libis-format 0.9.5-java

data/tools/fido/conf/fido-formats.xsd

@@ -0,0 +1,173 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--
+Usage of DC has been based on these references:
+  http://dublincore.org/documents/usageguide/elements.shtml
+  http://dublincore.org/documents/usageguide/qualifiers.shtml
+  http://www.dublincore.org/documents/dc-xml-guidelines/
+-->
+<xs:schema elementFormDefault="qualified" 
+  xmlns:xs="http://www.w3.org/2001/XMLSchema"  
+  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+  xmlns:dc="http://purl.org/dc/elements/1.1/"
+  xmlns:dcterms="http://purl.org/dc/terms/">
+  <xs:import namespace="http://purl.org/dc/elements/1.1/" schemaLocation="dc.xsd"/>
+  <xs:import namespace="http://purl.org/dc/terms/" schemaLocation="dcterms.xsd"/>
+<!-- http://www.dublincore.org/schemas/xmls/qdc/2003/04/02/dcterms.xsd" -->
+  <xs:element name="formats">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element maxOccurs="unbounded" ref="format"/>
+      </xs:sequence>
+      <xs:attribute name="version" type="xs:decimal" use="required"/>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="format">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element maxOccurs="1" minOccurs="1" ref="puid"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" ref="mime"/>
+        <xs:element minOccurs="0" ref="container"/>
+        <xs:element maxOccurs="1" minOccurs="1" ref="name"/>
+        <xs:element maxOccurs="1" minOccurs="0" name="version" type="xs:string"/>
+        <xs:element minOccurs="0" name="alias"  type="xs:string"/>
+        <xs:element maxOccurs="1" minOccurs="0" ref="pronom_id"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" ref="extension"/>
+        <xs:element maxOccurs="1" minOccurs="0" name="apple_uti" type="xs:string"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" ref="has_priority_over"/>
+        <xs:element maxOccurs="unbounded" ref="signature"/>
+        <xs:element minOccurs="0" ref="note"/>
+        <xs:element maxOccurs="1" ref="details"/>
+      </xs:sequence>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="mime" type="xs:string"/>
+  <xs:element name="puid" type="xs:string"/>
+  <xs:element name="container" type="container_type"/>
+  <xs:element name="pronom_id" type="xs:integer"/>
+  <xs:element name="extension" type="xs:string"/>
+  <xs:element name="has_priority_over" type="xs:string"/>
+  <xs:element name="signature">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element ref="name"/>
+        <xs:element ref="note" minOccurs="0" maxOccurs="unbounded"/>
+        <xs:element maxOccurs="unbounded" ref="pattern"/>
+      </xs:sequence>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="note" type="xs:string"/>
+  <xs:element name="pattern">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element ref="position"/>
+        <xs:element ref="pronom_pattern" minOccurs="0"/>
+        <xs:element ref="regex" minOccurs="1" maxOccurs="1" />
+      </xs:sequence>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="position" type="position_type"/>
+  <xs:element name="pronom_pattern" type="xs:string"/>
+  <xs:element name="regex" type="xs:string"/>
+  <xs:element name="name" type="xs:string"/>
+  <xs:simpleType name="position_type">
+    <xs:restriction base="xs:string">
+      <xs:enumeration value="BOF"/>
+      <xs:enumeration value="EOF"/>
+      <xs:enumeration value="VAR"/>
+    </xs:restriction>
+  </xs:simpleType>
+  <xs:simpleType name="container_type">
+    <xs:restriction base="xs:string">
+      <xs:enumeration value="zip"/>
+      <xs:enumeration value="tar"/>
+    </xs:restriction>
+  </xs:simpleType>
+  <xs:element name="details">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element maxOccurs="1" minOccurs="0" ref="dc:description"/>
+        <xs:element maxOccurs="1" minOccurs="0" ref="dcterms:available"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" ref="dc:creator"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" ref="dcterms:publisher"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" name="is_supertype_of" type="xs:string"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" name="is_subtype_of" type="xs:string"/>
+        <xs:element maxOccurs="1" minOccurs="0" name="content_type" type="xs:string"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" ref="reference"/>
+        <xs:element maxOccurs="unbounded" minOccurs="0" ref="example_file"/>
+        <xs:element maxOccurs="1" minOccurs="0" ref="record_metadata"/>
+      </xs:sequence>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="reference">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element maxOccurs="1" minOccurs="0" name="type" type="xs:string"/>
+        <xs:element maxOccurs="unbounded" ref="dc:any"/>
+      </xs:sequence>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="example_file">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element maxOccurs="unbounded" ref="dc:any"/>
+        <xs:element minOccurs="1" ref="checksum"/>
+      </xs:sequence>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="record_metadata">
+    <xs:complexType>
+      <xs:sequence>
+        <xs:element maxOccurs="1" minOccurs="1" ref="status"/>
+        <xs:element maxOccurs="unbounded" ref="dc:any"/>
+      </xs:sequence>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="checksum">
+    <xs:complexType mixed="true">
+      <xs:sequence>
+      </xs:sequence>
+      <xs:attribute name="type" use="required">
+       <xs:simpleType>
+        <xs:restriction base="xs:string">
+          <xs:enumeration value="md5"/>
+          <xs:enumeration value="sha256"/>
+        </xs:restriction>
+       </xs:simpleType>
+      </xs:attribute>
+    </xs:complexType>
+  </xs:element>
+  <xs:element name="status">
+    <xs:simpleType>
+        <xs:restriction base="xs:string">
+          <xs:enumeration value="unknown"/>
+          <xs:enumeration value="invalid"/>
+          <xs:enumeration value="incomplete"/>
+          <xs:enumeration value="stub"/>
+          <xs:enumeration value="adequate"/>
+          <xs:enumeration value="complete"/>
+          <xs:enumeration value="superb"/>
+          <xs:enumeration value="deprecated"/>
+        </xs:restriction>
+    </xs:simpleType>
+  </xs:element>
+  <xs:element name="content_type">
+    <xs:simpleType>
+        <xs:restriction base="xs:string">
+          <xs:enumeration value="raster"/>
+          <xs:enumeration value="vector"/>
+          <xs:enumeration value="image"/>
+          <xs:enumeration value="spreadsheet"/>
+          <xs:enumeration value="text"/>
+          <xs:enumeration value="word-processed"/>
+          <xs:enumeration value="page-description"/>
+          <xs:enumeration value="audio"/>
+          <xs:enumeration value="presentation"/>
+          <xs:enumeration value="mark-up"/>
+          <xs:enumeration value="database"/>
+          <xs:enumeration value="video"/>
+          <xs:enumeration value="email"/>
+          <xs:enumeration value="other"/>
+        </xs:restriction>
+    </xs:simpleType>
+  </xs:element>
+</xs:schema>

data/tools/fido/conf/format_extension_template.xml

@@ -0,0 +1,105 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<formats version="0.3" 
+  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+  xsi:noNamespaceSchemaLocation="fido-formats.xsd"
+  xmlns:dc="http://purl.org/dc/elements/1.1/"
+  xmlns:dcterms="http://purl.org/dc/terms/">
+  <format>
+	<!-- Unique identifier for this record. If an identifier is re-used, 
+	the later record takes precedence. This allows existing records to be overridden if required. -->
+    <puid>fido-fmt/189.word</puid>
+    <name>Microsoft Office Open XML - Word</name>
+    <version>1.0 Transitional</version>
+    <alias>Word OOXML</alias>
+    <extension>docx</extension>
+    <apple_uid/>
+    <!-- These are lower-priority signatures, identified by unique id. If a match is found, this record will be matched first, before those specified here. -->
+    <has_priority_over>x-fmt/263</has_priority_over>
+    <has_priority_over>fmt/189</has_priority_over>
+    <signature>
+      <name>Microsoft Office Open XML - Word</name>
+      <!-- Each signature is composed of one or more patterns, which must all match for an identification to be successful. 
+      BOF regex -->
+      <pattern>
+        <position>BOF</position>
+        <regex>(?s)\APK\x03\x04</regex>
+      </pattern>
+      <pattern>
+        <position>BOF</position>
+        <regex>(?s)\A.{30}\[Content_Types\]\.xml \xa2</regex>
+      </pattern>
+      <pattern>
+        <position>EOF</position>
+        <regex>(?s)\x00\x00word/.{1,20}\.xmlPK\x01\x02\x2d.{0,2000}\Z</regex>
+      </pattern>
+    </signature>
+      
+	<!-- These above information provide the core fido fields, i.e. those relating to identification. -->	
+	<!-- More detailed information beyond that required for identification goes in the details section. -->
+	<details>
+	  <!-- A description of the format -->
+      <dc:description></dc:description>
+      <!-- The release date -->
+      <dcterms:available/>
+      <!-- Who created the format. -->
+      <dc:creator>Microsoft</dc:creator>
+      <!-- Who published the format. -->
+      <dcterms:publisher />
+      <!-- Which formats (identified by puid) are super-types of this one.
+      e.g. all instances of the current type are also instances of this broader super-type. -->
+      <is_subtype_of/>
+      <!-- Which formats are sub-types to this one. -->
+      <is_supertype_of/>
+	  <!-- Content Type, free text, but should be one of:
+	raster|vector|image|spreadsheet|text|word-processed|page-description|audio|presentation|mark-up|database|video|email|other
+	  -->
+      <content_type>word-processed</content_type>
+      
+	  <!-- References to useful documentation or other resources. 
+	       You can use any Dublin Core or DCTerms elements here. -->
+      <reference>
+        <!-- The type of reference this is. Should be one of: 
+             information|specification|implementation|registry|related
+             -->
+      	<type>specification</type>
+        <dc:title/>
+        <dc:creator/>
+        <dc:publisher/>
+        <!-- If the specifcation is available at a URL, this should be included as an identifier. -->
+		<dc:identifier/>
+        <dc:description/>
+        <dcterms:created/>
+        <dcterms:modified/>
+        <dcterms:available/>
+        <dc:type>Authoritative</dc:type>
+        <dcterms:license/>
+      </reference>
+      
+	  <!-- Example files that should be successfully identified by the above signature. 
+           You can use any Dublin Core or DCTerms elements here. -->
+      <example_file>
+        <dc:identifier>./this.txt</dc:identifier>
+        <dc:rights>To the extent possible under law, Andrew Jackson has waived all copyright and related or neighbouring rights to this example file.</dc:rights>
+        <dcterms:license>http://creativecommons.org/publicdomain/zero/1.0/</dcterms:license>
+        <checksum type="md5">XXXXXX</checksum>
+      </example_file>
+      
+	  <!-- Metadata about this format record.
+           You can use any Dublin Core or DCTerms elements here. -->
+      <record_metadata>
+      	<!-- invalid|incomplete|stub|adequate|complete|superb|deprecated -->
+      	<status>complete</status>
+        <dc:creator>Your name goes here.</dc:creator>
+        <dc:description/>
+        <dcterms:license/>
+        <dcterms:created>11 Mar 2005</dcterms:created>
+        <dcterms:modified>02 Aug 2005</dcterms:modified>
+		<!-- If drawn from a reference, re-use the identifier from the references section. -->
+        <dc:source/>
+        <!-- Deprecated records should declare the new puid(s) here: -->
+        <dcterms:isReplacedBy/>
+      </record_metadata>
+    </details>
+    
+  </format>
+</formats>