Updated export schema: added an <upload> element which may be used in

<page>s alongside <revision>s. For image pages these will contain
file data: filename, size, URL at which it can be found.
This is used by the OAI updater system.

diff --git a/docs/export-0.2.xsd b/docs/export-0.2.xsd
new file mode 100644 (file)
index 0000000..8acbf54
--- /dev/null
+++ b/docs/export-0.2.xsd
@@ -0,0 +1,100 @@
+<?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.
+       
+       The canonical URL to the schema document is:
+       http://www.mediawiki.org/xml/export-0.2.xsd
+       
+       Use the namespace:
+       http://www.mediawiki.org/xml/export-0.2/
+-->
+<schema xmlns="http://www.w3.org/2001/XMLSchema"
+        xmlns:mw="http://www.mediawiki.org/xml/export-0.2/"
+        targetNamespace="http://www.mediawiki.org/xml/export-0.2/"
+        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="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="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>