the test</a> against the 1806 OASIS testsuite (except the Japanese tests since I don't support that 1807 encoding yet). This URL is rebuilt every couple of hours using the CVS 1808 head version.</li> 1809</ul> 1810 1811<h3>1.8.7: Mar 6 2000</h3> 1812<ul> 1813 <li>This is a bug fix release:</li> 1814 <li>It is possible to disable the ignorable blanks heuristic used by 1815 libxml-1.x, a new function xmlKeepBlanksDefault(0) will allow this. Note 1816 that for adherence to XML spec, this behaviour will be disabled by 1817 default in 2.x . The same function will allow to keep compatibility for 1818 old code.</li> 1819 <li>Blanks in <a> </a> constructs are not ignored anymore, 1820 avoiding heuristic is really the Right Way :-\</li> 1821 <li>The unchecked use of snprintf which was breaking libxml-1.8.6 1822 compilation on some platforms has been fixed</li> 1823 <li>nanoftp.c nanohttp.c: Fixed '#' and '?' stripping when processing 1824 URIs</li> 1825</ul> 1826 1827<h3>1.8.6: Jan 31 2000</h3> 1828<ul> 1829 <li>added a nanoFTP transport module, debugged until the new version of <a 1830 href="http://rpmfind.net/linux/rpm2html/rpmfind.html">rpmfind</a> can use 1831 it without troubles</li> 1832</ul> 1833 1834<h3>1.8.5: Jan 21 2000</h3> 1835<ul> 1836 <li>adding APIs to parse a well balanced chunk of XML (production <a 1837 href="http://www.w3.org/TR/REC-xml#NT-content">[43] content</a> of the 1838 XML spec)</li> 1839 <li>fixed a hideous bug in xmlGetProp pointed by Rune.Djurhuus@fast.no</li> 1840 <li>Jody Goldberg <jgoldberg@home.com> provided another patch trying 1841 to solve the zlib checks problems</li> 1842 <li>The current state in gnome CVS base is expected to ship as 1.8.5 with 1843 gnumeric soon</li> 1844</ul> 1845 1846<h3>1.8.4: Jan 13 2000</h3> 1847<ul> 1848 <li>bug fixes, reintroduced xmlNewGlobalNs(), fixed xmlNewNs()</li> 1849 <li>all exit() call should have been removed from libxml</li> 1850 <li>fixed a problem with INCLUDE_WINSOCK on WIN32 platform</li> 1851 <li>added newDocFragment()</li> 1852</ul> 1853 1854<h3>1.8.3: Jan 5 2000</h3> 1855<ul> 1856 <li>a Push interface for the XML and HTML parsers</li> 1857 <li>a shell-like interface to the document tree (try tester --shell :-)</li> 1858 <li>lots of bug fixes and improvement added over XMas holidays</li> 1859 <li>fixed the DTD parsing code to work with the xhtml DTD</li> 1860 <li>added xmlRemoveProp(), xmlRemoveID() and xmlRemoveRef()</li> 1861 <li>Fixed bugs in xmlNewNs()</li> 1862 <li>External entity loading code has been revamped, now it uses 1863 xmlLoadExternalEntity(), some fix on entities processing were added</li> 1864 <li>cleaned up WIN32 includes of socket stuff</li> 1865</ul> 1866 1867<h3>1.8.2: Dec 21 1999</h3> 1868<ul> 1869 <li>I got another problem with includes and C++, I hope this issue is fixed 1870 for good this time</li> 1871 <li>Added a few tree modification functions: xmlReplaceNode, 1872 xmlAddPrevSibling, xmlAddNextSibling, xmlNodeSetName and 1873 xmlDocSetRootElement</li> 1874 <li>Tried to improve the HTML output with help from <a 1875 href="mailto:clahey@umich.edu">Chris Lahey</a></li> 1876</ul> 1877 1878<h3>1.8.1: Dec 18 1999</h3> 1879<ul> 1880 <li>various patches to avoid troubles when using libxml with C++ compilers 1881 the "namespace" keyword and C escaping in include files</li> 1882 <li>a problem in one of the core macros IS_CHAR was corrected</li> 1883 <li>fixed a bug introduced in 1.8.0 breaking default namespace processing, 1884 and more specifically the Dia application</li> 1885 <li>fixed a posteriori validation (validation after parsing, or by using a 1886 Dtd not specified in the original document)</li> 1887 <li>fixed a bug in</li> 1888</ul> 1889 1890<h3>1.8.0: Dec 12 1999</h3> 1891<ul> 1892 <li>cleanup, especially memory wise</li> 1893 <li>the parser should be more reliable, especially the HTML one, it should 1894 not crash, whatever the input !</li> 1895 <li>Integrated various patches, especially a speedup improvement for large 1896 dataset from <a href="mailto:cnygard@bellatlantic.net">Carl Nygard</a>, 1897 configure with --with-buffers to enable them.</li> 1898 <li>attribute normalization, oops should have been added long ago !</li> 1899 <li>attributes defaulted from DTDs should be available, xmlSetProp() now 1900 does entities escaping by default.</li> 1901</ul> 1902 1903<h3>1.7.4: Oct 25 1999</h3> 1904<ul> 1905 <li>Lots of HTML improvement</li> 1906 <li>Fixed some errors when saving both XML and HTML</li> 1907 <li>More examples, the regression tests should now look clean</li> 1908 <li>Fixed a bug with contiguous charref</li> 1909</ul> 1910 1911<h3>1.7.3: Sep 29 1999</h3> 1912<ul> 1913 <li>portability problems fixed</li> 1914 <li>snprintf was used unconditionally, leading to link problems on system 1915 were it's not available, fixed</li> 1916</ul> 1917 1918<h3>1.7.1: Sep 24 1999</h3> 1919<ul> 1920 <li>The basic type for strings manipulated by libxml has been renamed in 1921 1.7.1 from <strong>CHAR</strong> to <strong>xmlChar</strong>. The reason 1922 is that CHAR was conflicting with a predefined type on Windows. However 1923 on non WIN32 environment, compatibility is provided by the way of a 1924 <strong>#define </strong>.</li> 1925 <li>Changed another error : the use of a structure field called errno, and 1926 leading to troubles on platforms where it's a macro</li> 1927</ul> 1928 1929<h3>1.7.0: Sep 23 1999</h3> 1930<ul> 1931 <li>Added the ability to fetch remote DTD or parsed entities, see the <a 1932 href="html/libxml-nanohttp.html">nanohttp</a> module.</li> 1933 <li>Added an errno to report errors by another mean than a simple printf 1934 like callback</li> 1935 <li>Finished ID/IDREF support and checking when validation</li> 1936 <li>Serious memory leaks fixed (there is now a <a 1937 href="html/libxml-xmlmemory.html">memory wrapper</a> module)</li> 1938 <li>Improvement of <a href="http://www.w3.org/TR/xpath">XPath</a> 1939 implementation</li> 1940 <li>Added an HTML parser front-end</li> 1941</ul> 1942 1943<h2><a name="XML">XML</a></h2> 1944 1945<p><a href="http://www.w3.org/TR/REC-xml">XML is a standard</a> for 1946markup-based structured documents. Here is <a name="example">an example XML 1947document</a>:</p> 1948<pre><?xml version="1.0"?> 1949<EXAMPLE prop1="gnome is great" prop2="&amp; linux too"> 1950 <head> 1951 <title>Welcome to Gnome</title> 1952 </head> 1953 <chapter> 1954 <title>The Linux adventure</title> 1955 <p>bla bla bla ...</p> 1956 <image href="linus.gif"/> 1957 <p>...</p> 1958 </chapter> 1959</EXAMPLE></pre> 1960 1961<p>The first line specifies that it is an XML document and gives useful 1962information about its encoding. Then the rest of the document is a text 1963format whose structure is specified by tags between brackets. <strong>Each 1964tag opened has to be closed</strong>. XML is pedantic about this. However, if 1965a tag is empty (no content), a single tag can serve as both the opening and 1966closing tag if it ends with <code>/></code> rather than with 1967<code>></code>. Note that, for example, the image tag has no content (just 1968an attribute) and is closed by ending the tag with <code>/></code>.</p> 1969 1970<p>XML can be applied successfully to a wide range of tasks, ranging from 1971long term structured document maintenance (where it follows the steps of 1972SGML) to simple data encoding mechanisms like configuration file formatting 1973(glade), spreadsheets (gnumeric), or even shorter lived documents such as 1974WebDAV where it is used to encode remote calls between a client and a 1975server.</p> 1976 1977<h2><a name="XSLT">XSLT</a></h2> 1978 1979<p>Check <a href="http://xmlsoft.org/XSLT">the separate libxslt page</a></p> 1980 1981<p><a href="http://www.w3.org/TR/xslt">XSL Transformations</a>, is a 1982language for transforming XML documents into other XML documents (or 1983HTML/textual output).</p> 1984 1985<p>A separate library called libxslt is available implementing XSLT-1.0 for 1986libxml2. This module "libxslt" too can be found in the Gnome CVS base.</p> 1987 1988<p>You can check the <a 1989href="http://cvs.gnome.org/lxr/source/libxslt/FEATURES">features</a> 1990supported and the progresses on the <a 1991href="http://cvs.gnome.org/lxr/source/libxslt/ChangeLog" 1992name="Changelog">Changelog</a>.</p> 1993 1994<h2><a name="Python">Python and bindings</a></h2> 1995 1996<p>There are a number of language bindings and wrappers available for 1997libxml2, the list below is not exhaustive. Please contact the <a 1998href="http://mail.gnome.org/mailman/listinfo/xml-bindings">xml-bindings@gnome.org</a> 1999(<a href="http://mail.gnome.org/archives/xml-bindings/">archives</a>) in 2000order to get updates to this list or to discuss the specific topic of libxml2 2001or libxslt wrappers or bindings:</p> 2002<ul> 2003 <li><a href="http://libxmlplusplus.sourceforge.net/">Libxml++</a> seems the 2004 most up-to-date C++ bindings for libxml2, check the <a 2005 href="http://libxmlplusplus.sourceforge.net/reference/html/hierarchy.html">documentation</a> 2006 and the <a 2007 href="http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/libxmlplusplus/libxml%2b%2b/examples/">examples</a>.</li> 2008 <li>There is another <a href="http://libgdome-cpp.berlios.de/">C++ wrapper 2009 based on the gdome2 bindings</a> maintained by Tobias Peters.</li> 2010 <li>and a third C++ wrapper by Peter Jones <pjones@pmade.org> 2011 <p>Website: <a 2012 href="http://pmade.org/pjones/software/xmlwrapp/">http://pmade.org/pjones/software/xmlwrapp/</a></p> 2013 </li> 2014 <li><a 2015 href="http://mail.gnome.org/archives/xml/2001-March/msg00014.html">Matt 2016 Sergeant</a> developed <a 2017 href="http://axkit.org/download/">XML::LibXSLT</a>, a Perl wrapper for 2018 libxml2/libxslt as part of the <a href="http://axkit.com/">AxKit XML 2019 application server</a>.</li> 2020 <li>If you're interested into scripting XML processing, have a look at <a 2021 href="http://xsh.sourceforge.net/">XSH</a> an XML editing shell based on 2022 Libxml2 Perl bindings.</li> 2023 <li><a href="mailto:dkuhlman@cutter.rexx.com">Dave Kuhlman</a> provides an 2024 earlier version of the libxml/libxslt <a 2025 href="http://www.rexx.com/~dkuhlman">wrappers for Python</a>.</li> 2026 <li>Gopal.V and Peter Minten develop <a 2027 href="http://savannah.gnu.org/projects/libxmlsharp">libxml#</a>, a set of 2028 C# libxml2 bindings.</li> 2029 <li>Petr Kozelka provides <a 2030 href="http://sourceforge.net/projects/libxml2-pas">Pascal units to glue 2031 libxml2</a> with Kylix, Delphi and other Pascal compilers.</li> 2032 <li>Uwe Fechner also provides <a 2033 href="http://sourceforge.net/projects/idom2-pas/">idom2</a>, a DOM2 2034 implementation for Kylix2/D5/D6 from Borland.</li> 2035 <li>Wai-Sun "Squidster" Chia provides <a 2036 href="http://www.rubycolor.org/arc/redist/">bindings for Ruby</a> and 2037 libxml2 bindings are also available in Ruby through the <a 2038 href="http://libgdome-ruby.berlios.de/">libgdome-ruby</a> module 2039 maintained by Tobias Peters.</li> 2040 <li>Steve Ball and contributors maintains <a 2041 href="http://tclxml.sourceforge.net/">libxml2 and libxslt bindings for 2042 Tcl</a>.</li> 2043 <li>There is support for libxml2 in the DOM module of PHP.</li> 2044 <li><a href="http://savannah.gnu.org/projects/classpathx/">LibxmlJ</a> is 2045 an effort to create a 100% JAXP-compatible Java wrapper for libxml2 and 2046 libxslt as part of GNU ClasspathX project.</li> 2047 <li>Patrick McPhee provides Rexx bindings fof libxml2 and libxslt, look for 2048 <a href="http://www.interlog.com/~ptjm/software.html">RexxXML</a>.</li> 2049</ul> 2050 2051<p>The distribution includes a set of Python bindings, which are guaranteed 2052to be maintained as part of the library in the future, though the Python 2053interface have not yet reached the completeness of the C API.</p> 2054 2055<p><a href="mailto:stephane.bidoul@softwareag.com">St�phane Bidoul</a> 2056maintains <a href="http://users.skynet.be/sbi/libxml-python/">a Windows port 2057of the Python bindings</a>.</p> 2058 2059<p>Note to people interested in building bindings, the API is formalized as 2060<a href="libxml2-api.xml">an XML API description file</a> which allows to 2061automate a large part of the Python bindings, this includes function 2062descriptions, enums, structures, typedefs, etc... The Python script used to 2063build the bindings is python/generator.py in the source distribution.</p> 2064 2065<p>To install the Python bindings there are 2 options:</p> 2066<ul> 2067 <li>If you use an RPM based distribution, simply install the <a 2068 href="http://rpmfind.net/linux/rpm2html/search.php?query=libxml2-python">libxml2-python 2069 RPM</a> (and if needed the <a 2070 href="http://rpmfind.net/linux/rpm2html/search.php?query=libxslt-python">libxslt-python 2071 RPM</a>).</li> 2072 <li>Otherwise use the <a href="ftp://xmlsoft.org/python/">libxml2-python 2073 module distribution</a> corresponding to your installed version of 2074 libxml2 and libxslt. Note that to install it you will need both libxml2 2075 and libxslt installed and run "python setup.py build install" in the 2076 module tree.</li> 2077</ul> 2078 2079<p>The distribution includes a set of examples and regression tests for the 2080python bindings in the <code>python/tests</code> directory. Here are some 2081excerpts from those tests:</p> 2082 2083<h3>tst.py:</h3> 2084 2085<p>This is a basic test of the file interface and DOM navigation:</p> 2086<pre>import libxml2, sys 2087 2088doc = libxml2.parseFile("tst.xml") 2089if doc.name != "tst.xml": 2090 print "doc.name failed" 2091 sys.exit(1) 2092root = doc.children 2093if root.name != "doc": 2094 print "root.name failed" 2095 sys.exit(1) 2096child = root.children 2097if child.name != "foo": 2098 print "child.name failed" 2099 sys.exit(1) 2100doc.freeDoc()</pre> 2101 2102<p>The Python module is called libxml2; parseFile is the equivalent of 2103xmlParseFile (most of the bindings are automatically generated, and the xml 2104prefix is removed and the casing convention are kept). All node seen at the 2105binding level share the same subset of accessors:</p> 2106<ul> 2107 <li><code>name</code> : returns the node name</li> 2108 <li><code>type</code> : returns a string indicating the node type</li> 2109 <li><code>content</code> : returns the content of the node, it is based on 2110 xmlNodeGetContent() and hence is recursive.</li> 2111 <li><code>parent</code> , <code>children</code>, <code>last</code>, 2112 <code>next</code>, <code>prev</code>, <code>doc</code>, 2113 <code>properties</code>: pointing to the associated element in the tree, 2114 those may return None in case no such link exists.</li> 2115</ul> 2116 2117<p>Also note the need to explicitly deallocate documents with freeDoc() . 2118Reference counting for libxml2 trees would need quite a lot of work to 2119function properly, and rather than risk memory leaks if not implemented 2120correctly it sounds safer to have an explicit function to free a tree. The 2121wrapper python objects like doc, root or child are them automatically garbage 2122collected.</p> 2123 2124<h3>validate.py:</h3> 2125 2126<p>This test check the validation interfaces and redirection of error 2127messages:</p> 2128<pre>import libxml2 2129 2130#deactivate error messages from the validation 2131def noerr(ctx, str): 2132 pass 2133 2134libxml2.registerErrorHandler(noerr, None) 2135 2136ctxt = libxml2.createFileParserCtxt("invalid.xml") 2137ctxt.validate(1) 2138ctxt.parseDocument() 2139doc = ctxt.doc() 2140valid = ctxt.isValid() 2141doc.freeDoc() 2142if valid != 0: 2143 print "validity check failed"</pre> 2144 2145<p>The first thing to notice is the call to registerErrorHandler(), it 2146defines a new error handler global to the library. It is used to avoid seeing 2147the error messages when trying to validate the invalid document.</p> 2148 2149<p>The main interest of that test is the creation of a parser context with 2150createFileParserCtxt() and how the behaviour can be changed before calling 2151parseDocument() . Similarly the informations resulting from the parsing phase 2152are also available using context methods.</p> 2153 2154<p>Contexts like nodes are defined as class and the libxml2 wrappers maps the 2155C function interfaces in terms of objects method as much as possible. The 2156best to get a complete view of what methods are supported is to look at the 2157libxml2.py module containing all the wrappers.</p> 2158 2159<h3>push.py:</h3> 2160 2161<p>This test show how to activate the push parser interface:</p> 2162<pre>import libxml2 2163 2164ctxt = libxml2.createPushParser(None, "<foo", 4, "test.xml") 2165ctxt.parseChunk("/>", 2, 1) 2166doc = ctxt.doc() 2167 2168doc.freeDoc()</pre> 2169 2170<p>The context is created with a special call based on the 2171xmlCreatePushParser() from the C library. The first argument is an optional 2172SAX callback object, then the initial set of data, the length and the name of 2173the resource in case URI-References need to be computed by the parser.</p> 2174 2175<p>Then the data are pushed using the parseChunk() method, the last call 2176setting the third argument terminate to 1.</p> 2177 2178<h3>pushSAX.py:</h3> 2179 2180<p>this test show the use of the event based parsing interfaces. In this case 2181the parser does not build a document, but provides callback information as 2182the parser makes progresses analyzing the data being provided:</p> 2183<pre>import libxml2 2184log = "" 2185 2186class callback: 2187 def startDocument(self): 2188 global log 2189 log = log + "startDocument:" 2190 2191 def endDocument(self): 2192 global log 2193 log = log + "endDocument:" 2194 2195 def startElement(self, tag, attrs): 2196 global log 2197 log = log + "startElement %s %s:" % (tag, attrs) 2198 2199 def endElement(self, tag): 2200 global log 2201 log = log + "endElement %s:" % (tag) 2202 2203 def characters(self, data): 2204 global log 2205 log = log + "characters: %s:" % (data) 2206 2207 def warning(self, msg): 2208 global log 2209 log = log + "warning: %s:" % (msg) 2210 2211 def error(self, msg): 2212 global log 2213 log = log + "error: %s:" % (msg) 2214 2215 def fatalError(self, msg): 2216 global log 2217 log = log + "fatalError: %s:" % (msg) 2218 2219handler = callback() 2220 2221ctxt = libxml2.createPushParser(handler, "<foo", 4, "test.xml") 2222chunk = " url='tst'>b" 2223ctxt.parseChunk(chunk, len(chunk), 0) 2224chunk = "ar</foo>" 2225ctxt.parseChunk(chunk, len(chunk), 1) 2226 2227reference = "startDocument:startElement foo {'url': 'tst'}:" + \ 2228 "characters: bar:endElement foo:endDocument:" 2229if log != reference: 2230 print "Error got: %s" % log 2231 print "Expected: %s" % reference</pre> 2232 2233<p>The key object in that test is the handler, it provides a number of entry 2234points which can be called by the parser as it makes progresses to indicate 2235the information set obtained. The full set of callback is larger than what 2236the callback class in that specific example implements (see the SAX 2237definition for a complete list). The wrapper will only call those supplied by 2238the object when activated. The startElement receives the names of the element 2239and a dictionary containing the attributes carried by this element.</p> 2240 2241<p>Also note that the reference string generated from the callback shows a 2242single character call even though the string "bar" is passed to the parser 2243from 2 different call to parseChunk()</p> 2244 2245<h3>xpath.py:</h3> 2246 2247<p>This is a basic test of XPath wrappers support</p> 2248<pre>import libxml2 2249 2250doc = libxml2.parseFile("tst.xml") 2251ctxt = doc.xpathNewContext() 2252res = ctxt.xpathEval("//*") 2253if len(res) != 2: 2254 print "xpath query: wrong node set size" 2255 sys.exit(1) 2256if res[0].name != "doc" or res[1].name != "foo": 2257 print "xpath query: wrong node set value" 2258 sys.exit(1) 2259doc.freeDoc() 2260ctxt.xpathFreeContext()</pre> 2261 2262<p>This test parses a file, then create an XPath context to evaluate XPath 2263expression on it. The xpathEval() method execute an XPath query and returns 2264the result mapped in a Python way. String and numbers are natively converted, 2265and node sets are returned as a tuple of libxml2 Python nodes wrappers. Like 2266the document, the XPath context need to be freed explicitly, also not that 2267the result of the XPath query may point back to the document tree and hence 2268the document must be freed after the result of the query is used.</p> 2269 2270<h3>xpathext.py:</h3> 2271 2272<p>This test shows how to extend the XPath engine with functions written in 2273python:</p> 2274<pre>import libxml2 2275 2276def foo(ctx, x): 2277 return x + 1 2278 2279doc = libxml2.parseFile("tst.xml") 2280ctxt = doc.xpathNewContext() 2281libxml2.registerXPathFunction(ctxt._o, "foo", None, foo) 2282res = ctxt.xpathEval("foo(1)") 2283if res != 2: 2284 print "xpath extension failure" 2285doc.freeDoc() 2286ctxt.xpathFreeContext()</pre> 2287 2288<p>Note how the extension function is registered with the context (but that 2289part is not yet finalized, this may change slightly in the future).</p> 2290 2291<h3>tstxpath.py:</h3> 2292 2293<p>This test is similar to the previous one but shows how the extension 2294function can access the XPath evaluation context:</p> 2295<pre>def foo(ctx, x): 2296 global called 2297 2298 # 2299 # test that access to the XPath evaluation contexts 2300 # 2301 pctxt = libxml2.xpathParserContext(_obj=ctx) 2302 ctxt = pctxt.context() 2303 called = ctxt.function() 2304 return x + 1</pre> 2305 2306<p>All the interfaces around the XPath parser(or rather evaluation) context 2307are not finalized, but it should be sufficient to do contextual work at the 2308evaluation point.</p> 2309 2310<h3>Memory debugging:</h3> 2311 2312<p>last but not least, all tests starts with the following prologue:</p> 2313<pre>#memory debug specific 2314libxml2.debugMemory(1)</pre> 2315 2316<p>and ends with the following epilogue:</p> 2317<pre>#memory debug specific 2318libxml2.cleanupParser() 2319if libxml2.debugMemory(1) == 0: 2320 print "OK" 2321else: 2322 print "Memory leak %d bytes" % (libxml2.debugMemory(1)) 2323 libxml2.dumpMemory()</pre> 2324 2325<p>Those activate the memory debugging interface of libxml2 where all 2326allocated block in the library are tracked. The prologue then cleans up the 2327library state and checks that all allocated memory has been freed. If not it 2328calls dumpMemory() which saves that list in a <code>.memdump</code> file.</p> 2329 2330<h2><a name="architecture">libxml2 architecture</a></h2> 2331 2332<p>Libxml2 is made of multiple components; some of them are optional, and 2333most of the block interfaces are public. The main components are:</p> 2334<ul> 2335 <li>an Input/Output layer</li> 2336 <li>FTP and HTTP client layers (optional)</li> 2337 <li>an Internationalization layer managing the encodings support</li> 2338 <li>a URI module</li> 2339 <li>the XML parser and its basic SAX interface</li> 2340 <li>an HTML parser using the same SAX interface (optional)</li> 2341 <li>a SAX tree module to build an in-memory DOM representation</li> 2342 <li>a tree module to manipulate the DOM representation</li> 2343 <li>a validation module using the DOM representation (optional)</li> 2344 <li>an XPath module for global lookup in a DOM representation 2345 (optional)</li> 2346 <li>a debug module (optional)</li> 2347</ul> 2348 2349<p>Graphically this gives the following:</p> 2350 2351<p><img src="libxml.gif" alt="a graphical view of the various"></p> 2352 2353<p></p> 2354 2355<h2><a name="tree">The tree output</a></h2> 2356 2357<p>The parser returns a tree built during the document analysis. The value 2358returned is an <strong>xmlDocPtr</strong> (i.e., a pointer to an 2359<strong>xmlDoc</strong> structure). This structure contains information such 2360as the file name, the document type, and a <strong>children</strong> pointer 2361which is the root of the document (or more exactly the first child under the 2362root which is the document). The tree is made of <strong>xmlNode</strong>s, 2363chained in double-linked lists of siblings and with a children<->parent 2364relationship. An xmlNode can also carry properties (a chain of xmlAttr 2365structures). An attribute may have a value which is a list of TEXT or 2366ENTITY_REF nodes.</p> 2367 2368<p>Here is an example (erroneous with respect to the XML spec since there 2369should be only one ELEMENT under the root):</p> 2370 2371<p><img src="structure.gif" alt=" structure.gif "></p> 2372 2373<p>In the source package there is a small program (not installed by default) 2374called <strong>xmllint</strong> which parses XML files given as argument and 2375prints them back as parsed. This is useful for detecting errors both in XML 2376code and in the XML parser itself. It has an option <strong>--debug</strong> 2377which prints the actual in-memory structure of the document; here is the 2378result with the <a href="#example">example</a> given before:</p> 2379<pre>DOCUMENT 2380version=1.0 2381standalone=true 2382 ELEMENT EXAMPLE 2383 ATTRIBUTE prop1 2384 TEXT 2385 content=gnome is great 2386 ATTRIBUTE prop2 2387 ENTITY_REF 2388 TEXT 2389 content= linux too 2390 ELEMENT head 2391 ELEMENT title 2392 TEXT 2393 content=Welcome to Gnome 2394 ELEMENT chapter 2395 ELEMENT title 2396 TEXT 2397 content=The Linux adventure 2398 ELEMENT p 2399 TEXT 2400 content=bla bla bla ... 2401 ELEMENT image 2402 ATTRIBUTE href 2403 TEXT 2404 content=linus.gif 2405 ELEMENT p 2406 TEXT 2407 content=...</pre> 2408 2409<p>This should be useful for learning the internal representation model.</p> 2410 2411<h2><a name="interface">The SAX interface</a></h2> 2412 2413<p>Sometimes the DOM tree output is just too large to fit reasonably into 2414memory. In that case (and if you don't expect to save back the XML document 2415loaded using libxml), it's better to use the SAX interface of libxml. SAX is 2416a <strong>callback-based interface</strong> to the parser. Before parsing, 2417the application layer registers a customized set of callbacks which are 2418called by the library as it progresses through the XML input.</p> 2419 2420<p>To get more detailed step-by-step guidance on using the SAX interface of 2421libxml, see the <a 2422href="http://www.daa.com.au/~james/gnome/xml-sax/xml-sax.html">nice 2423documentation</a>.written by <a href="mailto:james@daa.com.au">James 2424Henstridge</a>.</p> 2425 2426<p>You can debug the SAX behaviour by using the <strong>testSAX</strong> 2427program located in the gnome-xml module (it's usually not shipped in the 2428binary packages of libxml, but you can find it in the tar source 2429distribution). Here is the sequence of callbacks that would be reported by 2430testSAX when parsing the example XML document shown earlier:</p> 2431<pre>SAX.setDocumentLocator() 2432SAX.startDocument() 2433SAX.getEntity(amp) 2434SAX.startElement(EXAMPLE, prop1='gnome is great', prop2='&amp; linux too') 2435SAX.characters( , 3) 2436SAX.startElement(head) 2437SAX.characters( , 4) 2438SAX.startElement(title) 2439SAX.characters(Welcome to Gnome, 16) 2440SAX.endElement(title) 2441SAX.characters( , 3) 2442SAX.endElement(head) 2443SAX.characters( , 3) 2444SAX.startElement(chapter) 2445SAX.characters( , 4) 2446SAX.startElement(title) 2447SAX.characters(The Linux adventure, 19) 2448SAX.endElement(title) 2449SAX.characters( , 4) 2450SAX.startElement(p) 2451SAX.characters(bla bla bla ..., 15) 2452SAX.endElement(p) 2453SAX.characters( , 4) 2454SAX.startElement(image, href='linus.gif') 2455SAX.endElement(image) 2456SAX.characters( , 4) 2457SAX.startElement(p) 2458SAX.characters(..., 3) 2459SAX.endElement(p) 2460SAX.characters( , 3) 2461SAX.endElement(chapter) 2462SAX.characters( , 1) 2463SAX.endElement(EXAMPLE) 2464SAX.endDocument()</pre> 2465 2466<p>Most of the other interfaces of libxml2 are based on the DOM tree-building 2467facility, so nearly everything up to the end of this document presupposes the 2468use of the standard DOM tree build. Note that the DOM tree itself is built by 2469a set of registered default callbacks, without internal specific 2470interface.</p> 2471 2472<h2><a name="Validation">Validation & DTDs</a></h2> 2473 2474<p>Table of Content:</p> 2475<ol> 2476 <li><a href="#General5">General overview</a></li> 2477 <li><a href="#definition">The definition</a></li> 2478 <li><a href="#Simple">Simple rules</a> 2479 <ol> 2480 <li><a href="#reference">How to reference a DTD from a document</a></li> 2481 <li><a href="#Declaring">Declaring elements</a></li> 2482 <li><a href="#Declaring1">Declaring attributes</a></li> 2483 </ol> 2484 </li> 2485 <li><a href="#Some">Some examples</a></li> 2486 <li><a href="#validate">How to validate</a></li> 2487 <li><a href="#Other">Other resources</a></li> 2488</ol> 2489 2490<h3><a name="General5">General overview</a></h3> 2491 2492<p>Well what is validation and what is a DTD ?</p> 2493 2494<p>DTD is the acronym for Document Type Definition. This is a description of 2495the content for a family of XML files. This is part of the XML 1.0 2496specification, and allows one to describe and verify that a given document 2497instance conforms to the set of rules detailing its structure and content.</p> 2498 2499<p>Validation is the process of checking a document against a DTD (more 2500generally against a set of construction rules).</p> 2501 2502<p>The validation process and building DTDs are the two most difficult parts 2503of the XML life cycle. Briefly a DTD defines all the possible elements to be 2504found within your document, what is the formal shape of your document tree 2505(by defining the allowed content of an element; either text, a regular 2506expression for the allowed list of children, or mixed content i.e. both text 2507and children). The DTD also defines the valid attributes for all elements and 2508the types of those attributes.</p> 2509 2510<h3><a name="definition1">The definition</a></h3> 2511 2512<p>The <a href="http://www.w3.org/TR/REC-xml">W3C XML Recommendation</a> (<a 2513href="http://www.xml.com/axml/axml.html">Tim Bray's annotated version of 2514Rev1</a>):</p> 2515<ul> 2516 <li><a href="http://www.w3.org/TR/REC-xml#elemdecls">Declaring 2517 elements</a></li> 2518 <li><a href="http://www.w3.org/TR/REC-xml#attdecls">Declaring 2519 attributes</a></li> 2520</ul> 2521 2522<p>(unfortunately) all this is inherited from the SGML world, the syntax is 2523ancient...</p> 2524 2525<h3><a name="Simple1">Simple rules</a></h3> 2526 2527<p>Writing DTDs can be done in many ways. The rules to build them if you need 2528something permanent or something which can evolve over time can be radically 2529different. Really complex DTDs like DocBook ones are flexible but quite 2530harder to design. I will just focus on DTDs for a formats with a fixed simple 2531structure. It is just a set of basic rules, and definitely not exhaustive nor 2532usable for complex DTD design.</p> 2533 2534<h4><a name="reference1">How to reference a DTD from a document</a>:</h4> 2535 2536<p>Assuming the top element of the document is <code>spec</code> and the dtd 2537is placed in the file <code>mydtd</code> in the subdirectory 2538<code>dtds</code> of the directory from where the document were loaded:</p> 2539 2540<p><code><!DOCTYPE spec SYSTEM "dtds/mydtd"></code></p> 2541 2542<p>Notes:</p> 2543<ul> 2544 <li>The system string is actually an URI-Reference (as defined in <a 2545 href="http://www.ietf.org/rfc/rfc2396.txt">RFC 2396</a>) so you can use a 2546 full URL string indicating the location of your DTD on the Web. This is a 2547 really good thing to do if you want others to validate your document.</li> 2548 <li>It is also possible to associate a <code>PUBLIC</code> identifier (a 2549 magic string) so that the DTD is looked up in catalogs on the client side 2550 without having to locate it on the web.</li> 2551 <li>A DTD contains a set of element and attribute declarations, but they 2552 don't define what the root of the document should be. This is explicitly 2553 told to the parser/validator as the first element of the 2554 <code>DOCTYPE</code> declaration.</li> 2555</ul> 2556 2557<h4><a name="Declaring2">Declaring elements</a>:</h4> 2558 2559<p>The following declares an element <code>spec</code>:</p> 2560 2561<p><code><!ELEMENT spec (front, body, back?)></code></p> 2562 2563<p>It also expresses that the spec element contains one <code>front</code>, 2564one <code>body</code> and one optional <code>back</code> children elements in 2565this order. The declaration of one element of the structure and its content 2566are done in a single declaration. Similarly the following declares 2567<code>div1</code> elements:</p> 2568 2569<p><code><!ELEMENT div1 (head, (p | list | note)*, div2?)></code></p> 2570 2571<p>which means div1 contains one <code>head</code> then a series of optional 2572<code>p</code>, <code>list</code>s and <code>note</code>s and then an 2573optional <code>div2</code>. And last but not least an element can contain 2574text:</p> 2575 2576<p><code><!ELEMENT b (#PCDATA)></code></p> 2577 2578<p><code>b</code> contains text or being of mixed content (text and elements 2579in no particular order):</p> 2580 2581<p><code><!ELEMENT p (#PCDATA|a|ul|b|i|em)*></code></p> 2582 2583<p><code>p </code>can contain text or <code>a</code>, <code>ul</code>, 2584<code>b</code>, <code>i </code>or <code>em</code> elements in no particular 2585order.</p> 2586 2587<h4><a name="Declaring1">Declaring attributes</a>:</h4> 2588 2589<p>Again the attributes declaration includes their content definition:</p> 2590 2591<p><code><!ATTLIST termdef name CDATA #IMPLIED></code></p> 2592 2593<p>means that the element <code>termdef</code> can have a <code>name</code> 2594attribute containing text (<code>CDATA</code>) and which is optional 2595(<code>#IMPLIED</code>). The attribute value can also be defined within a 2596set:</p> 2597 2598<p><code><!ATTLIST list type (bullets|ordered|glossary) 2599"ordered"></code></p> 2600 2601<p>means <code>list</code> element have a <code>type</code> attribute with 3 2602allowed values "bullets", "ordered" or "glossary" and which default to 2603"ordered" if the attribute is not explicitly specified.</p> 2604 2605<p>The content type of an attribute can be text (<code>CDATA</code>), 2606anchor/reference/references 2607(<code>ID</code>/<code>IDREF</code>/<code>IDREFS</code>), entity(ies) 2608(<code>ENTITY</code>/<code>ENTITIES</code>) or name(s) 2609(<code>NMTOKEN</code>/<code>NMTOKENS</code>). The following defines that a 2610<code>chapter</code> element can have an optional <code>id</code> attribute 2611of type <code>ID</code>, usable for reference from attribute of type 2612IDREF:</p> 2613 2614<p><code><!ATTLIST chapter id ID #IMPLIED></code></p> 2615 2616<p>The last value of an attribute definition can be <code>#REQUIRED 2617</code>meaning that the attribute has to be given, <code>#IMPLIED</code> 2618meaning that it is optional, or the default value (possibly prefixed by 2619<code>#FIXED</code> if it is the only allowed).</p> 2620 2621<p>Notes:</p> 2622<ul> 2623 <li>Usually the attributes pertaining to a given element are declared in a 2624 single expression, but it is just a convention adopted by a lot of DTD 2625 writers: 2626 <pre><!ATTLIST termdef 2627 id ID #REQUIRED 2628 name CDATA #IMPLIED></pre> 2629 <p>The previous construct defines both <code>id</code> and 2630 <code>name</code> attributes for the element <code>termdef</code>.</p> 2631 </li> 2632</ul> 2633 2634<h3><a name="Some1">Some examples</a></h3> 2635 2636<p>The directory <code>test/valid/dtds/</code> in the libxml2 distribution 2637contains some complex DTD examples. The example in the file 2638<code>test/valid/dia.xml</code> shows an XML file where the simple DTD is 2639directly included within the document.</p> 2640 2641<h3><a name="validate1">How to validate</a></h3> 2642 2643<p>The simplest way is to use the xmllint program included with libxml. The 2644<code>--valid</code> option turns-on validation of the files given as input. 2645For example the following validates a copy of the first revision of the XML 26461.0 specification:</p> 2647 2648<p><code>xmllint --valid --noout test/valid/REC-xml-19980210.xml</code></p> 2649 2650<p>the -- noout is used to disable output of the resulting tree.</p> 2651 2652<p>The <code>--dtdvalid dtd</code> allows validation of the document(s) 2653against a given DTD.</p> 2654 2655<p>Libxml2 exports an API to handle DTDs and validation, check the <a 2656href="http://xmlsoft.org/html/libxml-valid.html">associated 2657description</a>.</p> 2658 2659<h3><a name="Other1">Other resources</a></h3> 2660 2661<p>DTDs are as old as SGML. So there may be a number of examples on-line, I 2662will just list one for now, others pointers welcome:</p> 2663<ul> 2664 <li><a href="http://www.xml101.com:8081/dtd/">XML-101 DTD</a></li> 2665</ul> 2666 2667<p>I suggest looking at the examples found under test/valid/dtd and any of 2668the large number of books available on XML. The dia example in test/valid 2669should be both simple and complete enough to allow you to build your own.</p> 2670 2671<p></p> 2672 2673<h2><a name="Memory">Memory Management</a></h2> 2674 2675<p>Table of Content:</p> 2676<ol> 2677 <li><a href="#General3">General overview</a></li> 2678 <li><a href="#setting">Setting libxml2 set of memory routines</a></li> 2679 <li><a href="#cleanup">Cleaning up after parsing</a></li> 2680 <li><a href="#Debugging">Debugging routines</a></li> 2681 <li><a href="#General4">General memory requirements</a></li> 2682</ol> 2683 2684<h3><a name="General3">General overview</a></h3> 2685 2686<p>The module <code><a 2687href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlmemory.h</a></code> 2688provides the interfaces to the libxml2 memory system:</p> 2689<ul> 2690 <li>libxml2 does not use the libc memory allocator directly but xmlFree(), 2691 xmlMalloc() and xmlRealloc()</li> 2692 <li>those routines can be reallocated to a specific set of routine, by 2693 default the libc ones i.e. free(), malloc() and realloc()</li> 2694 <li>the xmlmemory.c module includes a set of debugging routine</li> 2695</ul> 2696 2697<h3><a name="setting">Setting libxml2 set of memory routines</a></h3> 2698 2699<p>It is sometimes useful to not use the default memory allocator, either for 2700debugging, analysis or to implement a specific behaviour on memory management 2701(like on embedded systems). Two function calls are available to do so:</p> 2702<ul> 2703 <li><a href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemGet 2704 ()</a> which return the current set of functions in use by the parser</li> 2705 <li><a 2706 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemSetup()</a> 2707 which allow to set up a new set of memory allocation functions</li> 2708</ul> 2709 2710<p>Of course a call to xmlMemSetup() should probably be done before calling 2711any other libxml2 routines (unless you are sure your allocations routines are 2712compatibles).</p> 2713 2714<h3><a name="cleanup">Cleaning up after parsing</a></h3> 2715 2716<p>Libxml2 is not stateless, there is a few set of memory structures needing 2717allocation before the parser is fully functional (some encoding structures 2718for example). This also mean that once parsing is finished there is a tiny 2719amount of memory (a few hundred bytes) which can be recollected if you don't 2720reuse the parser immediately:</p> 2721<ul> 2722 <li><a href="http://xmlsoft.org/html/libxml-parser.html">xmlCleanupParser 2723 ()</a> is a centralized routine to free the parsing states. Note that it 2724 won't deallocate any produced tree if any (use the xmlFreeDoc() and 2725 related routines for this).</li> 2726 <li><a href="http://xmlsoft.org/html/libxml-parser.html">xmlInitParser 2727 ()</a> is the dual routine allowing to preallocate the parsing state 2728 which can be useful for example to avoid initialization reentrancy 2729 problems when using libxml2 in multithreaded applications</li> 2730</ul> 2731 2732<p>Generally xmlCleanupParser() is safe, if needed the state will be rebuild 2733at the next invocation of parser routines, but be careful of the consequences 2734in multithreaded applications.</p> 2735 2736<h3><a name="Debugging">Debugging routines</a></h3> 2737 2738<p>When configured using --with-mem-debug flag (off by default), libxml2 uses 2739a set of memory allocation debugging routines keeping track of all allocated 2740blocks and the location in the code where the routine was called. A couple of 2741other debugging routines allow to dump the memory allocated infos to a file 2742or call a specific routine when a given block number is allocated:</p> 2743<ul> 2744 <li><a 2745 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMallocLoc()</a> 2746 <a 2747 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlReallocLoc()</a> 2748 and <a 2749 href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemStrdupLoc()</a> 2750 are the memory debugging replacement allocation routines</li> 2751 <li><a href="http://xmlsoft.org/html/libxml-xmlmemory.html">xmlMemoryDump 2752 ()</a> dumps all the informations about the allocated memory block lefts 2753 in the <code>.memdump</code> file</li> 2754</ul> 2755 2756<p>When developing libxml2 memory debug is enabled, the tests programs call 2757xmlMemoryDump () and the "make test" regression tests will check for any 2758memory leak during the full regression test sequence, this helps a lot 2759ensuring that libxml2 does not leak memory and bullet proof memory 2760allocations use (some libc implementations are known to be far too permissive 2761resulting in major portability problems!).</p> 2762 2763<p>If the .memdump reports a leak, it displays the allocation function and 2764also tries to give some informations about the content and structure of the 2765allocated blocks left. This is sufficient in most cases to find the culprit, 2766but not always. Assuming the allocation problem is reproducible, it is 2767possible to find more easily:</p> 2768<ol> 2769 <li>write down the block number xxxx not allocated</li> 2770 <li>export the environment variable XML_MEM_BREAKPOINT=xxxx , the easiest 2771 when using GDB is to simply give the command 2772 <p><code>set environment XML_MEM_BREAKPOINT xxxx</code></p> 2773 <p>before running the program.</p> 2774 </li> 2775 <li>run the program under a debugger and set a breakpoint on 2776 xmlMallocBreakpoint() a specific function called when this precise block 2777 is allocated</li> 2778 <li>when the breakpoint is reached you can then do a fine analysis of the 2779 allocation an step to see the condition resulting in the missing 2780 deallocation.</li> 2781</ol> 2782 2783<p>I used to use a commercial tool to debug libxml2 memory problems but after 2784noticing that it was not detecting memory leaks that simple mechanism was 2785used and proved extremely efficient until now. Lately I have also used <a 2786href="http://developer.kde.org/~sewardj/">valgrind</a> with quite some 2787success, it is tied to the i386 architecture since it works by emulating the 2788processor and instruction set, it is slow but extremely efficient, i.e. it 2789spot memory usage errors in a very precise way.</p> 2790 2791<h3><a name="General4">General memory requirements</a></h3> 2792 2793<p>How much libxml2 memory require ? It's hard to tell in average it depends 2794of a number of things:</p> 2795<ul> 2796 <li>the parser itself should work in a fixed amount of memory, except for 2797 information maintained about the stacks of names and entities locations. 2798 The I/O and encoding handlers will probably account for a few KBytes. 2799 This is true for both the XML and HTML parser (though the HTML parser 2800 need more state).</li> 2801 <li>If you are generating the DOM tree then memory requirements will grow 2802 nearly linear with the size of the data. In general for a balanced 2803 textual document the internal memory requirement is about 4 times the 2804 size of the UTF8 serialization of this document (example the XML-1.0 2805 recommendation is a bit more of 150KBytes and takes 650KBytes of main 2806 memory when parsed). Validation will add a amount of memory required for 2807 maintaining the external Dtd state which should be linear with the 2808 complexity of the content model defined by the Dtd</li> 2809 <li>If you need to work with fixed memory requirements or don't need the 2810 full DOM tree then using the <a href="xmlreader.html">xmlReader 2811 interface</a> is probably the best way to proceed, it still allows to 2812 validate or operate on subset of the tree if needed.</li> 2813 <li>If you don't care about the advanced features of libxml2 like 2814 validation, DOM, XPath or XPointer, don't use entities, need to work with 2815 fixed memory requirements, and try to get the fastest parsing possible 2816 then the SAX interface should be used, but it has known restrictions.</li> 2817</ul> 2818 2819<p></p> 2820 2821<h2><a name="Encodings">Encodings support</a></h2> 2822 2823<p>Table of Content:</p> 2824<ol> 2825 <li><a href="encoding.html#What">What does internationalization support 2826 mean ?</a></li> 2827 <li><a href="encoding.html#internal">The internal encoding, how and 2828 why</a></li> 2829 <li><a href="encoding.html#implemente">How is it implemented ?</a></li> 2830 <li><a href="encoding.html#Default">Default supported encodings</a></li> 2831 <li><a href="encoding.html#extend">How to extend the existing 2832 support</a></li> 2833</ol> 2834 2835<h3><a name="What">What does internationalization support mean ?</a></h3> 2836 2837<p>If you are not really familiar with Internationalization (usual shortcut 2838is I18N) , Unicode, characters and glyphs, I suggest you read a <a 2839href="http://www.tbray.org/ongoing/When/200x/2003/04/06/Unicode">presentation</a> 2840by Tim Bray on Unicode and why you should care about it.</p> 2841 2842<p>XML was designed from the start to allow the support of any character set 2843by using Unicode. Any conformant XML parser has to support the UTF-8 and 2844UTF-16 default encodings which can both express the full unicode ranges. UTF8 2845is a variable length encoding whose greatest points are to reuse the same 2846encoding for ASCII and to save space for Western encodings, but it is a bit 2847more complex to handle in practice. UTF-16 use 2 bytes per character (and 2848sometimes combines two pairs), it makes implementation easier, but looks a 2849bit overkill for Western languages encoding. Moreover the XML specification 2850allows the document to be encoded in other encodings at the condition that 2851they are clearly labeled as such. For example the following is a wellformed 2852XML document encoded in ISO-8859-1 and using accentuated letters that we 2853French like for both markup and content:</p> 2854<pre><?xml version="1.0" encoding="ISO-8859-1"?> 2855<tr�s>l�</tr�s></pre> 2856 2857<p>Having internationalization support in libxml2 means the following:</p> 2858<ul> 2859 <li>the document is properly parsed</li> 2860 <li>informations about it's encoding are saved</li> 2861 <li>it can be modified</li> 2862 <li>it can be saved in its original encoding</li> 2863 <li>it can also be saved in another encoding supported by libxml2 (for 2864 example straight UTF8 or even an ASCII form)</li> 2865</ul> 2866 2867<p>Another very important point is that the whole libxml2 API, with the 2868exception of a few routines to read with a specific encoding or save to a 2869specific encoding, is completely agnostic about the original encoding of the 2870document.</p> 2871 2872<p>It should be noted too that the HTML parser embedded in libxml2 now obey 2873the same rules too, the following document will be (as of 2.2.2) handled in 2874an internationalized fashion by libxml2 too:</p> 2875<pre><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" 2876 "http://www.w3.org/TR/REC-html40/loose.dtd"> 2877<html lang="fr"> 2878<head> 2879 <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=ISO-8859-1"> 2880</head> 2881<body> 2882<p>W3C cr�e des standards pour le Web.</body> 2883</html></pre> 2884 2885<h3><a name="internal">The internal encoding, how and why</a></h3> 2886 2887<p>One of the core decisions was to force all documents to be converted to a 2888default internal encoding, and that encoding to be UTF-8, here are the 2889rationales for those choices:</p> 2890<ul> 2891 <li>keeping the native encoding in the internal form would force the libxml 2892 users (or the code associated) to be fully aware of the encoding of the 2893 original document, for examples when adding a text node to a document, 2894 the content would have to be provided in the document encoding, i.e. the 2895 client code would have to check it before hand, make sure it's conformant 2896 to the encoding, etc ... Very hard in practice, though in some specific 2897 cases this may make sense.</li> 2898 <li>the second decision was which encoding. From the XML spec only UTF8 and 2899 UTF16 really makes sense as being the two only encodings for which there 2900 is mandatory support. UCS-4 (32 bits fixed size encoding) could be 2901 considered an intelligent choice too since it's a direct Unicode mapping 2902 support. I selected UTF-8 on the basis of efficiency and compatibility 2903 with surrounding software: 2904 <ul> 2905 <li>UTF-8 while a bit more complex to convert from/to (i.e. slightly 2906 more costly to import and export CPU wise) is also far more compact 2907 than UTF-16 (and UCS-4) for a majority of the documents I see it used 2908 for right now (RPM RDF catalogs, advogato data, various configuration 2909 file formats, etc.) and the key point for today's computer 2910 architecture is efficient uses of caches. If one nearly double the 2911 memory requirement to store the same amount of data, this will trash 2912 caches (main memory/external caches/internal caches) and my take is 2913 that this harms the system far more than the CPU requirements needed 2914 for the conversion to UTF-8</li> 2915 <li>Most of libxml2 version 1 users were using it with straight ASCII 2916 most of the time, doing the conversion with an internal encoding 2917 requiring all their code to be rewritten was a serious show-stopper 2918 for using UTF-16 or UCS-4.</li> 2919 <li>UTF-8 is being used as the de-facto internal encoding standard for 2920 related code like the <a href="http://www.pango.org/">pango</a> 2921 upcoming Gnome text widget, and a lot of Unix code (yet another place 2922 where Unix programmer base takes a different approach from Microsoft 2923 - they are using UTF-16)</li> 2924 </ul> 2925 </li> 2926</ul> 2927 2928<p>What does this mean in practice for the libxml2 user:</p> 2929<ul> 2930 <li>xmlChar, the libxml2 data type is a byte, those bytes must be assembled 2931 as UTF-8 valid strings. The proper way to terminate an xmlChar * string 2932 is simply to append 0 byte, as usual.</li> 2933 <li>One just need to make sure that when using chars outside the ASCII set, 2934 the values has been properly converted to UTF-8</li> 2935</ul> 2936 2937<h3><a name="implemente">How is it implemented ?</a></h3> 2938 2939<p>Let's describe how all this works within libxml, basically the I18N 2940(internationalization) support get triggered only during I/O operation, i.e. 2941when reading a document or saving one. Let's look first at the reading 2942sequence:</p> 2943<ol> 2944 <li>when a document is processed, we usually don't know the encoding, a 2945 simple heuristic allows to detect UTF-16 and UCS-4 from encodings where 2946 the ASCII range (0-0x7F) maps with ASCII</li> 2947 <li>the xml declaration if available is parsed, including the encoding 2948 declaration. At that point, if the autodetected encoding is different 2949 from the one declared a call to xmlSwitchEncoding() is issued.</li> 2950 <li>If there is no encoding declaration, then the input has to be in either 2951 UTF-8 or UTF-16, if it is not then at some point when processing the 2952 input, the converter/checker of UTF-8 form will raise an encoding error. 2953 You may end-up with a garbled document, or no document at all ! Example: 2954 <pre>~/XML -> /xmllint err.xml 2955err.xml:1: error: Input is not proper UTF-8, indicate encoding ! 2956<tr�s>l�</tr�s> 2957 ^ 2958err.xml:1: error: Bytes: 0xE8 0x73 0x3E 0x6C 2959<tr�s>l�</tr�s> 2960 ^</pre> 2961 </li> 2962 <li>xmlSwitchEncoding() does an encoding name lookup, canonicalize it, and 2963 then search the default registered encoding converters for that encoding. 2964 If it's not within the default set and iconv() support has been compiled 2965 it, it will ask iconv for such an encoder. If this fails then the parser 2966 will report an error and stops processing: 2967 <pre>~/XML -> /xmllint err2.xml 2968err2.xml:1: error: Unsupported encoding UnsupportedEnc 2969<?xml version="1.0" encoding="UnsupportedEnc"?> 2970 ^</pre> 2971 </li> 2972 <li>From that point the encoder processes progressively the input (it is 2973 plugged as a front-end to the I/O module) for that entity. It captures 2974 and converts on-the-fly the document to be parsed to UTF-8. The parser 2975 itself just does UTF-8 checking of this input and process it 2976 transparently. The only difference is that the encoding information has 2977 been added to the parsing context (more precisely to the input 2978 corresponding to this entity).</li> 2979 <li>The result (when using DOM) is an internal form completely in UTF-8 2980 with just an encoding information on the document node.</li> 2981</ol> 2982 2983<p>Ok then what happens when saving the document (assuming you 2984collected/built an xmlDoc DOM like structure) ? It depends on the function 2985called, xmlSaveFile() will just try to save in the original encoding, while 2986xmlSaveFileTo() and xmlSaveFileEnc() can optionally save to a given 2987encoding:</p> 2988<ol> 2989 <li>if no encoding is given, libxml2 will look for an encoding value 2990 associated to the document and if it exists will try to save to that 2991 encoding, 2992 <p>otherwise everything is written in the internal form, i.e. UTF-8</p> 2993 </li> 2994 <li>so if an encoding was specified, either at the API level or on the 2995 document, libxml2 will again canonicalize the encoding name, lookup for a 2996 converter in the registered set or through iconv. If not found the 2997 function will return an error code</li> 2998 <li>the converter is placed before the I/O buffer layer, as another kind of 2999 buffer, then libxml2 will simply push the UTF-8 serialization to through 3000 that buffer, which will then progressively be converted and pushed onto 3001 the I/O layer.</li> 3002 <li>It is possible that the converter code fails on some input, for example 3003 trying to push an UTF-8 encoded Chinese character through the UTF-8 to 3004 ISO-8859-1 converter won't work. Since the encoders are progressive they 3005 will just report the error and the number of bytes converted, at that 3006 point libxml2 will decode the offending character, remove it from the 3007 buffer and replace it with the associated charRef encoding &#123; and 3008 resume the conversion. This guarantees that any document will be saved 3009 without losses (except for markup names where this is not legal, this is 3010 a problem in the current version, in practice avoid using non-ascii 3011 characters for tag or attribute names). A special "ascii" encoding name 3012 is used to save documents to a pure ascii form can be used when 3013 portability is really crucial</li> 3014</ol> 3015 3016<p>Here are a few examples based on the same test document:</p> 3017<pre>~/XML -> /xmllint isolat1 3018<?xml version="1.0" encoding="ISO-8859-1"?> 3019<tr�s>l�</tr�s> 3020~/XML -> /xmllint --encode UTF-8 isolat1 3021<?xml version="1.0" encoding="UTF-8"?> 3022<très>l� �</très> 3023~/XML -> </pre> 3024 3025<p>The same processing is applied (and reuse most of the code) for HTML I18N 3026processing. Looking up and modifying the content encoding is a bit more 3027difficult since it is located in a <meta> tag under the <head>, 3028so a couple of functions htmlGetMetaEncoding() and htmlSetMetaEncoding() have 3029been provided. The parser also attempts to switch encoding on the fly when 3030detecting such a tag on input. Except for that the processing is the same 3031(and again reuses the same code).</p> 3032 3033<h3><a name="Default">Default supported encodings</a></h3> 3034 3035<p>libxml2 has a set of default converters for the following encodings 3036(located in encoding.c):</p> 3037<ol> 3038 <li>UTF-8 is supported by default (null handlers)</li> 3039 <li>UTF-16, both little and big endian</li> 3040 <li>ISO-Latin-1 (ISO-8859-1) covering most western languages</li> 3041 <li>ASCII, useful mostly for saving</li> 3042 <li>HTML, a specific handler for the conversion of UTF-8 to ASCII with HTML 3043 predefined entities like &copy; for the Copyright sign.</li> 3044</ol> 3045 3046<p>More over when compiled on an Unix platform with iconv support the full 3047set of encodings supported by iconv can be instantly be used by libxml. On a 3048linux machine with glibc-2.1 the list of supported encodings and aliases fill 30493 full pages, and include UCS-4, the full set of ISO-Latin encodings, and the 3050various Japanese ones.</p> 3051 3052<h4>Encoding aliases</h4> 3053 3054<p>From 2.2.3, libxml2 has support to register encoding names aliases. The 3055goal is to be able to parse document whose encoding is supported but where 3056the name differs (for example from the default set of names accepted by 3057iconv). The following functions allow to register and handle new aliases for 3058existing encodings. Once registered libxml2 will automatically lookup the 3059aliases when handling a document:</p> 3060<ul> 3061 <li>int xmlAddEncodingAlias(const char *name, const char *alias);</li> 3062 <li>int xmlDelEncodingAlias(const char *alias);</li> 3063 <li>const char * xmlGetEncodingAlias(const char *alias);</li> 3064 <li>void xmlCleanupEncodingAliases(void);</li> 3065</ul> 3066 3067<h3><a name="extend">How to extend the existing support</a></h3> 3068 3069<p>Well adding support for new encoding, or overriding one of the encoders 3070(assuming it is buggy) should not be hard, just write input and output 3071conversion routines to/from UTF-8, and register them using 3072xmlNewCharEncodingHandler(name, xxxToUTF8, UTF8Toxxx), and they will be 3073called automatically if the parser(s) encounter such an encoding name 3074(register it uppercase, this will help). The description of the encoders, 3075their arguments and expected return values are described in the encoding.h 3076header.</p> 3077 3078<p>A quick note on the topic of subverting the parser to use a different 3079internal encoding than UTF-8, in some case people will absolutely want to 3080keep the internal encoding different, I think it's still possible (but the 3081encoding must be compliant with ASCII on the same subrange) though I didn't 3082tried it. The key is to override the default conversion routines (by 3083registering null encoders/decoders for your charsets), and bypass the UTF-8 3084checking of the parser by setting the parser context charset 3085(ctxt->charset) to something different than XML_CHAR_ENCODING_UTF8, but 3086there is no guarantee that this will work. You may also have some troubles 3087saving back.</p> 3088 3089<p>Basically proper I18N support is important, this requires at least 3090libxml-2.0.0, but a lot of features and corrections are really available only 3091starting 2.2.</p> 3092 3093<h2><a name="IO">I/O Interfaces</a></h2> 3094 3095<p>Table of Content:</p> 3096<ol> 3097 <li><a href="#General1">General overview</a></li> 3098 <li><a href="#basic">The basic buffer type</a></li> 3099 <li><a href="#Input">Input I/O handlers</a></li> 3100 <li><a href="#Output">Output I/O handlers</a></li> 3101 <li><a href="#entities">The entities loader</a></li> 3102 <li><a href="#Example2">Example of customized I/O</a></li> 3103</ol> 3104 3105<h3><a name="General1">General overview</a></h3> 3106 3107<p>The module <code><a 3108href="http://xmlsoft.org/html/libxml-xmlio.html">xmlIO.h</a></code> provides 3109the interfaces to the libxml2 I/O system. This consists of 4 main parts:</p> 3110<ul> 3111 <li>Entities loader, this is a routine which tries to fetch the entities 3112 (files) based on their PUBLIC and SYSTEM identifiers. The default loader 3113 don't look at the public identifier since libxml2 do not maintain a 3114 catalog. You can redefine you own entity loader by using 3115 <code>xmlGetExternalEntityLoader()</code> and 3116 <code>xmlSetExternalEntityLoader()</code>. <a href="#entities">Check the 3117 example</a>.</li> 3118 <li>Input I/O buffers which are a commodity structure used by the parser(s) 3119 input layer to handle fetching the informations to feed the parser. This 3120 provides buffering and is also a placeholder where the encoding 3121 converters to UTF8 are piggy-backed.</li> 3122 <li>Output I/O buffers are similar to the Input ones and fulfill similar 3123 task but when generating a serialization from a tree.</li> 3124 <li>A mechanism to register sets of I/O callbacks and associate them with 3125 specific naming schemes like the protocol part of the URIs. 3126 <p>This affect the default I/O operations and allows to use specific I/O 3127 handlers for certain names.</p> 3128 </li> 3129</ul> 3130 3131<p>The general mechanism used when loading http://rpmfind.net/xml.html for 3132example in the HTML parser is the following:</p> 3133<ol> 3134 <li>The default entity loader calls <code>xmlNewInputFromFile()</code> with 3135 the parsing context and the URI string.</li> 3136 <li>the URI string is checked against the existing registered handlers 3137 using their match() callback function, if the HTTP module was compiled 3138 in, it is registered and its match() function will succeeds</li> 3139 <li>the open() function of the handler is called and if successful will 3140 return an I/O Input buffer</li> 3141 <li>the parser will the start reading from this buffer and progressively 3142 fetch information from the resource, calling the read() function of the 3143 handler until the resource is exhausted</li> 3144 <li>if an encoding change is detected it will be installed on the input 3145 buffer, providing buffering and efficient use of the conversion 3146 routines</li> 3147 <li>once the parser has finished, the close() function of the handler is 3148 called once and the Input buffer and associated resources are 3149 deallocated.</li> 3150</ol> 3151 3152<p>The user defined callbacks are checked first to allow overriding of the 3153default libxml2 I/O routines.</p> 3154 3155<h3><a name="basic">The basic buffer type</a></h3> 3156 3157<p>All the buffer manipulation handling is done using the 3158<code>xmlBuffer</code> type define in <code><a 3159href="http://xmlsoft.org/html/libxml-tree.html">tree.h</a> </code>which is a 3160resizable memory buffer. The buffer allocation strategy can be selected to be 3161either best-fit or use an exponential doubling one (CPU vs. memory use 3162trade-off). The values are <code>XML_BUFFER_ALLOC_EXACT</code> and 3163<code>XML_BUFFER_ALLOC_DOUBLEIT</code>, and can be set individually or on a 3164system wide basis using <code>xmlBufferSetAllocationScheme()</code>. A number 3165of functions allows to manipulate buffers with names starting with the 3166<code>xmlBuffer...</code> prefix.</p> 3167 3168<h3><a name="Input">Input I/O handlers</a></h3> 3169 3170<p>An Input I/O handler is a simple structure 3171<code>xmlParserInputBuffer</code> containing a context associated to the 3172resource (file descriptor, or pointer to a protocol handler), the read() and 3173close() callbacks to use and an xmlBuffer. And extra xmlBuffer and a charset 3174encoding handler are also present to support charset conversion when 3175needed.</p> 3176 3177<h3><a name="Output">Output I/O handlers</a></h3> 3178 3179<p>An Output handler <code>xmlOutputBuffer</code> is completely similar to an 3180Input one except the callbacks are write() and close().</p> 3181 3182<h3><a name="entities">The entities loader</a></h3> 3183 3184<p>The entity loader resolves requests for new entities and create inputs for 3185the parser. Creating an input from a filename or an URI string is done 3186through the xmlNewInputFromFile() routine. The default entity loader do not 3187handle the PUBLIC identifier associated with an entity (if any). So it just 3188calls xmlNewInputFromFile() with the SYSTEM identifier (which is mandatory in 3189XML).</p> 3190 3191<p>If you want to hook up a catalog mechanism then you simply need to 3192override the default entity loader, here is an example:</p> 3193<pre>#include <libxml/xmlIO.h> 3194 3195xmlExternalEntityLoader defaultLoader = NULL; 3196 3197xmlParserInputPtr 3198xmlMyExternalEntityLoader(const char *URL, const char *ID, 3199 xmlParserCtxtPtr ctxt) { 3200 xmlParserInputPtr ret; 3201 const char *fileID = NULL; 3202 /* lookup for the fileID depending on ID */ 3203 3204 ret = xmlNewInputFromFile(ctxt, fileID); 3205 if (ret != NULL) 3206 return(ret); 3207 if (defaultLoader != NULL) 3208 ret = defaultLoader(URL, ID, ctxt); 3209 return(ret); 3210} 3211 3212int main(..) { 3213 ... 3214 3215 /* 3216 * Install our own entity loader 3217 */ 3218 defaultLoader = xmlGetExternalEntityLoader(); 3219 xmlSetExternalEntityLoader(xmlMyExternalEntityLoader); 3220 3221 ... 3222}</pre> 3223 3224<h3><a name="Example2">Example of customized I/O</a></h3> 3225 3226<p>This example come from <a href="http://xmlsoft.org/messages/0708.html">a 3227real use case</a>, xmlDocDump() closes the FILE * passed by the application 3228and this was a problem. The <a 3229href="http://xmlsoft.org/messages/0711.html">solution</a> was to redefine a 3230new output handler with the closing call deactivated:</p> 3231<ol> 3232 <li>First define a new I/O output allocator where the output don't close 3233 the file: 3234 <pre>xmlOutputBufferPtr 3235xmlOutputBufferCreateOwn(FILE *file, xmlCharEncodingHandlerPtr encoder) { 3236����xmlOutputBufferPtr ret; 3237���� 3238����if (xmlOutputCallbackInitialized == 0) 3239��������xmlRegisterDefaultOutputCallbacks(); 3240 3241����if (file == NULL) return(NULL); 3242����ret = xmlAllocOutputBuffer(encoder); 3243����if (ret != NULL) { 3244��������ret->context = file; 3245��������ret->writecallback = xmlFileWrite; 3246��������ret->closecallback = NULL; /* No close callback */ 3247����} 3248����return(ret); 3249} </pre> 3250 </li> 3251 <li>And then use it to save the document: 3252 <pre>FILE *f; 3253xmlOutputBufferPtr output; 3254xmlDocPtr doc; 3255int res; 3256 3257f = ... 3258doc = .... 3259 3260output = xmlOutputBufferCreateOwn(f, NULL); 3261res = xmlSaveFileTo(output, doc, NULL); 3262 </pre> 3263 </li> 3264</ol> 3265 3266<h2><a name="Catalog">Catalog support</a></h2> 3267 3268<p>Table of Content:</p> 3269<ol> 3270 <li><a href="General2">General overview</a></li> 3271 <li><a href="#definition">The definition</a></li> 3272 <li><a href="#Simple">Using catalogs</a></li> 3273 <li><a href="#Some">Some examples</a></li> 3274 <li><a href="#reference">How to tune catalog usage</a></li> 3275 <li><a href="#validate">How to debug catalog processing</a></li> 3276 <li><a href="#Declaring">How to create and maintain catalogs</a></li> 3277 <li><a href="#implemento">The implementor corner quick review of the 3278 API</a></li> 3279 <li><a href="#Other">Other resources</a></li> 3280</ol> 3281 3282<h3><a name="General2">General overview</a></h3> 3283 3284<p>What is a catalog? Basically it's a lookup mechanism used when an entity 3285(a file or a remote resource) references another entity. The catalog lookup 3286is inserted between the moment the reference is recognized by the software 3287(XML parser, stylesheet processing, or even images referenced for inclusion 3288in a rendering) and the time where loading that resource is actually 3289started.</p> 3290 3291<p>It is basically used for 3 things:</p> 3292<ul> 3293 <li>mapping from "logical" names, the public identifiers and a more 3294 concrete name usable for download (and URI). For example it can associate 3295 the logical name 3296 <p>"-//OASIS//DTD DocBook XML V4.1.2//EN"</p> 3297 <p>of the DocBook 4.1.2 XML DTD with the actual URL where it can be 3298 downloaded</p> 3299 <p>http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd</p> 3300 </li> 3301 <li>remapping from a given URL to another one, like an HTTP indirection 3302 saying that 3303 <p>"http://www.oasis-open.org/committes/tr.xsl"</p> 3304 <p>should really be looked at</p> 3305 <p>"http://www.oasis-open.org/committes/entity/stylesheets/base/tr.xsl"</p> 3306 </li> 3307 <li>providing a local cache mechanism allowing to load the entities 3308 associated to public identifiers or remote resources, this is a really 3309 important feature for any significant deployment of XML or SGML since it 3310 allows to avoid the aleas and delays associated to fetching remote 3311 resources.</li> 3312</ul> 3313 3314<h3><a name="definition">The definitions</a></h3> 3315 3316<p>Libxml, as of 2.4.3 implements 2 kind of catalogs:</p> 3317<ul> 3318 <li>the older SGML catalogs, the official spec is SGML Open Technical 3319 Resolution TR9401:1997, but is better understood by reading <a 3320 href="http://www.jclark.com/sp/catalog.htm">the SP Catalog page</a> from 3321 James Clark. This is relatively old and not the preferred mode of 3322 operation of libxml.</li> 3323 <li><a href="http://www.oasis-open.org/committees/entity/spec.html">XML 3324 Catalogs</a> is far more flexible, more recent, uses an XML syntax and 3325 should scale quite better. This is the default option of libxml.</li> 3326</ul> 3327 3328<p></p> 3329 3330<h3><a name="Simple">Using catalog</a></h3> 3331 3332<p>In a normal environment libxml2 will by default check the presence of a 3333catalog in /etc/xml/catalog, and assuming it has been correctly populated, 3334the processing is completely transparent to the document user. To take a 3335concrete example, suppose you are authoring a DocBook document, this one 3336starts with the following DOCTYPE definition:</p> 3337<pre><?xml version='1.0'?> 3338<!DOCTYPE book PUBLIC "-//Norman Walsh//DTD DocBk XML V3.1.4//EN" 3339 "http://nwalsh.com/docbook/xml/3.1.4/db3xml.dtd"></pre> 3340 3341<p>When validating the document with libxml, the catalog will be 3342automatically consulted to lookup the public identifier "-//Norman Walsh//DTD 3343DocBk XML V3.1.4//EN" and the system identifier 3344"http://nwalsh.com/docbook/xml/3.1.4/db3xml.dtd", and if these entities have 3345been installed on your system and the catalogs actually point to them, libxml 3346will fetch them from the local disk.</p> 3347 3348<p style="font-size: 10pt"><strong>Note</strong>: Really don't use this 3349DOCTYPE example it's a really old version, but is fine as an example.</p> 3350 3351<p>Libxml2 will check the catalog each time that it is requested to load an 3352entity, this includes DTD, external parsed entities, stylesheets, etc ... If 3353your system is correctly configured all the authoring phase and processing 3354should use only local files, even if your document stays portable because it 3355uses the canonical public and system ID, referencing the remote document.</p> 3356 3357<h3><a name="Some">Some examples:</a></h3> 3358 3359<p>Here is a couple of fragments from XML Catalogs used in libxml2 early 3360regression tests in <code>test/catalogs</code> :</p> 3361<pre><?xml version="1.0"?> 3362<!DOCTYPE catalog PUBLIC 3363 "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" 3364 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 3365<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"> 3366 <public publicId="-//OASIS//DTD DocBook XML V4.1.2//EN" 3367 uri="http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"/> 3368...</pre> 3369 3370<p>This is the beginning of a catalog for DocBook 4.1.2, XML Catalogs are 3371written in XML, there is a specific namespace for catalog elements 3372"urn:oasis:names:tc:entity:xmlns:xml:catalog". The first entry in this 3373catalog is a <code>public</code> mapping it allows to associate a Public 3374Identifier with an URI.</p> 3375<pre>... 3376 <rewriteSystem systemIdStartString="http://www.oasis-open.org/docbook/" 3377 rewritePrefix="file:///usr/share/xml/docbook/"/> 3378...</pre> 3379 3380<p>A <code>rewriteSystem</code> is a very powerful instruction, it says that 3381any URI starting with a given prefix should be looked at another URI 3382constructed by replacing the prefix with an new one. In effect this acts like 3383a cache system for a full area of the Web. In practice it is extremely useful 3384with a file prefix if you have installed a copy of those resources on your 3385local system.</p> 3386<pre>... 3387<delegatePublic publicIdStartString="-//OASIS//DTD XML Catalog //" 3388 catalog="file:///usr/share/xml/docbook.xml"/> 3389<delegatePublic publicIdStartString="-//OASIS//ENTITIES DocBook XML" 3390 catalog="file:///usr/share/xml/docbook.xml"/> 3391<delegatePublic publicIdStartString="-//OASIS//DTD DocBook XML" 3392 catalog="file:///usr/share/xml/docbook.xml"/> 3393<delegateSystem systemIdStartString="http://www.oasis-open.org/docbook/" 3394 catalog="file:///usr/share/xml/docbook.xml"/> 3395<delegateURI uriStartString="http://www.oasis-open.org/docbook/" 3396 catalog="file:///usr/share/xml/docbook.xml"/> 3397...</pre> 3398 3399<p>Delegation is the core features which allows to build a tree of catalogs, 3400easier to maintain than a single catalog, based on Public Identifier, System 3401Identifier or URI prefixes it instructs the catalog software to look up 3402entries in another resource. This feature allow to build hierarchies of 3403catalogs, the set of entries presented should be sufficient to redirect the 3404resolution of all DocBook references to the specific catalog in 3405<code>/usr/share/xml/docbook.xml</code> this one in turn could delegate all 3406references for DocBook 4.2.1 to a specific catalog installed at the same time 3407as the DocBook resources on the local machine.</p> 3408 3409<h3><a name="reference">How to tune catalog usage:</a></h3> 3410 3411<p>The user can change the default catalog behaviour by redirecting queries 3412to its own set of catalogs, this can be done by setting the 3413<code>XML_CATALOG_FILES</code> environment variable to a list of catalogs, an 3414empty one should deactivate loading the default <code>/etc/xml/catalog</code> 3415default catalog</p> 3416 3417<h3><a name="validate">How to debug catalog processing:</a></h3> 3418 3419<p>Setting up the <code>XML_DEBUG_CATALOG</code> environment variable will 3420make libxml2 output debugging informations for each catalog operations, for 3421example:</p> 3422<pre>orchis:~/XML -> xmllint --memory --noout test/ent2 3423warning: failed to load external entity "title.xml" 3424orchis:~/XML -> export XML_DEBUG_CATALOG= 3425orchis:~/XML -> xmllint --memory --noout test/ent2 3426Failed to parse catalog /etc/xml/catalog 3427Failed to parse catalog /etc/xml/catalog 3428warning: failed to load external entity "title.xml" 3429Catalogs cleanup 3430orchis:~/XML -> </pre> 3431 3432<p>The test/ent2 references an entity, running the parser from memory makes 3433the base URI unavailable and the the "title.xml" entity cannot be loaded. 3434Setting up the debug environment variable allows to detect that an attempt is 3435made to load the <code>/etc/xml/catalog</code> but since it's not present the 3436resolution fails.</p> 3437 3438<p>But the most advanced way to debug XML catalog processing is to use the 3439<strong>xmlcatalog</strong> command shipped with libxml2, it allows to load 3440catalogs and make resolution queries to see what is going on. This is also 3441used for the regression tests:</p> 3442<pre>orchis:~/XML -> /xmlcatalog test/catalogs/docbook.xml \ 3443 "-//OASIS//DTD DocBook XML V4.1.2//EN" 3444http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd 3445orchis:~/XML -> </pre> 3446 3447<p>For debugging what is going on, adding one -v flags increase the verbosity 3448level to indicate the processing done (adding a second flag also indicate 3449what elements are recognized at parsing):</p> 3450<pre>orchis:~/XML -> /xmlcatalog -v test/catalogs/docbook.xml \ 3451 "-//OASIS//DTD DocBook XML V4.1.2//EN" 3452Parsing catalog test/catalogs/docbook.xml's content 3453Found public match -//OASIS//DTD DocBook XML V4.1.2//EN 3454http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd 3455Catalogs cleanup 3456orchis:~/XML -> </pre> 3457 3458<p>A shell interface is also available to debug and process multiple queries 3459(and for regression tests):</p> 3460<pre>orchis:~/XML -> /xmlcatalog -shell test/catalogs/docbook.xml \ 3461 "-//OASIS//DTD DocBook XML V4.1.2//EN" 3462> help 3463Commands available: 3464public PublicID: make a PUBLIC identifier lookup 3465system SystemID: make a SYSTEM identifier lookup 3466resolve PublicID SystemID: do a full resolver lookup 3467add 'type' 'orig' 'replace' : add an entry 3468del 'values' : remove values 3469dump: print the current catalog state 3470debug: increase the verbosity level 3471quiet: decrease the verbosity level 3472exit: quit the shell 3473> public "-//OASIS//DTD DocBook XML V4.1.2//EN" 3474http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd 3475> quit 3476orchis:~/XML -> </pre> 3477 3478<p>This should be sufficient for most debugging purpose, this was actually 3479used heavily to debug the XML Catalog implementation itself.</p> 3480 3481<h3><a name="Declaring">How to create and maintain</a> catalogs:</h3> 3482 3483<p>Basically XML Catalogs are XML files, you can either use XML tools to 3484manage them or use <strong>xmlcatalog</strong> for this. The basic step is 3485to create a catalog the -create option provide this facility:</p> 3486<pre>orchis:~/XML -> /xmlcatalog --create tst.xml 3487<?xml version="1.0"?> 3488<!DOCTYPE catalog PUBLIC "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" 3489 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 3490<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"/> 3491orchis:~/XML -> </pre> 3492 3493<p>By default xmlcatalog does not overwrite the original catalog and save the 3494result on the standard output, this can be overridden using the -noout 3495option. The <code>-add</code> command allows to add entries in the 3496catalog:</p> 3497<pre>orchis:~/XML -> /xmlcatalog --noout --create --add "public" \ 3498 "-//OASIS//DTD DocBook XML V4.1.2//EN" \ 3499 http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd tst.xml 3500orchis:~/XML -> cat tst.xml 3501<?xml version="1.0"?> 3502<!DOCTYPE catalog PUBLIC "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" \ 3503 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 3504<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"> 3505<public publicId="-//OASIS//DTD DocBook XML V4.1.2//EN" 3506 uri="http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd"/> 3507</catalog> 3508orchis:~/XML -> </pre> 3509 3510<p>The <code>-add</code> option will always take 3 parameters even if some of 3511the XML Catalog constructs (like nextCatalog) will have only a single 3512argument, just pass a third empty string, it will be ignored.</p> 3513 3514<p>Similarly the <code>-del</code> option remove matching entries from the 3515catalog:</p> 3516<pre>orchis:~/XML -> /xmlcatalog --del \ 3517 "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" tst.xml 3518<?xml version="1.0"?> 3519<!DOCTYPE catalog PUBLIC "-//OASIS//DTD Entity Resolution XML Catalog V1.0//EN" 3520 "http://www.oasis-open.org/committees/entity/release/1.0/catalog.dtd"> 3521<catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog"/> 3522orchis:~/XML -> </pre> 3523 3524<p>The catalog is now empty. Note that the matching of <code>-del</code> is 3525exact and would have worked in a similar fashion with the Public ID 3526string.</p> 3527 3528<p>This is rudimentary but should be sufficient to manage a not too complex 3529catalog tree of resources.</p> 3530 3531<h3><a name="implemento">The implementor corner quick review of the 3532API:</a></h3> 3533 3534<p>First, and like for every other module of libxml, there is an 3535automatically generated <a href="html/libxml-catalog.html">API page for 3536catalog support</a>.</p> 3537 3538<p>The header for the catalog interfaces should be included as:</p> 3539<pre>#include <libxml/catalog.h></pre> 3540 3541<p>The API is voluntarily kept very simple. First it is not obvious that 3542applications really need access to it since it is the default behaviour of 3543libxml2 (Note: it is possible to completely override libxml2 default catalog 3544by using <a href="html/libxml-parser.html">xmlSetExternalEntityLoader</a> to 3545plug an application specific resolver).</p> 3546 3547<p>Basically libxml2 support 2 catalog lists:</p> 3548<ul> 3549 <li>the default one, global shared by all the application</li> 3550 <li>a per-document catalog, this one is built if the document uses the 3551 <code>oasis-xml-catalog</code> PIs to specify its own catalog list, it is 3552 associated to the parser context and destroyed when the parsing context 3553 is destroyed.</li> 3554</ul> 3555 3556<p>the document one will be used first if it exists.</p> 3557 3558<h4>Initialization routines:</h4> 3559 3560<p>xmlInitializeCatalog(), xmlLoadCatalog() and xmlLoadCatalogs() should be 3561used at startup to initialize the catalog, if the catalog should be 3562initialized with specific values xmlLoadCatalog() or xmlLoadCatalogs() 3563should be called before xmlInitializeCatalog() which would otherwise do a 3564default initialization first.</p> 3565 3566<p>The xmlCatalogAddLocal() call is used by the parser to grow the document 3567own catalog list if needed.</p> 3568 3569<h4>Preferences setup:</h4> 3570 3571<p>The XML Catalog spec requires the possibility to select default 3572preferences between public and system delegation, 3573xmlCatalogSetDefaultPrefer() allows this, xmlCatalogSetDefaults() and 3574xmlCatalogGetDefaults() allow to control if XML Catalogs resolution should 3575be forbidden, allowed for global catalog, for document catalog or both, the 3576default is to allow both.</p> 3577 3578<p>And of course xmlCatalogSetDebug() allows to generate debug messages 3579(through the xmlGenericError() mechanism).</p> 3580 3581<h4>Querying routines:</h4> 3582 3583<p>xmlCatalogResolve(), xmlCatalogResolveSystem(), xmlCatalogResolvePublic() 3584and xmlCatalogResolveURI() are relatively explicit if you read the XML 3585Catalog specification they correspond to section 7 algorithms, they should 3586also work if you have loaded an SGML catalog with a simplified semantic.</p> 3587 3588<p>xmlCatalogLocalResolve() and xmlCatalogLocalResolveURI() are the same but 3589operate on the document catalog list</p> 3590 3591<h4>Cleanup and Miscellaneous:</h4> 3592 3593<p>xmlCatalogCleanup() free-up the global catalog, xmlCatalogFreeLocal() is 3594the per-document equivalent.</p> 3595 3596<p>xmlCatalogAdd() and xmlCatalogRemove() are used to dynamically modify the 3597first catalog in the global list, and xmlCatalogDump() allows to dump a 3598catalog state, those routines are primarily designed for xmlcatalog, I'm not 3599sure that exposing more complex interfaces (like navigation ones) would be 3600really useful.</p> 3601 3602<p>The xmlParseCatalogFile() is a function used to load XML Catalog files, 3603it's similar as xmlParseFile() except it bypass all catalog lookups, it's 3604provided because this functionality may be useful for client tools.</p> 3605 3606<h4>threaded environments:</h4> 3607 3608<p>Since the catalog tree is built progressively, some care has been taken to 3609try to avoid troubles in multithreaded environments. The code is now thread 3610safe assuming that the libxml2 library has been compiled with threads 3611support.</p> 3612 3613<p></p> 3614 3615<h3><a name="Other">Other resources</a></h3> 3616 3617<p>The XML Catalog specification is relatively recent so there isn't much 3618literature to point at:</p> 3619<ul> 3620 <li>You can find a good rant from Norm Walsh about <a 3621 href="http://www.arbortext.com/Think_Tank/XML_Resources/Issue_Three/issue_three.html">the 3622 need for catalogs</a>, it provides a lot of context informations even if 3623 I don't agree with everything presented. Norm also wrote a more recent 3624 article <a 3625 href="http://wwws.sun.com/software/xml/developers/resolver/article/">XML 3626 entities and URI resolvers</a> describing them.</li> 3627 <li>An <a href="http://home.ccil.org/~cowan/XML/XCatalog.html">old XML 3628 catalog proposal</a> from John Cowan</li> 3629 <li>The <a href="http://www.rddl.org/">Resource Directory Description 3630 Language</a> (RDDL) another catalog system but more oriented toward 3631 providing metadata for XML namespaces.</li> 3632 <li>the page from the OASIS Technical <a 3633 href="http://www.oasis-open.org/committees/entity/">Committee on Entity 3634 Resolution</a> who maintains XML Catalog, you will find pointers to the 3635 specification update, some background and pointers to others tools 3636 providing XML Catalog support</li> 3637 <li>There is a <a href="buildDocBookCatalog">shell script</a> to generate 3638 XML Catalogs for DocBook 4.1.2 . If it can write to the /etc/xml/ 3639 directory, it will set-up /etc/xml/catalog and /etc/xml/docbook based on 3640 the resources found on the system. Otherwise it will just create 3641 ~/xmlcatalog and ~/dbkxmlcatalog and doing: 3642 <p><code>export XML_CATALOG_FILES=$HOME/xmlcatalog</code></p> 3643 <p>should allow to process DocBook documentations without requiring 3644 network accesses for the DTD or stylesheets</p> 3645 </li> 3646 <li>I have uploaded <a href="ftp://xmlsoft.org/test/dbk412catalog.tar.gz">a 3647 small tarball</a> containing XML Catalogs for DocBook 4.1.2 which seems 3648 to work fine for me too</li> 3649 <li>The <a href="http://www.xmlsoft.org/xmlcatalog_man.html">xmlcatalog 3650 manual page</a></li> 3651</ul> 3652 3653<p>If you have suggestions for corrections or additions, simply contact 3654me:</p> 3655 3656<h2><a name="library">The parser interfaces</a></h2> 3657 3658<p>This section is directly intended to help programmers getting bootstrapped 3659using the XML tollkit from the C language. It is not intended to be 3660extensive. I hope the automatically generated documents will provide the 3661completeness required, but as a separate set of documents. The interfaces of 3662the XML parser are by principle low level, Those interested in a higher level 3663API should <a href="#DOM">look at DOM</a>.</p> 3664 3665<p>The <a href="html/libxml-parser.html">parser interfaces for XML</a> are 3666separated from the <a href="html/libxml-htmlparser.html">HTML parser 3667interfaces</a>. Let's have a look at how the XML parser can be called:</p> 3668 3669<h3><a name="Invoking">Invoking the parser : the pull method</a></h3> 3670 3671<p>Usually, the first thing to do is to read an XML input. The parser accepts 3672documents either from in-memory strings or from files. The functions are 3673defined in "parser.h":</p> 3674<dl> 3675 <dt><code>xmlDocPtr xmlParseMemory(char *buffer, int size);</code></dt> 3676 <dd><p>Parse a null-terminated string containing the document.</p> 3677 </dd> 3678</dl> 3679<dl> 3680 <dt><code>xmlDocPtr xmlParseFile(const char *filename);</code></dt> 3681 <dd><p>Parse an XML document contained in a (possibly compressed) 3682 file.</p> 3683 </dd> 3684</dl> 3685 3686<p>The parser returns a pointer to the document structure (or NULL in case of 3687failure).</p> 3688 3689<h3 id="Invoking1">Invoking the parser: the push method</h3> 3690 3691<p>In order for the application to keep the control when the document is 3692being fetched (which is common for GUI based programs) libxml2 provides a 3693push interface, too, as of version 1.8.3. Here are the interface 3694functions:</p> 3695<pre>xmlParserCtxtPtr xmlCreatePushParserCtxt(xmlSAXHandlerPtr sax, 3696 void *user_data, 3697 const char *chunk, 3698 int size, 3699 const char *filename); 3700int xmlParseChunk (xmlParserCtxtPtr ctxt, 3701 const char *chunk, 3702 int size, 3703 int terminate);</pre> 3704 3705<p>and here is a simple example showing how to use the interface:</p> 3706<pre> FILE *f; 3707 3708 f = fopen(filename, "r"); 3709 if (f != NULL) { 3710 int res, size = 1024; 3711 char chars[1024]; 3712 xmlParserCtxtPtr ctxt; 3713 3714 res = fread(chars, 1, 4, f); 3715 if (res > 0) { 3716 ctxt = xmlCreatePushParserCtxt(NULL, NULL, 3717 chars, res, filename); 3718 while ((res = fread(chars, 1, size, f)) > 0) { 3719 xmlParseChunk(ctxt, chars, res, 0); 3720 } 3721 xmlParseChunk(ctxt, chars, 0, 1); 3722 doc = ctxt->myDoc; 3723 xmlFreeParserCtxt(ctxt); 3724 } 3725 }</pre> 3726 3727<p>The HTML parser embedded into libxml2 also has a push interface; the 3728functions are just prefixed by "html" rather than "xml".</p> 3729 3730<h3 id="Invoking2">Invoking the parser: the SAX interface</h3> 3731 3732<p>The tree-building interface makes the parser memory-hungry, first loading 3733the document in memory and then building the tree itself. Reading a document 3734without building the tree is possible using the SAX interfaces (see SAX.h and 3735<a href="http://www.daa.com.au/~james/gnome/xml-sax/xml-sax.html">James 3736Henstridge's documentation</a>). Note also that the push interface can be 3737limited to SAX: just use the two first arguments of 3738<code>xmlCreatePushParserCtxt()</code>.</p> 3739 3740<h3><a name="Building">Building a tree from scratch</a></h3> 3741 3742<p>The other way to get an XML tree in memory is by building it. Basically 3743there is a set of functions dedicated to building new elements. (These are 3744also described in <libxml/tree.h>.) For example, here is a piece of 3745code that produces the XML document used in the previous examples:</p> 3746<pre> #include <libxml/tree.h> 3747 xmlDocPtr doc; 3748 xmlNodePtr tree, subtree; 3749 3750 doc = xmlNewDoc("1.0"); 3751 doc->children = xmlNewDocNode(doc, NULL, "EXAMPLE", NULL); 3752 xmlSetProp(doc->children, "prop1", "gnome is great"); 3753 xmlSetProp(doc->children, "prop2", "& linux too"); 3754 tree = xmlNewChild(doc->children, NULL, "head", NULL); 3755 subtree = xmlNewChild(tree, NULL, "title", "Welcome to Gnome"); 3756 tree = xmlNewChild(doc->children, NULL, "chapter", NULL); 3757 subtree = xmlNewChild(tree, NULL, "title", "The Linux adventure"); 3758 subtree = xmlNewChild(tree, NULL, "p", "bla bla bla ..."); 3759 subtree = xmlNewChild(tree, NULL, "image", NULL); 3760 xmlSetProp(subtree, "href", "linus.gif");</pre> 3761 3762<p>Not really rocket science ...</p> 3763 3764<h3><a name="Traversing">Traversing the tree</a></h3> 3765 3766<p>Basically by <a href="html/libxml-tree.html">including "tree.h"</a> your 3767code has access to the internal structure of all the elements of the tree. 3768The names should be somewhat simple like <strong>parent</strong>, 3769<strong>children</strong>, <strong>next</strong>, <strong>prev</strong>, 3770<strong>properties</strong>, etc... For example, still with the previous 3771example:</p> 3772<pre><code>doc->children->children->children</code></pre> 3773 3774<p>points to the title element,</p> 3775<pre>doc->children->children->next->children->children</pre> 3776 3777<p>points to the text node containing the chapter title "The Linux 3778adventure".</p> 3779 3780<p><strong>NOTE</strong>: XML allows <em>PI</em>s and <em>comments</em> to be 3781present before the document root, so <code>doc->children</code> may point 3782to an element which is not the document Root Element; a function 3783<code>xmlDocGetRootElement()</code> was added for this purpose.</p> 3784 3785<h3><a name="Modifying">Modifying the tree</a></h3> 3786 3787<p>Functions are provided for reading and writing the document content. Here 3788is an excerpt from the <a href="html/libxml-tree.html">tree API</a>:</p> 3789<dl> 3790 <dt><code>xmlAttrPtr xmlSetProp(xmlNodePtr node, const xmlChar *name, const 3791 xmlChar *value);</code></dt> 3792 <dd><p>This sets (or changes) an attribute carried by an ELEMENT node. 3793 The value can be NULL.</p> 3794 </dd> 3795</dl> 3796<dl> 3797 <dt><code>const xmlChar *xmlGetProp(xmlNodePtr node, const xmlChar 3798 *name);</code></dt> 3799 <dd><p>This function returns a pointer to new copy of the property 3800 content. Note that the user must deallocate the result.</p> 3801 </dd> 3802</dl> 3803 3804<p>Two functions are provided for reading and writing the text associated 3805with elements:</p> 3806<dl> 3807 <dt><code>xmlNodePtr xmlStringGetNodeList(xmlDocPtr doc, const xmlChar 3808 *value);</code></dt> 3809 <dd><p>This function takes an "external" string and converts it to one 3810 text node or possibly to a list of entity and text nodes. All 3811 non-predefined entity references like &Gnome; will be stored 3812 internally as entity nodes, hence the result of the function may not be 3813 a single node.</p> 3814 </dd> 3815</dl> 3816<dl> 3817 <dt><code>xmlChar *xmlNodeListGetString(xmlDocPtr doc, xmlNodePtr list, int 3818 inLine);</code></dt> 3819 <dd><p>This function is the inverse of 3820 <code>xmlStringGetNodeList()</code>. It generates a new string 3821 containing the content of the text and entity nodes. Note the extra 3822 argument inLine. If this argument is set to 1, the function will expand 3823 entity references. For example, instead of returning the &Gnome; 3824 XML encoding in the string, it will substitute it with its value (say, 3825 "GNU Network Object Model Environment").</p> 3826 </dd> 3827</dl> 3828 3829<h3><a name="Saving">Saving a tree</a></h3> 3830 3831<p>Basically 3 options are possible:</p> 3832<dl> 3833 <dt><code>void xmlDocDumpMemory(xmlDocPtr cur, xmlChar**mem, int 3834 *size);</code></dt> 3835 <dd><p>Returns a buffer into which the document has been saved.</p> 3836 </dd> 3837</dl> 3838<dl> 3839 <dt><code>extern void xmlDocDump(FILE *f, xmlDocPtr doc);</code></dt> 3840 <dd><p>Dumps a document to an open file descriptor.</p> 3841 </dd> 3842</dl> 3843<dl> 3844 <dt><code>int xmlSaveFile(const char *filename, xmlDocPtr cur);</code></dt> 3845 <dd><p>Saves the document to a file. In this case, the compression 3846 interface is triggered if it has been turned on.</p> 3847 </dd> 3848</dl> 3849 3850<h3><a name="Compressio">Compression</a></h3> 3851 3852<p>The library transparently handles compression when doing file-based 3853accesses. The level of compression on saves can be turned on either globally 3854or individually for one file:</p> 3855<dl> 3856 <dt><code>int xmlGetDocCompressMode (xmlDocPtr doc);</code></dt> 3857 <dd><p>Gets the document compression ratio (0-9).</p> 3858 </dd> 3859</dl> 3860<dl> 3861 <dt><code>void xmlSetDocCompressMode (xmlDocPtr doc, int mode);</code></dt> 3862 <dd><p>Sets the document compression ratio.</p> 3863 </dd> 3864</dl> 3865<dl> 3866 <dt><code>int xmlGetCompressMode(void);</code></dt> 3867 <dd><p>Gets the default compression ratio.</p> 3868 </dd> 3869</dl> 3870<dl> 3871 <dt><code>void xmlSetCompressMode(int mode);</code></dt> 3872 <dd><p>Sets the default compression ratio.</p> 3873 </dd> 3874</dl> 3875 3876<h2><a name="Entities">Entities or no entities</a></h2> 3877 3878<p>Entities in principle are similar to simple C macros. An entity defines an 3879abbreviation for a given string that you can reuse many times throughout the 3880content of your document. Entities are especially useful when a given string 3881may occur frequently within a document, or to confine the change needed to a 3882document to a restricted area in the internal subset of the document (at the 3883beginning). Example:</p> 3884<pre>1 <?xml version="1.0"?> 38852 <!DOCTYPE EXAMPLE SYSTEM "example.dtd" [ 38863 <!ENTITY xml "Extensible Markup Language"> 38874 ]> 38885 <EXAMPLE> 38896 &xml; 38907 </EXAMPLE></pre> 3891 3892<p>Line 3 declares the xml entity. Line 6 uses the xml entity, by prefixing 3893its name with '&' and following it by ';' without any spaces added. There 3894are 5 predefined entities in libxml2 allowing you to escape characters with 3895predefined meaning in some parts of the xml document content: 3896<strong>&lt;</strong> for the character '<', <strong>&gt;</strong> 3897for the character '>', <strong>&apos;</strong> for the character ''', 3898<strong>&quot;</strong> for the character '"', and 3899<strong>&amp;</strong> for the character '&'.</p> 3900 3901<p>One of the problems related to entities is that you may want the parser to 3902substitute an entity's content so that you can see the replacement text in 3903your application. Or you may prefer to keep entity references as such in the 3904content to be able to save the document back without losing this usually 3905precious information (if the user went through the pain of explicitly 3906defining entities, he may have a a rather negative attitude if you blindly 3907substitute them as saving time). The <a 3908href="html/libxml-parser.html#xmlSubstituteEntitiesDefault">xmlSubstituteEntitiesDefault()</a> 3909function allows you to check and change the behaviour, which is to not 3910substitute entities by default.</p> 3911 3912<p>Here is the DOM tree built by libxml2 for the previous document in the 3913default case:</p> 3914<pre>/gnome/src/gnome-xml -> /xmllint --debug test/ent1 3915DOCUMENT 3916version=1.0 3917 ELEMENT EXAMPLE 3918 TEXT 3919 content= 3920 ENTITY_REF 3921 INTERNAL_GENERAL_ENTITY xml 3922 content=Extensible Markup Language 3923 TEXT 3924 content=</pre> 3925 3926<p>And here is the result when substituting entities:</p> 3927<pre>/gnome/src/gnome-xml -> /tester --debug --noent test/ent1 3928DOCUMENT 3929version=1.0 3930 ELEMENT EXAMPLE 3931 TEXT 3932 content= Extensible Markup Language</pre> 3933 3934<p>So, entities or no entities? Basically, it depends on your use case. I 3935suggest that you keep the non-substituting default behaviour and avoid using 3936entities in your XML document or data if you are not willing to handle the 3937entity references elements in the DOM tree.</p> 3938 3939<p>Note that at save time libxml2 enforces the conversion of the predefined 3940entities where necessary to prevent well-formedness problems, and will also 3941transparently replace those with chars (i.e. it will not generate entity 3942reference elements in the DOM tree or call the reference() SAX callback when 3943finding them in the input).</p> 3944 3945<p><span style="background-color: #FF0000">WARNING</span>: handling entities 3946on top of the libxml2 SAX interface is difficult!!! If you plan to use 3947non-predefined entities in your documents, then the learning curve to handle 3948then using the SAX API may be long. If you plan to use complex documents, I 3949strongly suggest you consider using the DOM interface instead and let libxml 3950deal with the complexity rather than trying to do it yourself.</p> 3951 3952<h2><a name="Namespaces">Namespaces</a></h2> 3953 3954<p>The libxml2 library implements <a 3955href="http://www.w3.org/TR/REC-xml-names/">XML namespaces</a> support by 3956recognizing namespace constructs in the input, and does namespace lookup 3957automatically when building the DOM tree. A namespace declaration is 3958associated with an in-memory structure and all elements or attributes within 3959that namespace point to it. Hence testing the namespace is a simple and fast 3960equality operation at the user level.</p> 3961 3962<p>I suggest that people using libxml2 use a namespace, and declare it in the 3963root element of their document as the default namespace. Then they don't need 3964to use the prefix in the content but we will have a basis for future semantic 3965refinement and merging of data from different sources. This doesn't increase 3966the size of the XML output significantly, but significantly increases its 3967value in the long-term. Example:</p> 3968<pre><mydoc xmlns="http://mydoc.example.org/schemas/"> 3969 <elem1>...</elem1> 3970 <elem2>...</elem2> 3971</mydoc></pre> 3972 3973<p>The namespace value has to be an absolute URL, but the URL doesn't have to 3974point to any existing resource on the Web. It will bind all the element and 3975attributes with that URL. I suggest to use an URL within a domain you 3976control, and that the URL should contain some kind of version information if 3977possible. For example, <code>"http://www.gnome.org/gnumeric/1.0/"</code> is a 3978good namespace scheme.</p> 3979 3980<p>Then when you load a file, make sure that a namespace carrying the 3981version-independent prefix is installed on the root element of your document, 3982and if the version information don't match something you know, warn the user 3983and be liberal in what you accept as the input. Also do *not* try to base 3984namespace checking on the prefix value. <foo:text> may be exactly the 3985same as <bar:text> in another document. What really matters is the URI 3986associated with the element or the attribute, not the prefix string (which is 3987just a shortcut for the full URI). In libxml, element and attributes have an 3988<code>ns</code> field pointing to an xmlNs structure detailing the namespace 3989prefix and its URI.</p> 3990 3991<p>@@Interfaces@@</p> 3992<pre>xmlNodePtr node; 3993if(!strncmp(node->name,"mytag",5) 3994 && node->ns 3995 && !strcmp(node->ns->href,"http://www.mysite.com/myns/1.0")) { 3996 ... 3997}</pre> 3998 3999<p>Usually people object to using namespaces together with validity checking. 4000I will try to make sure that using namespaces won't break validity checking, 4001so even if you plan to use or currently are using validation I strongly 4002suggest adding namespaces to your document. A default namespace scheme 4003<code>xmlns="http://...."</code> should not break validity even on less 4004flexible parsers. Using namespaces to mix and differentiate content coming 4005from multiple DTDs will certainly break current validation schemes. To check 4006such documents one needs to use schema-validation, which is supported in 4007libxml2 as well. See <a href="http://www.relaxng.org/">relagx-ng</a> and <a 4008href="http://www.w3c.org/XML/Schema">w3c-schema</a>.</p> 4009 4010<h2><a name="Upgrading">Upgrading 1.x code</a></h2> 4011 4012<p>Incompatible changes:</p> 4013 4014<p>Version 2 of libxml2 is the first version introducing serious backward 4015incompatible changes. The main goals were:</p> 4016<ul> 4017 <li>a general cleanup. A number of mistakes inherited from the very early 4018 versions couldn't be changed due to compatibility constraints. Example 4019 the "childs" element in the nodes.</li> 4020 <li>Uniformization of the various nodes, at least for their header and link 4021 parts (doc, parent, children, prev, next), the goal is a simpler 4022 programming model and simplifying the task of the DOM implementors.</li> 4023 <li>better conformances to the XML specification, for example version 1.x 4024 had an heuristic to try to detect ignorable white spaces. As a result the 4025 SAX event generated were ignorableWhitespace() while the spec requires 4026 character() in that case. This also mean that a number of DOM node 4027 containing blank text may populate the DOM tree which were not present 4028 before.</li> 4029</ul> 4030 4031<h3>How to fix libxml-1.x code:</h3> 4032 4033<p>So client code of libxml designed to run with version 1.x may have to be 4034changed to compile against version 2.x of libxml. Here is a list of changes 4035that I have collected, they may not be sufficient, so in case you find other 4036change which are required, <a href="mailto:Daniel.Veillard@w3.org">drop me a 4037mail</a>:</p> 4038<ol> 4039 <li>The package name have changed from libxml to libxml2, the library name 4040 is now -lxml2 . There is a new xml2-config script which should be used to 4041 select the right parameters libxml2</li> 4042 <li>Node <strong>childs</strong> field has been renamed 4043 <strong>children</strong> so s/childs/children/g should be applied 4044 (probability of having "childs" anywhere else is close to 0+</li> 4045 <li>The document don't have anymore a <strong>root</strong> element it has 4046 been replaced by <strong>children</strong> and usually you will get a 4047 list of element here. For example a Dtd element for the internal subset 4048 and it's declaration may be found in that list, as well as processing 4049 instructions or comments found before or after the document root element. 4050 Use <strong>xmlDocGetRootElement(doc)</strong> to get the root element of 4051 a document. Alternatively if you are sure to not reference DTDs nor have 4052 PIs or comments before or after the root element 4053 s/->root/->children/g will probably do it.</li> 4054 <li>The white space issue, this one is more complex, unless special case of 4055 validating parsing, the line breaks and spaces usually used for indenting 4056 and formatting the document content becomes significant. So they are 4057 reported by SAX and if your using the DOM tree, corresponding nodes are 4058 generated. Too approach can be taken: 4059 <ol> 4060 <li>lazy one, use the compatibility call 4061 <strong>xmlKeepBlanksDefault(0)</strong> but be aware that you are 4062 relying on a special (and possibly broken) set of heuristics of 4063 libxml to detect ignorable blanks. Don't complain if it breaks or 4064 make your application not 100% clean w.r.t. to it's input.</li> 4065 <li>the Right Way: change you code to accept possibly insignificant 4066 blanks characters, or have your tree populated with weird blank text 4067 nodes. You can spot them using the commodity function 4068 <strong>xmlIsBlankNode(node)</strong> returning 1 for such blank 4069 nodes.</li> 4070 </ol> 4071 <p>Note also that with the new default the output functions don't add any 4072 extra indentation when saving a tree in order to be able to round trip 4073 (read and save) without inflating the document with extra formatting 4074 chars.</p> 4075 </li> 4076 <li>The include path has changed to $prefix/libxml/ and the includes 4077 themselves uses this new prefix in includes instructions... If you are 4078 using (as expected) the 4079 <pre>xml2-config --cflags</pre> 4080 <p>output to generate you compile commands this will probably work out of 4081 the box</p> 4082 </li> 4083 <li>xmlDetectCharEncoding takes an extra argument indicating the length in 4084 byte of the head of the document available for character detection.</li> 4085</ol> 4086 4087<h3>Ensuring both libxml-1.x and libxml-2.x compatibility</h3> 4088 4089<p>Two new version of libxml (1.8.11) and libxml2 (2.3.4) have been released 4090to allow smooth upgrade of existing libxml v1code while retaining 4091compatibility. They offers the following:</p> 4092<ol> 4093 <li>similar include naming, one should use 4094 <strong>#include<libxml/...></strong> in both cases.</li> 4095 <li>similar identifiers defined via macros for the child and root fields: 4096 respectively <strong>xmlChildrenNode</strong> and 4097 <strong>xmlRootNode</strong></li> 4098 <li>a new macro <strong>LIBXML_TEST_VERSION</strong> which should be 4099 inserted once in the client code</li> 4100</ol> 4101 4102<p>So the roadmap to upgrade your existing libxml applications is the 4103following:</p> 4104<ol> 4105 <li>install the libxml-1.8.8 (and libxml-devel-1.8.8) packages</li> 4106 <li>find all occurrences where the xmlDoc <strong>root</strong> field is 4107 used and change it to <strong>xmlRootNode</strong></li> 4108 <li>similarly find all occurrences where the xmlNode 4109 <strong>childs</strong> field is used and change it to 4110 <strong>xmlChildrenNode</strong></li> 4111 <li>add a <strong>LIBXML_TEST_VERSION</strong> macro somewhere in your 4112 <strong>main()</strong> or in the library init entry point</li> 4113 <li>Recompile, check compatibility, it should still work</li> 4114 <li>Change your configure script to look first for xml2-config and fall 4115 back using xml-config . Use the --cflags and --libs output of the command 4116 as the Include and Linking parameters needed to use libxml.</li> 4117 <li>install libxml2-2.3.x and libxml2-devel-2.3.x (libxml-1.8.y and 4118 libxml-devel-1.8.y can be kept simultaneously)</li> 4119 <li>remove your config.cache, relaunch your configuration mechanism, and 4120 recompile, if steps 2 and 3 were done right it should compile as-is</li> 4121 <li>Test that your application is still running correctly, if not this may 4122 be due to extra empty nodes due to formating spaces being kept in libxml2 4123 contrary to libxml1, in that case insert xmlKeepBlanksDefault(1) in your 4124 code before calling the parser (next to 4125 <strong>LIBXML_TEST_VERSION</strong> is a fine place).</li> 4126</ol> 4127 4128<p>Following those steps should work. It worked for some of my own code.</p> 4129 4130<p>Let me put some emphasis on the fact that there is far more changes from 4131libxml 1.x to 2.x than the ones you may have to patch for. The overall code 4132has been considerably cleaned up and the conformance to the XML specification 4133has been drastically improved too. Don't take those changes as an excuse to 4134not upgrade, it may cost a lot on the long term ...</p> 4135 4136<h2><a name="Thread">Thread safety</a></h2> 4137 4138<p>Starting with 2.4.7, libxml2 makes provisions to ensure that concurrent 4139threads can safely work in parallel parsing different documents. There is 4140however a couple of things to do to ensure it:</p> 4141<ul> 4142 <li>configure the library accordingly using the --with-threads options</li> 4143 <li>call xmlInitParser() in the "main" thread before using any of the 4144 libxml2 API (except possibly selecting a different memory allocator)</li> 4145</ul> 4146 4147<p>Note that the thread safety cannot be ensured for multiple threads sharing 4148the same document, the locking must be done at the application level, libxml 4149exports a basic mutex and reentrant mutexes API in <libxml/threads.h>. 4150The parts of the library checked for thread safety are:</p> 4151<ul> 4152 <li>concurrent loading</li> 4153 <li>file access resolution</li> 4154 <li>catalog access</li> 4155 <li>catalog building</li> 4156 <li>entities lookup/accesses</li> 4157 <li>validation</li> 4158 <li>global variables per-thread override</li> 4159 <li>memory handling</li> 4160</ul> 4161 4162<p>XPath is supposed to be thread safe now, but this wasn't tested 4163seriously.</p> 4164 4165<h2><a name="DOM"></a><a name="Principles">DOM Principles</a></h2> 4166 4167<p><a href="http://www.w3.org/DOM/">DOM</a> stands for the <em>Document 4168Object Model</em>; this is an API for accessing XML or HTML structured 4169documents. Native support for DOM in Gnome is on the way (module gnome-dom), 4170and will be based on gnome-xml. This will be a far cleaner interface to 4171manipulate XML files within Gnome since it won't expose the internal 4172structure.</p> 4173 4174<p>The current DOM implementation on top of libxml2 is the <a 4175href="http://cvs.gnome.org/lxr/source/gdome2/">gdome2 Gnome module</a>, this 4176is a full DOM interface, thanks to Paolo Casarini, check the <a 4177href="http://www.cs.unibo.it/~casarini/gdome2/">Gdome2 homepage</a> for more 4178informations.</p> 4179 4180<h2><a name="Example"></a><a name="real">A real example</a></h2> 4181 4182<p>Here is a real size example, where the actual content of the application 4183data is not kept in the DOM tree but uses internal structures. It is based on 4184a proposal to keep a database of jobs related to Gnome, with an XML based 4185storage structure. Here is an <a href="gjobs.xml">XML encoded jobs 4186base</a>:</p> 4187<pre><?xml version="1.0"?> 4188<gjob:Helping xmlns:gjob="http://www.gnome.org/some-location"> 4189 <gjob:Jobs> 4190 4191 <gjob:Job> 4192 <gjob:Project ID="3"/> 4193 <gjob:Application>GBackup</gjob:Application> 4194 <gjob:Category>Development</gjob:Category> 4195 4196 <gjob:Update> 4197 <gjob:Status>Open</gjob:Status> 4198 <gjob:Modified>Mon, 07 Jun 1999 20:27:45 -0400 MET DST</gjob:Modified> 4199 <gjob:Salary>USD 0.00</gjob:Salary> 4200 </gjob:Update> 4201 4202 <gjob:Developers> 4203 <gjob:Developer> 4204 </gjob:Developer> 4205 </gjob:Developers> 4206 4207 <gjob:Contact> 4208 <gjob:Person>Nathan Clemons</gjob:Person> 4209 <gjob:Email>nathan@windsofstorm.net</gjob:Email> 4210 <gjob:Company> 4211 </gjob:Company> 4212 <gjob:Organisation> 4213 </gjob:Organisation> 4214 <gjob:Webpage> 4215 </gjob:Webpage> 4216 <gjob:Snailmail> 4217 </gjob:Snailmail> 4218 <gjob:Phone> 4219 </gjob:Phone> 4220 </gjob:Contact> 4221 4222 <gjob:Requirements> 4223 The program should be released as free software, under the GPL. 4224 </gjob:Requirements> 4225 4226 <gjob:Skills> 4227 </gjob:Skills> 4228 4229 <gjob:Details> 4230 A GNOME based system that will allow a superuser to configure 4231 compressed and uncompressed files and/or file systems to be backed 4232 up with a supported media in the system. This should be able to 4233 perform via find commands generating a list of files that are passed 4234 to tar, dd, cpio, cp, gzip, etc., to be directed to the tape machine 4235 or via operations performed on the filesystem itself. Email 4236 notification and GUI status display very important. 4237 </gjob:Details> 4238 4239 </gjob:Job> 4240 4241 </gjob:Jobs> 4242</gjob:Helping></pre> 4243 4244<p>While loading the XML file into an internal DOM tree is a matter of 4245calling only a couple of functions, browsing the tree to gather the data and 4246generate the internal structures is harder, and more error prone.</p> 4247 4248<p>The suggested principle is to be tolerant with respect to the input 4249structure. For example, the ordering of the attributes is not significant, 4250the XML specification is clear about it. It's also usually a good idea not to 4251depend on the order of the children of a given node, unless it really makes 4252things harder. Here is some code to parse the information for a person:</p> 4253<pre>/* 4254 * A person record 4255 */ 4256typedef struct person { 4257 char *name; 4258 char *email; 4259 char *company; 4260 char *organisation; 4261 char *smail; 4262 char *webPage; 4263 char *phone; 4264} person, *personPtr; 4265 4266/* 4267 * And the code needed to parse it 4268 */ 4269personPtr parsePerson(xmlDocPtr doc, xmlNsPtr ns, xmlNodePtr cur) { 4270 personPtr ret = NULL; 4271 4272DEBUG("parsePerson\n"); 4273 /* 4274 * allocate the struct 4275 */ 4276 ret = (personPtr) malloc(sizeof(person)); 4277 if (ret == NULL) { 4278 fprintf(stderr,"out of memory\n"); 4279 return(NULL); 4280 } 4281 memset(ret, 0, sizeof(person)); 4282 4283 /* We don't care what the top level element name is */ 4284 cur = cur->xmlChildrenNode; 4285 while (cur != NULL) { 4286 if ((!strcmp(cur->name, "Person")) && (cur->ns == ns)) 4287 ret->name = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 4288 if ((!strcmp(cur->name, "Email")) && (cur->ns == ns)) 4289 ret->email = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 4290 cur = cur->next; 4291 } 4292 4293 return(ret); 4294}</pre> 4295 4296<p>Here are a couple of things to notice:</p> 4297<ul> 4298 <li>Usually a recursive parsing style is the more convenient one: XML data 4299 is by nature subject to repetitive constructs and usually exhibits highly 4300 structured patterns.</li> 4301 <li>The two arguments of type <em>xmlDocPtr</em> and <em>xmlNsPtr</em>, 4302 i.e. the pointer to the global XML document and the namespace reserved to 4303 the application. Document wide information are needed for example to 4304 decode entities and it's a good coding practice to define a namespace for 4305 your application set of data and test that the element and attributes 4306 you're analyzing actually pertains to your application space. This is 4307 done by a simple equality test (cur->ns == ns).</li> 4308 <li>To retrieve text and attributes value, you can use the function 4309 <em>xmlNodeListGetString</em> to gather all the text and entity reference 4310 nodes generated by the DOM output and produce an single text string.</li> 4311</ul> 4312 4313<p>Here is another piece of code used to parse another level of the 4314structure:</p> 4315<pre>#include <libxml/tree.h> 4316/* 4317 * a Description for a Job 4318 */ 4319typedef struct job { 4320 char *projectID; 4321 char *application; 4322 char *category; 4323 personPtr contact; 4324 int nbDevelopers; 4325 personPtr developers[100]; /* using dynamic alloc is left as an exercise */ 4326} job, *jobPtr; 4327 4328/* 4329 * And the code needed to parse it 4330 */ 4331jobPtr parseJob(xmlDocPtr doc, xmlNsPtr ns, xmlNodePtr cur) { 4332 jobPtr ret = NULL; 4333 4334DEBUG("parseJob\n"); 4335 /* 4336 * allocate the struct 4337 */ 4338 ret = (jobPtr) malloc(sizeof(job)); 4339 if (ret == NULL) { 4340 fprintf(stderr,"out of memory\n"); 4341 return(NULL); 4342 } 4343 memset(ret, 0, sizeof(job)); 4344 4345 /* We don't care what the top level element name is */ 4346 cur = cur->xmlChildrenNode; 4347 while (cur != NULL) { 4348 4349 if ((!strcmp(cur->name, "Project")) && (cur->ns == ns)) { 4350 ret->projectID = xmlGetProp(cur, "ID"); 4351 if (ret->projectID == NULL) { 4352 fprintf(stderr, "Project has no ID\n"); 4353 } 4354 } 4355 if ((!strcmp(cur->name, "Application")) && (cur->ns == ns)) 4356 ret->application = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 4357 if ((!strcmp(cur->name, "Category")) && (cur->ns == ns)) 4358 ret->category = xmlNodeListGetString(doc, cur->xmlChildrenNode, 1); 4359 if ((!strcmp(cur->name, "Contact")) && (cur->ns == ns)) 4360 ret->contact = parsePerson(doc, ns, cur); 4361 cur = cur->next; 4362 } 4363 4364 return(ret); 4365}</pre> 4366 4367<p>Once you are used to it, writing this kind of code is quite simple, but 4368boring. Ultimately, it could be possible to write stubbers taking either C 4369data structure definitions, a set of XML examples or an XML DTD and produce 4370the code needed to import and export the content between C data and XML 4371storage. This is left as an exercise to the reader :-)</p> 4372 4373<p>Feel free to use <a href="example/gjobread.c">the code for the full C 4374parsing example</a> as a template, it is also available with Makefile in the 4375Gnome CVS base under gnome-xml/example</p> 4376 4377<h2><a name="Contributi">Contributions</a></h2> 4378<ul> 4379 <li>Bjorn Reese, William Brack and Thomas Broyer have provided a number of 4380 patches, Gary Pennington worked on the validation API, threading support 4381 and Solaris port.</li> 4382 <li>John Fleck helps maintaining the documentation and man pages.</li> 4383 <li><a href="mailto:igor@zlatkovic.com">Igor Zlatkovic</a> is now the 4384 maintainer of the Windows port, <a 4385 href="http://www.zlatkovic.com/projects/libxml/index.html">he provides 4386 binaries</a></li> 4387 <li><a href="mailto:Gary.Pennington@sun.com">Gary Pennington</a> provides 4388 <a href="http://garypennington.net/libxml2/">Solaris binaries</a></li> 4389 <li><a 4390 href="http://mail.gnome.org/archives/xml/2001-March/msg00014.html">Matt 4391 Sergeant</a> developed <a 4392 href="http://axkit.org/download/">XML::LibXSLT</a>, a Perl wrapper for 4393 libxml2/libxslt as part of the <a href="http://axkit.com/">AxKit XML 4394 application server</a></li> 4395 <li><a href="mailto:fnatter@gmx.net">Felix Natter</a> and <a 4396 href="mailto:geertk@ai.rug.nl">Geert Kloosterman</a> provide <a 4397 href="libxml-doc.el">an emacs module</a> to lookup libxml(2) functions 4398 documentation</li> 4399 <li><a href="mailto:sherwin@nlm.nih.gov">Ziying Sherwin</a> provided <a 4400 href="http://xmlsoft.org/messages/0488.html">man pages</a></li> 4401 <li>there is a module for <a 4402 href="http://acs-misc.sourceforge.net/nsxml.html">libxml/libxslt support 4403 in OpenNSD/AOLServer</a></li> 4404 <li><a href="mailto:dkuhlman@cutter.rexx.com">Dave Kuhlman</a> provided the 4405 first version of libxml/libxslt <a 4406 href="http://www.rexx.com/~dkuhlman">wrappers for Python</a></li> 4407 <li>Petr Kozelka provides <a 4408 href="http://sourceforge.net/projects/libxml2-pas">Pascal units to glue 4409 libxml2</a> with Kylix and Delphi and other Pascal compilers</li> 4410 <li><a href="mailto:aleksey@aleksey.com">Aleksey Sanin</a> implemented the 4411 <a href="http://www.w3.org/Signature/">XML Canonicalization and XML 4412 Digital Signature</a> <a 4413 href="http://www.aleksey.com/xmlsec/">implementations for libxml2</a></li> 4414 <li><a href="mailto:Steve.Ball@zveno.com">Steve Ball</a>, <a 4415 href="http://www.zveno.com/">Zveno</a> and contributors maintain <a 4416 href="http://tclxml.sourceforge.net/">tcl bindings for libxml2 and 4417 libxslt</a>, as well as <a 4418 href="http://tclxml.sf.net/tkxmllint.html">tkxmllint</a> a GUI for 4419 xmllint and <a href="http://tclxml.sf.net/tkxsltproc.html">tkxsltproc</a> 4420 a GUI for xsltproc.</li> 4421</ul> 4422 4423<p></p> 4424</body> 4425</html> 4426
