Update export schema to version 0.3, including some site configuration data.

diff --git a/docs/export-0.3.xsd b/docs/export-0.3.xsd
new file mode 100644 (file)
index 0000000..02403c6
--- /dev/null
+++ b/docs/export-0.3.xsd
@@ -0,0 +1,146 @@
+<?xml version="1.0" encoding="UTF-8" ?>
+<!--
+       This is an XML Schema description of the format
+       output by MediaWiki's Special:Export system.
+       
+       Version 0.2 adds optional basic file upload info support,
+       which is used by our OAI export/import submodule.
+       
+       Version 0.3 adds some site configuration information such
+       as a list of defined namespaces.
+       
+       The canonical URL to the schema document is:
+       http://www.mediawiki.org/xml/export-0.3.xsd
+       
+       Use the namespace:
+       http://www.mediawiki.org/xml/export-0.3/
+-->
+<schema xmlns="http://www.w3.org/2001/XMLSchema"
+        xmlns:mw="http://www.mediawiki.org/xml/export-0.3/"
+        targetNamespace="http://www.mediawiki.org/xml/export-0.3/"
+        elementFormDefault="qualified">
+
+       <annotation>
+               <documentation xml:lang="en">
+                       MediaWiki's page export format
+               </documentation>
+       </annotation>
+       
+       <!-- Need this to reference xml:lang -->
+       <import namespace="http://www.w3.org/XML/1998/namespace"
+               schemaLocation="http://www.w3.org/2001/xml.xsd"/>
+       
+       <!-- Our root element -->
+       <element name="mediawiki" type="mw:MediaWikiType"/>
+       
+       <complexType name="MediaWikiType">
+               <sequence>
+                       <element name="siteinfo" type="mw:SiteInfoType"
+                               minOccurs="0" maxOccurs="1"/>
+                       <element name="page" type="mw:PageType"
+                               minOccurs="0" maxOccurs="unbounded"/>
+               </sequence>
+               <attribute name="version" type="string" use="required"/>
+               <attribute ref="xml:lang" use="required"/>
+       </complexType>
+       
+       <complexType name="SiteInfoType">
+               <sequence>
+                       <element name="sitename" type="string" minOccurs="0" />
+                       <element name="base" type="anyURI" minOccurs="0" />
+                       <element name="generator" type="string" minOccurs="0" />
+                       <element name="case" type="mw:CaseType" minOccurs="0" />
+                       <element name="namespaces" type="mw:NamespacesType" minOccurs="0" />
+               </sequence>
+       </complexType>
+       
+       <simpleType name="CaseType">
+               <restriction base="NMTOKEN">
+                       <!-- Cannot have two titles differing only by case of first letter. -->
+                       <!-- Default behavior through 1.5, $wgCapitalLinks = true -->
+                       <enumeration value="first-letter" />
+                       
+                       <!-- Complete title is case-sensitive -->
+                       <!-- Behavior when $wgCapitalLinks = false -->
+                       <enumeration value="case-sensitive" />
+                       
+                       <!-- Cannot have two titles differing only by case. -->
+                       <!-- Not yet implemented as of MediaWiki 1.5 -->
+                       <enumeration value="case-insensitive" />
+               </restriction>
+       </simpleType>
+       
+       <complexType name="NamespacesType">
+               <sequence>
+                       <element name="namespace" type="mw:NamespaceType"
+                               minOccurs="0" maxOccurs="unbounded" />
+               </sequence>
+       </complexType>
+       
+       <complexType name="NamespaceType">
+               <simpleContent>
+                       <extension base="string">
+                               <attribute name="key" type="integer" />
+                       </extension>
+               </simpleContent>
+       </complexType>
+       
+       <complexType name="PageType">
+               <sequence>
+                       <!-- Title in text form. (Using spaces, not underscores; with namespace ) -->
+                       <element name="title" type="string"/>
+                       
+                       <!-- optional page ID number -->
+                       <element name="id" type="positiveInteger" minOccurs="0"/>
+                       
+                       <!-- comma-separated list of string tokens, if present -->
+                       <element name="restrictions" type="string" minOccurs="0"/>
+                       
+                       <!-- Zero or more sets of revision or upload data -->
+                       <choice minOccurs="0" maxOccurs="unbounded">
+                               <element name="revision" type="mw:RevisionType" />
+                               <element name="upload" type="mw:UploadType" />
+                       </choice>
+               </sequence>
+       </complexType>
+       
+       <complexType name="RevisionType">
+               <sequence>
+                       <element name="id" type="positiveInteger" minOccurs="0"/>
+                       <element name="timestamp" type="dateTime"/>
+                       <element name="contributor" type="mw:ContributorType"/>
+                       <element name="minor" minOccurs="0" />
+                       <element name="comment" type="string" minOccurs="0"/>
+                       <element name="text" type="string"/>
+               </sequence>
+       </complexType>
+       
+       <complexType name="ContributorType">
+               <sequence>
+                       <element name="username" type="string" minOccurs="0"/>
+                       <element name="id" type="positiveInteger" minOccurs="0" />
+                       
+                       <element name="ip" type="string" minOccurs="0"/>
+               </sequence>
+       </complexType>
+       
+       <complexType name="UploadType">
+               <sequence>
+                       <!-- Revision-style data... -->
+                       <element name="timestamp" type="dateTime"/>
+                       <element name="contributor" type="mw:ContributorType"/>
+                       <element name="comment" type="string" minOccurs="0"/>
+                       
+                       <!-- Filename. (Using underscores, not spaces. No 'Image:' namespace marker.) -->
+                       <element name="filename" type="string"/>
+                       
+                       <!-- URI at which this resource can be obtained -->
+                       <element name="src" type="anyURI"/>
+                       
+                       <element name="size" type="positiveInteger" />
+                       
+                       <!-- TODO: add other metadata fields -->
+               </sequence>
+       </complexType>
+       
+</schema>

diff --git a/docs/export-demo.xml b/docs/export-demo.xml
new file mode 100644 (file)
index 0000000..941a097
--- /dev/null
+++ b/docs/export-demo.xml
@@ -0,0 +1,115 @@
+<mediawiki xmlns="http://www.mediawiki.org/xml/export-0.3/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.mediawiki.org/xml/export-0.3/ http://www.mediawiki.org/xml/export-0.3.xsd" version="0.3" xml:lang="en">
+
+<!-- Optional global configuration info -->
+<siteinfo>
+  <!-- Site name, as set in $wgSitename -->
+  <sitename>DemoWiki</sitename>
+  
+  <!-- Forgot where you got this set? -->
+  <base>http://example.com/wiki/Main_Page</base>
+  
+  <!-- Source software version -->
+  <generator>MediaWiki 1.5.0</generator>
+  
+  <!-- Title case sensitivity options of the wiki this data came from -->
+  <!-- May be 'first-letter', 'case-sensitive', or 'case-insensitive' -->
+  <case>first-letter</case>
+  
+  <!-- Defined namespace keys on the source wiki. -->
+  <!-- Titles can be substring-split to obtain the symbolic numeric key -->
+  <namespaces>
+    <namespace key="-2">Media</namespace>
+    <namespace key="-1">Special</namespace>
+    <namespace key="0"></namespace>
+    <namespace key="1">Talk</namespace>
+    <namespace key="2">User</namespace>
+    <namespace key="3">User talk</namespace>
+    <namespace key="4">DemoWiki</namespace>
+    <namespace key="5">DemoWIki talk</namespace>
+    <namespace key="6">Image</namespace>
+    <namespace key="7">Image talk</namespace>
+    <namespace key="8">MediaWiki</namespace>
+    <namespace key="9">MediaWiki talk</namespace>
+    <namespace key="10">Template</namespace>
+    <namespace key="11">Template talk</namespace>
+    <namespace key="12">Help</namespace>
+    <namespace key="13">Help talk</namespace>
+    <namespace key="14">Category</namespace>
+    <namespace key="15">Category talk</namespace>
+  </namespaces>
+</siteinfo>
+
+<!-- The rest of the data will be a series of page records -->
+<page>
+  <!-- Titles are listed here in text form, with namespace prefix   -->
+  <!-- if any, and spaces rather than the underscores used in URLs. -->
+  <title>Page title</title>
+  
+  <!-- The page's immutable page_id number in the source database.  -->
+  <!-- Page ID numbers are kept across page moves, but may change   -->
+  <!-- if a page is deleted and recreated.                          -->
+  <id>1</id>
+  
+  <!-- If restricted, the ACL is listed here raw. -->
+  <restrictions>edit=sysop:move=sysop</restrictions>
+  
+  <!-- With a series of revision records... -->
+  
+  <!-- Remember this is XML; if you must use a regex-based extractor -->
+  <!-- in place of a standard XML parser, be very careful.           -->
+  <!-- * Don't forget to decode character entities!                  -->
+  <!-- * If using a 'loose' XML parser, ensure that whitespace is    -->
+  <!--    preserved in the <text> elements.                          -->
+  <revision>
+    <!-- Unique revision ID number (rev_id) in the source database.  -->
+    <!-- This number uniquely identifies the revision on that wiki.  -->
+    <id>100</id>
+    
+    <timestamp>2001-01-15T13:15:00Z</timestamp>
+    <contributor><username>Foobar</username><id>42</id></contributor>
+    <minor />
+    <comment>I have just one thing to say!</comment>
+    <text>A bunch of [[text]] here.</text>
+  </revision>
+  
+  <revision>
+    <timestamp>2001-01-15T13:10:27Z</timestamp>
+    <contributor><ip>10.0.0.2</ip></contributor>
+    <comment>new!</comment>
+    <text>An earlier [[revision]].</text>
+  </revision>
+</page>
+
+<page>
+  <title>Talk:Page title</title>
+  <id>2</id>
+  <revision>
+    <id>101</id>
+    <timestamp>2001-01-15T14:03:00Z</timestamp>
+    <contributor><ip>10.0.0.2</ip></contributor>
+    <comment>hey</comment>
+    <text>WHYD YOU LOCK PAGE??!!! i was editing that jerk</text>
+  </revision>
+</page>
+
+<page>
+  <title>Image:Some image.jpg</title>
+  <id>3</id>
+  <revision>
+    <id>102</id>
+    <timestamp>2001-01-15T20:34:12Z</timestamp>
+    <contributor><username>Foobar</username><id>42</id></contributor>
+    <comment>My awesomeest image!</comment>
+    <text>This is an awesome little imgae. I lurves it. {{PD}}</text>
+  </revision>
+  <upload>
+    <timestamp>2001-01-15T20:34:12Z</timestamp>
+    <contributor><username>Foobar</username><id>42</id></contributor>
+    <comment>My awesomeest image!</comment>
+    <filename>Some_image.jpg</filename>
+    <src>http://upload.wikimedia.org/commons/2/22/Some_image.jpg</src>
+    <size>12345</size>
+  </upload>
+</page>
+
+</mediawiki>