docs/export-0.8.xsd

   1 <?xml version="1.0" encoding="UTF-8" ?>
   2 <!--
   3         This is an XML Schema description of the format
   4         output by MediaWiki's Special:Export system.
   5
   6         Version 0.2 adds optional basic file upload info support,
   7         which is used by our OAI export/import submodule.
   8
   9         Version 0.3 adds some site configuration information such
  10         as a list of defined namespaces.
  11
  12         Version 0.4 adds per-revision delete flags, log exports,
  13         discussion threading data, a per-page redirect flag, and
  14         per-namespace capitalization.
  15
  16         Version 0.5 adds byte count per revision.
  17
  18         Version 0.6 adds a separate namespace tag, and resolves the
  19         redirect target and adds a separate sha1 tag for each revision.
  20
  21         Version 0.7 adds a unique identity constraint for both page and
  22         revision identifiers. See also bug 4220.
  23         Fix type for <ns> from "positiveInteger" to "nonNegativeInteger" to allow 0
  24         Moves <logitem> to its right location.
  25         Add parentid to revision.
  26         Fix type for <id> within <contributor> to "nonNegativeInteger"
  27
  28         Version 0.8 adds support for a <model> and a <format> tag for
  29         each revision. See contenthandler.txt.
  30
  31         The canonical URL to the schema document is:
  32         http://www.mediawiki.org/xml/export-0.8.xsd
  33
  34         Use the namespace:
  35         http://www.mediawiki.org/xml/export-0.8/
  36 -->
  37 <schema xmlns="http://www.w3.org/2001/XMLSchema"
  38                 xmlns:mw="http://www.mediawiki.org/xml/export-0.8/"
  39                 targetNamespace="http://www.mediawiki.org/xml/export-0.8/"
  40                 elementFormDefault="qualified">
  41
  42         <annotation>
  43                 <documentation xml:lang="en">
  44                         MediaWiki's page export format
  45                 </documentation>
  46         </annotation>
  47
  48         <!-- Need this to reference xml:lang -->
  49         <import namespace="http://www.w3.org/XML/1998/namespace"
  50                         schemaLocation="http://www.w3.org/2001/xml.xsd" />
  51
  52         <!-- Our root element -->
  53         <element name="mediawiki" type="mw:MediaWikiType">
  54                 <!-- Page ID contraint, see bug 4220 -->
  55                 <unique name="PageIDConstraint">
  56                         <selector xpath="mw:page" />
  57                         <field xpath="mw:id" />
  58                 </unique>
  59                 <!-- Revision ID contraint, see bug 4220 -->
  60                 <unique name="RevIDConstraint">
  61                         <selector xpath="mw:page/mw:revision" />
  62                         <field xpath="mw:id" />
  63                 </unique>
  64         </element>
  65
  66         <complexType name="MediaWikiType">
  67                 <sequence>
  68                         <element name="siteinfo" type="mw:SiteInfoType"
  69                                          minOccurs="0" maxOccurs="1" />
  70                         <element name="page" type="mw:PageType"
  71                                          minOccurs="0" maxOccurs="unbounded" />
  72                         <element name="logitem" type="mw:LogItemType"
  73                                          minOccurs="0" maxOccurs="unbounded" />
  74                 </sequence>
  75                 <attribute name="version" type="string" use="required" />
  76                 <attribute ref="xml:lang" use="required" />
  77         </complexType>
  78
  79         <complexType name="SiteInfoType">
  80                 <sequence>
  81                         <element name="sitename" type="string" minOccurs="0" />
  82                         <element name="base" type="anyURI" minOccurs="0" />
  83                         <element name="generator" type="string" minOccurs="0" />
  84                         <element name="case" type="mw:CaseType" minOccurs="0" />
  85                         <element name="namespaces" type="mw:NamespacesType" minOccurs="0" />
  86                 </sequence>
  87         </complexType>
  88
  89         <simpleType name="CaseType">
  90                 <restriction base="NMTOKEN">
  91                         <!-- Cannot have two titles differing only by case of first letter. -->
  92                         <!-- Default behavior through 1.5, $wgCapitalLinks = true -->
  93                         <enumeration value="first-letter" />
  94
  95                         <!-- Complete title is case-sensitive -->
  96                         <!-- Behavior when $wgCapitalLinks = false -->
  97                         <enumeration value="case-sensitive" />
  98
  99                         <!-- Cannot have non-case senstitive titles eg [[FOO]] == [[Foo]] -->
 100                         <!-- Not yet implemented as of MediaWiki 1.18 -->
 101                         <enumeration value="case-insensitive" />
 102                 </restriction>
 103         </simpleType>
 104
 105         <simpleType name="DeletedFlagType">
 106                 <restriction base="NMTOKEN">
 107                         <enumeration value="deleted" />
 108                 </restriction>
 109         </simpleType>
 110
 111         <complexType name="NamespacesType">
 112                 <sequence>
 113                         <element name="namespace" type="mw:NamespaceType"
 114                                          minOccurs="0" maxOccurs="unbounded" />
 115                 </sequence>
 116         </complexType>
 117
 118         <complexType name="NamespaceType">
 119                 <simpleContent>
 120                         <extension base="string">
 121                                 <attribute name="key" type="integer" />
 122                                 <attribute name="case" type="mw:CaseType" />
 123                         </extension>
 124                 </simpleContent>
 125         </complexType>
 126
 127         <complexType name="RedirectType">
 128                 <simpleContent>
 129                         <extension base="string">
 130                                 <attribute name="title" type="string" />
 131                         </extension>
 132                 </simpleContent>
 133         </complexType>
 134
 135         <simpleType name="ContentModelType">
 136                 <restriction base="string">
 137                         <pattern value="[a-zA-Z][-+./a-zA-Z0-9]*" />
 138                 </restriction>
 139         </simpleType>
 140
 141         <simpleType name="ContentFormatType">
 142                 <restriction base="string">
 143                         <pattern value="[a-zA-Z][-+.a-zA-Z0-9]*/[a-zA-Z][-+.a-zA-Z0-9]*" />
 144                 </restriction>
 145         </simpleType>
 146
 147         <complexType name="PageType">
 148                 <sequence>
 149                         <!-- Title in text form. (Using spaces, not underscores; with namespace ) -->
 150                         <element name="title" type="string" />
 151
 152                         <!-- Namespace in canonical form -->
 153                         <element name="ns" type="nonNegativeInteger" />
 154
 155                         <!-- optional page ID number -->
 156                         <element name="id" type="positiveInteger" />
 157
 158                         <!-- flag if the current revision is a redirect -->
 159                         <element name="redirect" type="mw:RedirectType" minOccurs="0" maxOccurs="1" />
 160
 161                         <!-- comma-separated list of string tokens, if present -->
 162                         <element name="restrictions" type="string" minOccurs="0" />
 163
 164                         <!-- Zero or more sets of revision or upload data -->
 165                         <choice minOccurs="0" maxOccurs="unbounded">
 166                                 <element name="revision" type="mw:RevisionType" />
 167                                 <element name="upload" type="mw:UploadType" />
 168                         </choice>
 169
 170                         <!-- Zero or One sets of discussion threading data -->
 171                         <element name="discussionthreadinginfo" minOccurs="0" maxOccurs="1" type="mw:DiscussionThreadingInfo" />
 172                 </sequence>
 173         </complexType>
 174
 175         <complexType name="RevisionType">
 176                 <sequence>
 177                         <element name="id" type="positiveInteger" />
 178                         <element name="parentid" type="positiveInteger" minOccurs="0" />
 179                         <element name="timestamp" type="dateTime" />
 180                         <element name="contributor" type="mw:ContributorType" />
 181                         <element name="minor" minOccurs="0" maxOccurs="1" />
 182                         <element name="comment" type="mw:CommentType" minOccurs="0" maxOccurs="1" />
 183                         <element name="text" type="mw:TextType" />
 184                         <element name="sha1" type="string" />
 185                         <element name="model" type="mw:ContentModelType" />
 186                         <element name="format" type="mw:ContentFormatType" />
 187                 </sequence>
 188         </complexType>
 189
 190         <complexType name="LogItemType">
 191                 <sequence>
 192                         <element name="id" type="positiveInteger" />
 193                         <element name="timestamp" type="dateTime" />
 194                         <element name="contributor" type="mw:ContributorType" />
 195                         <element name="comment" type="mw:CommentType" minOccurs="0" />
 196                         <element name="type" type="string" />
 197                         <element name="action" type="string" />
 198                         <element name="text" type="mw:LogTextType" minOccurs="0" maxOccurs="1" />
 199                         <element name="logtitle" type="string" minOccurs="0" maxOccurs="1" />
 200                         <element name="params" type="mw:LogParamsType" minOccurs="0" maxOccurs="1" />
 201                 </sequence>
 202         </complexType>
 203
 204         <complexType name="CommentType">
 205                 <simpleContent>
 206                         <extension base="string">
 207                                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 208                                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 209                         </extension>
 210                 </simpleContent>
 211         </complexType>
 212
 213         <complexType name="TextType">
 214                 <simpleContent>
 215                         <extension base="string">
 216                                 <attribute ref="xml:space" use="optional" default="preserve" />
 217                                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 218                                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 219                                 <!-- This isn't a good idea; we should be using "ID" instead of "NMTOKEN" -->
 220                                 <!-- However, "NMTOKEN" is strictest definition that is both compatible with existing -->
 221                                 <!-- usage ([0-9]+) and with the "ID" type. -->
 222                                 <attribute name="id" type="NMTOKEN" />
 223                                 <attribute name="bytes" use="optional" type="nonNegativeInteger" />
 224                         </extension>
 225                 </simpleContent>
 226         </complexType>
 227
 228         <complexType name="LogTextType">
 229                 <simpleContent>
 230                         <extension base="string">
 231                                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 232                                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 233                         </extension>
 234                 </simpleContent>
 235         </complexType>
 236
 237         <complexType name="LogParamsType">
 238                 <simpleContent>
 239                         <extension base="string">
 240                                 <attribute ref="xml:space" use="optional" default="preserve" />
 241                         </extension>
 242                 </simpleContent>
 243         </complexType>
 244
 245         <complexType name="ContributorType">
 246                 <sequence>
 247                         <element name="username" type="string" minOccurs="0" />
 248                         <element name="id" type="nonNegativeInteger" minOccurs="0" />
 249
 250                         <element name="ip" type="string" minOccurs="0" />
 251                 </sequence>
 252                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 253                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 254         </complexType>
 255
 256         <complexType name="UploadType">
 257                 <sequence>
 258                         <!-- Revision-style data... -->
 259                         <element name="timestamp" type="dateTime" />
 260                         <element name="contributor" type="mw:ContributorType" />
 261                         <element name="comment" type="string" minOccurs="0" />
 262
 263                         <!-- Filename. (Using underscores, not spaces. No 'File:' namespace marker.) -->
 264                         <element name="filename" type="string" />
 265
 266                         <!-- URI at which this resource can be obtained -->
 267                         <element name="src" type="anyURI" />
 268
 269                         <element name="size" type="positiveInteger" />
 270
 271                         <!-- TODO: add other metadata fields -->
 272                 </sequence>
 273         </complexType>
 274
 275         <!-- Discussion threading data for LiquidThreads -->
 276         <complexType name="DiscussionThreadingInfo">
 277                 <sequence>
 278                         <element name="ThreadSubject" type="string" />
 279                         <element name="ThreadParent" type="positiveInteger" />
 280                         <element name="ThreadAncestor" type="positiveInteger" />
 281                         <element name="ThreadPage" type="string" />
 282                         <element name="ThreadID" type="positiveInteger" />
 283                         <element name="ThreadAuthor" type="string" />
 284                         <element name="ThreadEditStatus" type="string" />
 285                         <element name="ThreadType" type="string" />
 286                 </sequence>
 287         </complexType>
 288
 289 </schema>