Commit to fix bug 30513.
authorDiederik van Liere <diederik@users.mediawiki.org>
Wed, 9 Nov 2011 21:15:48 +0000 (21:15 +0000)
committerDiederik van Liere <diederik@users.mediawiki.org>
Wed, 9 Nov 2011 21:15:48 +0000 (21:15 +0000)
docs/export-0.5.xsd
docs/export-0.6.xsd [new file with mode: 0644]
includes/Export.php

index ca54c93..469446d 100644 (file)
@@ -12,6 +12,8 @@
        Version 0.4 adds per-revision delete flags, log exports,
        discussion threading data, a per-page redirect flag, and
        per-namespace capitalization.
+       
+       Version 0.5 adds byte count per revision.
 
        The canonical URL to the schema document is:
        http://www.mediawiki.org/xml/export-0.5.xsd
diff --git a/docs/export-0.6.xsd b/docs/export-0.6.xsd
new file mode 100644 (file)
index 0000000..a225e52
--- /dev/null
@@ -0,0 +1,225 @@
+<?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.
+       
+       Version 0.4 adds per-revision delete flags, log exports,
+       discussion threading data, a per-page redirect flag, and
+       per-namespace capitalization.
+
+       Version 0.5 adds byte count per revision.
+       
+       Version 0.6 adds a separate namespace tag, locates the
+       redirect target and strips of the namespace from the title. 
+       
+       The canonical URL to the schema document is:
+       http://www.mediawiki.org/xml/export-0.6.xsd
+       
+       Use the namespace:
+       http://www.mediawiki.org/xml/export-0.6/
+-->
+<schema xmlns="http://www.w3.org/2001/XMLSchema"
+        xmlns:mw="http://www.mediawiki.org/xml/export-0.6/"
+        targetNamespace="http://www.mediawiki.org/xml/export-0.6/"
+        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 non-case senstitive titles eg [[FOO]] == [[Foo]] -->
+                       <!-- Not yet implemented as of MediaWiki 1.18 -->
+                       <enumeration value="case-insensitive" />
+               </restriction>
+       </simpleType>
+       
+       <simpleType name="DeletedFlagType">
+               <restriction base="NMTOKEN">
+                       <enumeration value="deleted"/>
+               </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" />
+                               <attribute name="case" type="mw:CaseType" />
+                       </extension>
+               </simpleContent>
+       </complexType>
+       
+       <complexType name="PageType">
+               <sequence>
+                       <!-- Title in text form. (Using spaces, not underscores; with namespace ) -->
+                       <element name="title" type="string"/>
+                       
+                       <!-- Namesapce in text form -->
+                       <element name="ns" type="string"/>
+                       
+                       <!-- optional page ID number -->
+                       <element name="id" type="positiveInteger" minOccurs="0"/>
+                       
+                       <!-- flag if the current revision is a redirect -->
+                       <element name="redirect" type="string" 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" />
+                               <element name="logitem" type="mw:LogItemType" />
+                       </choice>
+                       
+                       <!-- Zero or One sets of discussion threading data -->
+                       <element name="discussionthreadinginfo" minOccurs="0" maxOccurs="1" type="mw:DiscussionThreadingInfo" />
+               </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="mw:CommentType" minOccurs="0"/>
+                       <element name="text" type="mw:TextType" />
+               </sequence>
+       </complexType>
+
+       <complexType name="LogItemType">
+               <sequence>
+                       <element name="id" type="positiveInteger" minOccurs="0"/>
+                       <element name="timestamp" type="dateTime"/>
+                       <element name="contributor" type="mw:ContributorType"/>
+                       <element name="comment" type="mw:CommentType" minOccurs="0"/>
+                       <element name="type" type="string" />
+                       <element name="action" type="string" />
+                       <element name="text" type="mw:TextType" />
+               </sequence>
+       </complexType>
+       
+       <complexType name="CommentType">
+               <simpleContent>
+                       <extension base="string">
+                               <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
+                               <attribute name="deleted" use="optional" type="mw:DeletedFlagType"/>
+                       </extension>
+               </simpleContent>
+       </complexType>
+
+
+       <complexType name="TextType">
+               <simpleContent>
+                       <extension base="string">
+                               <attribute ref="xml:space" use="optional" default="preserve" />
+                               <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
+                               <attribute name="deleted" use="optional" type="mw:DeletedFlagType"/>    
+                               <!-- This isn't a good idea; we should be using "ID" instead of "NMTOKEN" -->
+                               <!-- However, "NMTOKEN" is strictest definition that is both compatible with existing -->
+                               <!-- usage ([0-9]+) and with the "ID" type. -->
+                               <attribute name="id" type="NMTOKEN"/>
+                               <attribute name="bytes" use="optional" type="nonNegativeInteger"/>
+                       </extension>
+               </simpleContent>
+       </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>
+               <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
+               <attribute name="deleted" use="optional" type="mw:DeletedFlagType"/>
+       </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>
+       
+       <!-- Discussion threading data for LiquidThreads -->
+       <complexType name="DiscussionThreadingInfo">
+               <sequence>
+                       <element name="ThreadSubject" type="string" />
+                       <element name="ThreadParent" type="positiveInteger" />
+                       <element name="ThreadAncestor" type="positiveInteger" />
+                       <element name="ThreadPage" type="string" />
+                       <element name="ThreadID" type="positiveInteger" />
+                       <element name="ThreadAuthor" type="string" />
+                       <element name="ThreadEditStatus" type="string" />
+                       <element name="ThreadType" type="string" />
+               </sequence>
+       </complexType>
+       
+</schema>
index a9f29cb..53211e1 100644 (file)
@@ -380,7 +380,7 @@ class XmlDumpWriter {
         * @return string
         */
        function schemaVersion() {
-               return "0.5";
+               return "0.6";
        }
 
        /**
@@ -476,10 +476,15 @@ class XmlDumpWriter {
        function openPage( $row ) {
                $out = "  <page>\n";
                $title = Title::makeTitle( $row->page_namespace, $row->page_title );
-               $out .= '    ' . Xml::elementClean( 'title', array(), $title->getPrefixedText() ) . "\n";
+               $out .= '    ' . Xml::elementClean( 'title', array(), strval( $title->getPrefixedText() ) ) . "\n";
+               $out .= '    ' . Xml::element( 'ns', array(), strval( $row->page_namespace) ) . "\n";
                $out .= '    ' . Xml::element( 'id', array(), strval( $row->page_id ) ) . "\n";
                if ( $row->page_is_redirect ) {
-                       $out .= '    ' . Xml::element( 'redirect', array() ) . "\n";
+                       $page = WikiPage::factory( $title );
+                       $redirect = $page->getRedirectTarget();
+                       if ( $redirect !== null ) {
+                               $out .= '    ' . Xml::element( 'redirect', array( 'title' => strval( $redirect ) ) ) . "\n";
+                       }
                }
                if ( $row->page_restrictions != '' ) {
                        $out .= '    ' . Xml::element( 'restrictions', array(),