docs/export-0.9.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         Version 0.9 adds the database name to the site information.
  32
  33         The canonical URL to the schema document is:
  34         http://www.mediawiki.org/xml/export-0.9.xsd
  35
  36         Use the namespace:
  37         http://www.mediawiki.org/xml/export-0.9/
  38 -->
  39 <schema xmlns="http://www.w3.org/2001/XMLSchema"
  40                 xmlns:mw="http://www.mediawiki.org/xml/export-0.9/"
  41                 targetNamespace="http://www.mediawiki.org/xml/export-0.9/"
  42                 elementFormDefault="qualified">
  43
  44         <annotation>
  45                 <documentation xml:lang="en">
  46                         MediaWiki's page export format
  47                 </documentation>
  48         </annotation>
  49
  50         <!-- Need this to reference xml:lang -->
  51         <import namespace="http://www.w3.org/XML/1998/namespace"
  52                         schemaLocation="http://www.w3.org/2001/xml.xsd" />
  53
  54         <!-- Our root element -->
  55         <element name="mediawiki" type="mw:MediaWikiType">
  56                 <!-- Page ID contraint, see bug 4220 -->
  57                 <unique name="PageIDConstraint">
  58                         <selector xpath="mw:page" />
  59                         <field xpath="mw:id" />
  60                 </unique>
  61                 <!-- Revision ID contraint, see bug 4220 -->
  62                 <unique name="RevIDConstraint">
  63                         <selector xpath="mw:page/mw:revision" />
  64                         <field xpath="mw:id" />
  65                 </unique>
  66         </element>
  67
  68         <complexType name="MediaWikiType">
  69                 <sequence>
  70                         <element name="siteinfo" type="mw:SiteInfoType"
  71                                          minOccurs="0" maxOccurs="1" />
  72                         <element name="page" type="mw:PageType"
  73                                          minOccurs="0" maxOccurs="unbounded" />
  74                         <element name="logitem" type="mw:LogItemType"
  75                                          minOccurs="0" maxOccurs="unbounded" />
  76                 </sequence>
  77                 <attribute name="version" type="string" use="required" />
  78                 <attribute ref="xml:lang" use="required" />
  79         </complexType>
  80
  81         <complexType name="SiteInfoType">
  82                 <sequence>
  83                         <element name="sitename" type="string" minOccurs="0" />
  84             <element name="dbname" type="string" minOccurs="0" />
  85                         <element name="base" type="anyURI" minOccurs="0" />
  86                         <element name="generator" type="string" minOccurs="0" />
  87                         <element name="case" type="mw:CaseType" minOccurs="0" />
  88                         <element name="namespaces" type="mw:NamespacesType" minOccurs="0" />
  89                 </sequence>
  90         </complexType>
  91
  92         <simpleType name="CaseType">
  93                 <restriction base="NMTOKEN">
  94                         <!-- Cannot have two titles differing only by case of first letter. -->
  95                         <!-- Default behavior through 1.5, $wgCapitalLinks = true -->
  96                         <enumeration value="first-letter" />
  97
  98                         <!-- Complete title is case-sensitive -->
  99                         <!-- Behavior when $wgCapitalLinks = false -->
 100                         <enumeration value="case-sensitive" />
 101
 102                         <!-- Cannot have non-case senstitive titles eg [[FOO]] == [[Foo]] -->
 103                         <!-- Not yet implemented as of MediaWiki 1.18 -->
 104                         <enumeration value="case-insensitive" />
 105                 </restriction>
 106         </simpleType>
 107
 108         <simpleType name="DeletedFlagType">
 109                 <restriction base="NMTOKEN">
 110                         <enumeration value="deleted" />
 111                 </restriction>
 112         </simpleType>
 113
 114         <complexType name="NamespacesType">
 115                 <sequence>
 116                         <element name="namespace" type="mw:NamespaceType"
 117                                          minOccurs="0" maxOccurs="unbounded" />
 118                 </sequence>
 119         </complexType>
 120
 121         <complexType name="NamespaceType">
 122                 <simpleContent>
 123                         <extension base="string">
 124                                 <attribute name="key" type="integer" />
 125                                 <attribute name="case" type="mw:CaseType" />
 126                         </extension>
 127                 </simpleContent>
 128         </complexType>
 129
 130         <complexType name="RedirectType">
 131                 <simpleContent>
 132                         <extension base="string">
 133                                 <attribute name="title" type="string" />
 134                         </extension>
 135                 </simpleContent>
 136         </complexType>
 137
 138         <simpleType name="ContentModelType">
 139                 <restriction base="string">
 140                         <pattern value="[a-zA-Z][-+./a-zA-Z0-9]*" />
 141                 </restriction>
 142         </simpleType>
 143
 144         <simpleType name="ContentFormatType">
 145                 <restriction base="string">
 146                         <pattern value="[a-zA-Z][-+.a-zA-Z0-9]*/[a-zA-Z][-+.a-zA-Z0-9]*" />
 147                 </restriction>
 148         </simpleType>
 149
 150         <complexType name="PageType">
 151                 <sequence>
 152                         <!-- Title in text form. (Using spaces, not underscores; with namespace ) -->
 153                         <element name="title" type="string" />
 154
 155                         <!-- Namespace in canonical form -->
 156                         <element name="ns" type="nonNegativeInteger" />
 157
 158                         <!-- optional page ID number -->
 159                         <element name="id" type="positiveInteger" />
 160
 161                         <!-- flag if the current revision is a redirect -->
 162                         <element name="redirect" type="mw:RedirectType" minOccurs="0" maxOccurs="1" />
 163
 164                         <!-- comma-separated list of string tokens, if present -->
 165                         <element name="restrictions" type="string" minOccurs="0" />
 166
 167                         <!-- Zero or more sets of revision or upload data -->
 168                         <choice minOccurs="0" maxOccurs="unbounded">
 169                                 <element name="revision" type="mw:RevisionType" />
 170                                 <element name="upload" type="mw:UploadType" />
 171                         </choice>
 172
 173                         <!-- Zero or One sets of discussion threading data -->
 174                         <element name="discussionthreadinginfo" minOccurs="0" maxOccurs="1" type="mw:DiscussionThreadingInfo" />
 175                 </sequence>
 176         </complexType>
 177
 178         <complexType name="RevisionType">
 179                 <sequence>
 180                         <element name="id" type="positiveInteger" />
 181                         <element name="parentid" type="positiveInteger" minOccurs="0" />
 182                         <element name="timestamp" type="dateTime" />
 183                         <element name="contributor" type="mw:ContributorType" />
 184                         <element name="minor" minOccurs="0" maxOccurs="1" />
 185                         <element name="comment" type="mw:CommentType" minOccurs="0" maxOccurs="1" />
 186                         <element name="text" type="mw:TextType" />
 187                         <element name="sha1" type="string" />
 188                         <element name="model" type="mw:ContentModelType" />
 189                         <element name="format" type="mw:ContentFormatType" />
 190                 </sequence>
 191         </complexType>
 192
 193         <complexType name="LogItemType">
 194                 <sequence>
 195                         <element name="id" type="positiveInteger" />
 196                         <element name="timestamp" type="dateTime" />
 197                         <element name="contributor" type="mw:ContributorType" />
 198                         <element name="comment" type="mw:CommentType" minOccurs="0" />
 199                         <element name="type" type="string" />
 200                         <element name="action" type="string" />
 201                         <element name="text" type="mw:LogTextType" minOccurs="0" maxOccurs="1" />
 202                         <element name="logtitle" type="string" minOccurs="0" maxOccurs="1" />
 203                         <element name="params" type="mw:LogParamsType" minOccurs="0" maxOccurs="1" />
 204                 </sequence>
 205         </complexType>
 206
 207         <complexType name="CommentType">
 208                 <simpleContent>
 209                         <extension base="string">
 210                                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 211                                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 212                         </extension>
 213                 </simpleContent>
 214         </complexType>
 215
 216         <complexType name="TextType">
 217                 <simpleContent>
 218                         <extension base="string">
 219                                 <attribute ref="xml:space" use="optional" default="preserve" />
 220                                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 221                                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 222                                 <!-- This isn't a good idea; we should be using "ID" instead of "NMTOKEN" -->
 223                                 <!-- However, "NMTOKEN" is strictest definition that is both compatible with existing -->
 224                                 <!-- usage ([0-9]+) and with the "ID" type. -->
 225                                 <attribute name="id" type="NMTOKEN" />
 226                                 <attribute name="bytes" use="optional" type="nonNegativeInteger" />
 227                         </extension>
 228                 </simpleContent>
 229         </complexType>
 230
 231         <complexType name="LogTextType">
 232                 <simpleContent>
 233                         <extension base="string">
 234                                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 235                                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 236                         </extension>
 237                 </simpleContent>
 238         </complexType>
 239
 240         <complexType name="LogParamsType">
 241                 <simpleContent>
 242                         <extension base="string">
 243                                 <attribute ref="xml:space" use="optional" default="preserve" />
 244                         </extension>
 245                 </simpleContent>
 246         </complexType>
 247
 248         <complexType name="ContributorType">
 249                 <sequence>
 250                         <element name="username" type="string" minOccurs="0" />
 251                         <element name="id" type="nonNegativeInteger" minOccurs="0" />
 252
 253                         <element name="ip" type="string" minOccurs="0" />
 254                 </sequence>
 255                 <!-- This allows deleted=deleted on non-empty elements, but XSD is not omnipotent -->
 256                 <attribute name="deleted" use="optional" type="mw:DeletedFlagType" />
 257         </complexType>
 258
 259         <complexType name="UploadType">
 260                 <sequence>
 261                         <!-- Revision-style data... -->
 262                         <element name="timestamp" type="dateTime" />
 263                         <element name="contributor" type="mw:ContributorType" />
 264                         <element name="comment" type="string" minOccurs="0" />
 265
 266                         <!-- Filename. (Using underscores, not spaces. No 'File:' namespace marker.) -->
 267                         <element name="filename" type="string" />
 268
 269                         <!-- URI at which this resource can be obtained -->
 270                         <element name="src" type="anyURI" />
 271
 272                         <element name="size" type="positiveInteger" />
 273
 274                         <!-- TODO: add other metadata fields -->
 275                 </sequence>
 276         </complexType>
 277
 278         <!-- Discussion threading data for LiquidThreads -->
 279         <complexType name="DiscussionThreadingInfo">
 280                 <sequence>
 281                         <element name="ThreadSubject" type="string" />
 282                         <element name="ThreadParent" type="positiveInteger" />
 283                         <element name="ThreadAncestor" type="positiveInteger" />
 284                         <element name="ThreadPage" type="string" />
 285                         <element name="ThreadID" type="positiveInteger" />
 286                         <element name="ThreadAuthor" type="string" />
 287                         <element name="ThreadEditStatus" type="string" />
 288                         <element name="ThreadType" type="string" />
 289                 </sequence>
 290         </complexType>
 291
 292 </schema>