JP2005352945A - Style sheet creation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a style sheet creation device for precisely and automatically performing a conversion between structured documents. <P>SOLUTION: This style sheet creation device comprises a schema file analysis means 202 analyzing first and second schema files describing, for first and second structured documents, the structures thereof, respectively; a correspondence file analysis means 201 reading and analyzing a correlation file describing the associations of items between the first and second structured documents; a conversion pattern determination means 203 determining a conversion pattern from the first structured document to the second structured document based on the analysis result of the correspondence file and the analysis result of the schema files; and a style sheet output means outputting, based on the conversion pattern, a style sheet for performing the conversion of data structure between different structured documents of the first and second structured document. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a style sheet creation apparatus for converting a data structure between structured documents having different data structures.

  In recent years, with the spread of the Internet and digital broadcasting, the number of contents has increased explosively. Among these many contents, in order to easily grasp and search the contents of desired contents, cases of adding various metadata to the contents are increasing. The metadata is described in XML (eXtensible Markup Language), which is a meta language, in consideration of data transmission / reception between computers and affinity with the Internet. The data structure of description such as XML is formulated by various organizations. Further, the data structure of the meta language differs depending on the type of content to which the metadata is added and the purpose for using the metadata. Therefore, even if an XML document based on metadata is created in conformity with the content, if the content is used for other purposes, it is necessary to create a new XML document having a data structure suitable for the other purposes. is there. Since it is effective to create a newly created XML document using an already created XML document, the data structure of the XML document is converted. Here, when the data structure between XML documents is converted, XSLT (XML Stylesheet Language Transformations) conversion is performed on the XML document. At this time, an XSLT file is created as a style sheet that specifies what conversion is to be performed.

  In Patent Document 1, a component for converting a structured document is presented to a metadata producer using a GUI (Graphical User Interface). Here, the component is a component of the form description describing the conversion rule. The form description is a description of control rules for converting a structured document into a format that is displayed in a format that is arranged on a display device. When the creator selects a desired part from the presented parts, the parts are combined to form an XSLT file. Therefore, even a person who does not have knowledge about XSLT can easily create an XSLT file.

Moreover, in patent document 2, all the XML schemas of the XML document to be converted are converted into an ontology model. Then, the correspondence of items between XML documents is extracted by the converted ontology model, and an XSLT file is automatically created.
JP 2003-303183 A JP 2003-233528 A

  However, in Japanese Patent Laid-Open No. 2004-228561, a creator who converts an XML document must manually select parts necessary for converting the XML document. Here, since the conversion of the XML document is a conversion that performs XML structure conversion, it is necessary to consider the correspondence between components between XML documents, that is, items to be described later. That is, it is necessary to select a necessary component in consideration of whether the item is, for example, “element”, “attribute”, “element value”, or “attribute value”. Therefore, the XSLT file cannot be automatically created, and it takes time to create the XSLT file.

  In Patent Document 2, it is necessary to convert all XML schemas of an XML document to be converted into an ontology model when extracting a tag relationship between XML documents using an ontology model. Therefore, when an XML schema language that does not describe a semantic structure is converted into an ontology that describes a semantic structure, problems such as missing information in items and changes in the meaning of information occur.

  SUMMARY An advantage of some aspects of the invention is that it provides a style sheet creation apparatus for accurately and automatically converting between structured documents.

  A first invention of the present application is a style sheet creation device for converting a data structure between a first and a second structured document having different data structures in order to solve the above-mentioned problems. Schema file analysis means for analyzing the first and second schema files describing the structure of each of the two structured documents, and the first and second structured documents for items which are information units included in each structured document A correspondence file analysis unit that reads the correspondence file describing the association between the items and analyzes the correspondence relationship of the items, and the first structure based on the analysis result of the correspondence file and the analysis result of the schema file Conversion pattern determining means for determining a conversion pattern from the structured document to the second structured document, and based on the conversion pattern, Providing a style sheet producing apparatus which comprises a style sheet output means for outputting a style sheet for converting the data structure between serial different structured document of the first and second structured document.

  Here, the item is, for example, a tag for clearly indicating an element position in a structured document and storing an attribute. As described above, the items can be automatically and efficiently extracted from the schema file by using the correspondence file. Also, the correspondence pattern between the first structured document and the second structured document is analyzed by analyzing the correspondence between items by the correspondence file analyzing means and analyzing the hierarchical structure and type between items by the schema file analyzing means. Can be determined for each item. For example, when there are two types of items of the structured document, “element” and “attribute”, “element” → “element”, “element” → “element” → for each item to be converted in the first structured document. A conversion pattern such as “attribute”, “attribute” → “element” can be determined. Therefore, a style sheet can be created automatically, efficiently and easily based on the determined conversion pattern.

  The style sheet is created based on the analysis result of the correspondence file and the schema file without conversion from the schema file to the model of the correspondence file. Therefore, it is possible to prevent a lack of item information or a change in the meaning of information associated with the conversion.

  According to a second aspect of the present invention, in the first aspect, the schema file analysis unit includes the first and second items obtained by analyzing the correspondence relationship between the first and second structured documents in the correspondence file analysis unit. 2. A style sheet creation apparatus characterized by extracting information from an extracted item from a schema file and analyzing the extracted item information.

  The item analyzed by the correspondence file analysis unit based on the correspondence file is an item to be converted. Of the items in the first and second schema files, only the items to be converted can be efficiently created by analyzing the hierarchical structure and type of items.

  A third invention of the present application provides the style sheet creating apparatus according to the second invention, wherein the item information is a type of the item and / or a hierarchical structure of the item.

  By analyzing the hierarchical structure of items, an accurate data structure can be converted between structured documents having different hierarchical structures. Also, by analyzing the types of items, it is possible to convert the data structure accurately even when the handling of items differs between structured documents.

  A fourth invention of the present application provides the style sheet creating apparatus according to the third invention, wherein the type of the item includes an element, an attribute, an element value, and / or an attribute value.

  A fifth invention of the present application provides the style sheet creating apparatus according to the first invention, wherein the correspondence file is an ontology file described by OWL (Web Ontology Language).

  By using the ontology file as the correspondence file describing the correspondence between items, it is possible to efficiently retrieve the correspondence between items between structured documents. Further, by using an ontology file used on the Internet, information such as an HTML document can be easily converted into metadata of video data.

  According to a sixth aspect of the present invention, in the first aspect, the schema file analyzing means includes the first and second items obtained by analyzing the correspondence between the first and second structured documents in the correspondence file analyzing means. 2. Extracting from a schema file and analyzing the information, the conversion pattern determining means determines a conversion pattern for each item by replacing the correspondence of the extracted item with the correspondence of the item information. A style sheet creating apparatus is provided.

  The correspondence relationship file analysis unit analyzes, for example, that item 1 of the first structured document is equal to item 2 of the second structured document (item 1 → item 2). Further, the schema file analysis means specifies, for example, the type of item as the analysis result of the schema file. Here, the item type is the content of a tag such as “element”, “attribute”, “element value”, or “attribute value”. For example, when the type of item 1 is “element” and the type of item 2 is “attribute”, the conversion pattern is calculated based on the correspondence (item 1 → item 2) analyzed from the correspondence file and the type of each item. The determining means determines the conversion pattern of “element” → “attribute”. Thus, the conversion pattern is automatically and efficiently determined based on the analysis result of the correspondence file and the schema file. Therefore, a style sheet for converting the data structure between the first structured document and the second structured document can be created automatically, efficiently and easily based on this conversion pattern. The style sheet is created based on the analysis result of the correspondence file and the schema file without conversion from the schema file to the model of the correspondence file. Therefore, it is possible to prevent a lack of item information or a change in the meaning of information associated with the conversion.

  A seventh invention of the present application provides the style sheet creation device according to the first invention, wherein the conversion pattern determining means further determines a conversion pattern from the second structured document to the first structured document. To do.

  In addition to the style sheet for converting from the first structured document to the second structured document, a style sheet for converting from the second structured document to the first structured document can be created.

  According to an eighth aspect of the present invention, in the first aspect, the style sheet output means includes a template that defines a conversion pattern from the item of the first structured document to the item of the second structured document in a predetermined style sheet language. The conversion pattern storage means stored for each conversion pattern and the template corresponding to each conversion pattern determined by the conversion pattern determination means are extracted from the conversion pattern storage means, and the style sheet is extracted based on the extracted template. There is provided a style sheet creating apparatus comprising template description means for describing the conversion pattern in the style sheet language.

  Using a template corresponding to the conversion pattern from the first structured document item to the second structured document item, a style sheet can be easily created in a style sheet language.

  A ninth invention of the present application provides the style sheet creating apparatus according to the eighth invention, wherein the conversion pattern storage means stores a template defining a conversion pattern of item information.

  It is possible to create a style sheet that can accurately convert the data structure even when item information differs between structured documents. For example, even when the hierarchical structure is different between structured documents or when the types of structures are different, an accurate data structure can be converted.

  In a tenth aspect of the present invention, in the first aspect, the structured document is an XML (eXtensible Markup Language) document, the schema file is an XML schema file, and the style sheet is an XSLT (XML Stylesheet Language Transformations) file. Provided is a style sheet creating apparatus characterized by the above.

  The eleventh invention of the present application is a style sheet creation method for converting the data structure between the first and second structured documents having different data structures, and the structure of each of the first and second structured documents. Schema file analysis step for analyzing the described first and second schema files, and the correspondence relationship describing the association between the first and second structured documents with respect to items which are information units included in each structured document Based on the correspondence file analysis step of reading the file and analyzing the correspondence of the items, and the analysis result of the correspondence file and the analysis result of the schema file, the first structured document to the second structured document A conversion pattern determining step for determining a conversion pattern into the first and second based on the conversion pattern Providing a style sheet creation method characterized by including a style sheet output step of outputting a style sheet for converting the data structure in different structured documents of the structured document.

  By using a correspondence file describing the correspondence between items between structured documents, a style sheet can be created automatically, efficiently and easily. The style sheet is created based on the analysis result of the correspondence file and the schema file without conversion from the schema file to the model of the correspondence file. Therefore, it is possible to prevent a lack of item information or a change in the meaning of information associated with the conversion.

<Summary of invention>
Hereinafter, the meta language is a language for describing the logical structure of data. Examples of the meta language include SGML (Standard Generalized Markup Language) and HTML (HyperText Markup Language) in addition to the above-described XML. A structured document is a document created using a meta language, and examples include an XML document, an SGML document, and HTML. A schema file is a file in which a structure that can be taken by a structured document is described in a schema language. For example, an XML schema file is used. Note that the file is not limited to an XML schema file as long as it can read information on items and a hierarchical structure to be described later. For example, an RDF (Resource Description Framework) file or a CSV (Comma Separated Value) file can be used. A style sheet describes a conversion rule for converting a data structure between structured documents having different data structures, and includes, for example, an XML style sheet. A style sheet language of the style sheet includes an XSLT (XML Stylesheet Language Transformations) file.

  The correspondence file is a file that describes the correspondence between items between structured documents, and examples thereof include synonym dictionary files such as a thesaurus and an ontology. The item is, for example, a tag for clearly indicating an element position in a structured document and storing an attribute. The item has information such as the type of item and the hierarchical structure of the item. The item type is the content of a tag such as “element”, “attribute”, “element value”, or “attribute value”.

  An outline of a method of creating a style sheet for converting the data structure between the structured document A and the structured document B having different data structures will be described. First, a correspondence file describing the association of items between the structured document A and the structured document B is read, and the correspondence between items is analyzed. For example, it is analyzed from the correspondence file that the item IA of the structured document A and the item IB of the structured document B are equal (item IA → item IB).

  Next, items are extracted from each of the schema file SA of the structured document A and the schema file SB of the structured document B, and the hierarchical structure and type between the items are analyzed. The item extraction may be performed for the item IA and the item IB whose correspondence is analyzed in the correspondence file. For example, the item IA is extracted from the schema file SA, and the item IB is extracted from the schema file SB. Then, the information of the extracted item IA and item IB is analyzed. As the item information, for example, it is analyzed that the type of the item IA is “element” and the type of the item IB is “attribute”.

  A conversion pattern from the structured document A to the structured document B is determined on the basis of the correspondence relationship between items analyzed from the correspondence file (item IA → item IB) and the types of the items IA and IB. Here, a conversion pattern of “element” → [attribute] is determined, and a style sheet for converting the data structure between the structured document A and the structured document B is created based on the conversion pattern.

  As described above, items can be automatically and efficiently extracted from the schema file by using the correspondence file describing the correspondence of items between structured documents. In addition, the correspondence file analysis unit analyzes the correspondence between items, and the schema file analysis unit analyzes the hierarchical structure and type between items, thereby converting the conversion pattern from the structured document A to the structured document B. Can be determined for each. Therefore, a style sheet can be created automatically, efficiently and easily based on the determined conversion pattern.

The style sheet is created based on the analysis result of the correspondence file and the schema file without conversion from the schema file to the model of the correspondence file. Therefore, it is possible to prevent a lack of item information or a change in the meaning of information associated with the conversion.
<First embodiment>
[Constitution]
Hereinafter, XML is used as a meta language, an XML document described in XML is used as a structured document, an XML schema file described in XML schema is used as a schema file, and an ontology file is used as a correspondence file. An example of creating a style sheet from an XSLT file will be described.
(1) Overall Configuration An XML document conversion apparatus 100 having the XSLT file creation apparatus 101 of the present invention will be described in detail with reference to a first embodiment. FIG. 1 is an explanatory diagram showing data structure conversion between XML documents according to the first embodiment. Here, a case where conversion is performed between different data structures of metadata A described in MPEG (Moving Picture Coding Experts Group) 7 and metadata B described in TV Anytime will be described as an example. Hereinafter, a document in which metadata A is described in XML is referred to as XML document A, and a document in which metadata B is described in XML is referred to as XML document B.

  The XML document conversion apparatus 100 converts the data structure between the XML document A and the XML document B described in a data structure different from the XML document A. The XML document conversion apparatus 100 includes an XSLT file generation apparatus 101 that generates an XSLT file that is a style sheet, and an XSLT parser 102 that converts an XML data structure between XML documents based on the generated XSLT file. The XSLT file creation apparatus 101 creates an XSLT file based on the analysis result of the ontology file 107 and the analysis result of the XML schema file SA describing the structure of the XML document A and the XML schema file SB describing the structure of the XML document B. To do. The XSLT parser 102 converts the XML data structure between the XML document A and the XML document B based on the XSLT file output from the XSLT file creation apparatus 101.

(2) Configuration of XSLT File Creation Device Next, creation of a style sheet by the XSLT file creation device 101 will be described. FIG. 2 is a block diagram showing the configuration of the XSLT file creation apparatus 101. The XSLT file creation apparatus 101 includes ontology analysis means 201, schema file analysis means 202, conversion pattern determination means 203, template description means 204, and template DB 205. Here, the template description means 204 and the template DB 205 are included in the style sheet output means in the claims. The XSLT file creation apparatus 101 performs the four functions shown in FIG.
◎ Analysis of ontology file 107 (step S1)
◎ Analysis of XML schema files SA and SB (step S2)
◎ Determination of conversion pattern (step S3)
◎ Output of XSLT file based on conversion pattern (step S4)
The above four functions will be described in detail below.

(2-1) Ontology File Analysis The ontology analysis unit 201 reads the ontology file 107, extracts items to be converted from the ontology file 107, and extracts correspondences between the items.

(2-1-1) Ontology file The ontology file 107 describes the correspondence of items between the XML document A and the XML document B to be converted. For example, the ontology file 107 is described using OWL (Web Ontology language) formulated by W3C (World Wide Web Consortium). In addition, if it can describe the relationship of an item, it will not be limited to OWL. An item is a tag for clearly indicating an element position and storing an attribute in an XML document, for example. In the ontology file 107, in addition to the correspondence between items, information such as item types and item hierarchical structures can be newly defined and described. Here, the item type is, for example, “element”, “attribute”, “element value”, and / or “attribute value”. The item hierarchical structure is a parent-child relationship between items.

  When analyzing an XML schema file described later, there may be a plurality of the same items in the same XML document. Therefore, it is preferable to describe the ontology file 107 so that items to be converted can be identified. For example, if there are items whose hierarchical structure is XXX / YYY and items whose structure is ZZZ / NNN / YYY in the same XML document, simply writing YYY in the ontology file 107 will change the item to be converted. It cannot be uniquely identified. Therefore, the YYY hierarchical structure is described in the ontology file 107 to identify YYY. When NNN in ZZZ / NNN / YYY is an item that can be uniquely identified, YYY is identified by describing the relative position from the item “NNN” as NNN / YYY, for example.

(2-1-2) Conversion Pattern Defined by Ontology File Next, an example of a conversion pattern defined by the ontology file will be described. In the conversion of the data structure between XML documents, conversion is mainly performed between “element”, “attribute”, “element value”, or “attribute value”. Therefore, as a pattern to be converted, for example, a variation of the conversion pattern shown in FIG. 4 is given as an example. In FIG. 4, the “No” column indicates an ID number for uniquely identifying a conversion pattern, and the “Conversion Pattern” column indicates the conversion direction between “element”, “attribute”, “element value”, or “attribute value”. Indicates the conversion pattern. The conversion pattern “element” → “element” of No. 1 in FIG. 4 indicates that “element” of one XML document is converted to “element” of the other XML document. That is, the conversion pattern “element” → “element” is a conversion pattern in which the value of the “element” of one XML document is the same as the value of the “element” of the other XML document.

  Among the above conversion patterns, conversion pattern No. 1 to No. FIG. 5 shows an XML document conversion example up to 3. In FIG. 5, the data structure is converted from “XML document A” to “XML document B”. For example, conversion pattern No. Focusing on “element” → “element” of 1, “element” <AAA> before conversion is converted to “element” <BBB>. At this time, it can be seen that the value α of the “element” <AAA> and the value α of the “element” <BBB> are the same.

  Further, the ontology file 107 connects, for example, {a value obtained by adding the value of the item IA and the value of the item IB and the value of the item IC} or {a character string of the value of the item IA and a character string of the value of the item IB The relationship that the character string and the character string of the value of the item IC are the same value} may be described. In addition, {if "element A" is present, "element B" is converted to "element C"} conversion pattern, {"plural elements" → "element"} conversion pattern, {"plural elements" → Various conversion patterns such as a conversion pattern “multiple elements”}. Furthermore, the conversion pattern is not limited to the conversion pattern based on the type of item to be converted such as “element” or “attribute”, but may be a conversion pattern based on the number of items, contents, or the like. For example, as the conversion pattern, a conversion pattern of {input the number of “element A” included in the XML document A as the value of “element B” in the XML document B}, or the like is cited.

(2-1-3) Example of ontology description FIG. 6 shows an example in which the correspondence between these items is described by ontology based on the conversion pattern shown in FIG. FIG. 6 shows an example of ontology description corresponding to the “No” column and the “conversion pattern” column shown in FIG. 4 in the “OWL description” column. Characters such as “A”, “B”, “C”, “D”, and “E” in the “OWL description” column and the “conversion pattern” column correspond to each other. Based on the OWL description in FIG. 6, an ontology file describing the correspondence between items between XML documents is created. An example of the ontology file 107 is shown in FIG. The ontology file 107 in FIG. 7 describes the correspondence between the items of “element”, “attribute”, “element value”, and “attribute value” between the XML document A and the XML document B in accordance with the OWL description in FIG. ing.

(2-1-4) Extraction of Conversion Target Item and Analysis of Correspondence Next, ontology analysis unit 201 reads an ontology file that has read an item to be converted at the time of conversion from XML document A to XML document B. Extract from 107 description. That is, it is extracted which item and which item are converted between XML documents. The extraction is performed, for example, by reading a class or property declared in the ontology file 107 and based on which item is a conversion target. Specifically, for example, an item declared with <ow: Class rdf: ID = “...>” In the OWL description, an item called with rdf: resource = “...”, And a NameSpace ( The items to be converted are extracted by extracting the items called from the namespace. Furthermore, the ontology analysis unit 201 analyzes the correspondence between the extracted items.

  Taking the ontology file 107 as an example, the extraction and analysis of items to be converted by the ontology analysis means 201 will be described. First, in L200 of FIG. 7, a schema file to be called is declared. aaa: indicates the XML schema file SA of the XML document A, and bbb: indicates the XML schema file SB of the XML document B. L201 to L202 describe that “TypeContent” and “aaa: AnalogIC” are equal. Further, in L203 to L204, it is described that “bbb: Type” holds “TypeContent” as a value. Therefore, the ontology analyzing unit 201 analyzes the correspondence relationship that the values possessed by the item “AnalogIC” of the XML document A and the item “Type” of the XML document B are the same value by the description of L201 to L204. Here, the conversion of “Analog IC” → “Type Content” is referred to as “conversion correspondence 1”. Similarly, in L205 to L206, the correspondence relationship that the item “PartsName” of the XML document A and the item “Name” of the XML document B are the same value is analyzed. The conversion “PartsName” → “Name” is referred to as “conversion correspondence 2”. Further, in L207 to L208, the item “InputOffsetVoltage” of the XML document A and the item “unit” of the XML document B take “V”, and the value of the item “V_IO” of the XML document B is the same value. Analyze the correspondence. This conversion of “InputOffsetVoltage” + “unit” + “V” → “V_IO” is referred to as conversion correspondence 3. Then, the ontology analysis unit 201 outputs the extracted items to be converted and the correspondence between the items to the schema file analysis unit 202.

(2-2) Analysis of XML Schema File Schema file analysis means 202 reads the XML schema file SA and XML schema file SB that are conversion targets, and the hierarchy of items to be converted extracted by the analysis of ontology file 107 Analyze item information such as structure and item type. Here, it is assumed that Xpath is analyzed as a hierarchical structure. Note that the present invention is not limited to Xpath, and other position designation methods such as Xpointer may be analyzed.

  The schema file analysis unit 202 analyzes the type of “element”, “attribute”, “element value”, or “attribute value” for each item to be converted. For example, it is determined that an item designated as Element in the XML schema file is “element”, and an item designated as “Attribute” is determined as “attribute”. Here, when the item type is “element value”, “attribute value”, or the like, the item type cannot be specified only from the XML schema file. Therefore, the schema file analysis unit 202 identifies the type of item from the description of the ontology file 107. For example, the type of the item is specified assuming that the type of the item for which “hasValue” is specified in the ontology file 107 is “element value” and “attribute value”. The schema file analysis unit 202 outputs the analysis result to the conversion pattern determination unit 203.

(2-2-1) Example of Analysis of XML Schema File The analysis by the above-described schema file analysis unit 202 will be described in more detail using a schema file described in the XML schema language. 8 and 9 are examples of the XML schema file SA and the XML schema file SB, respectively.

  First, analysis of the XML schema file SA in FIG. 8 will be described. The schema file analysis unit 202 reads the item to be converted extracted by the analysis of the ontology file 107 in FIG. 5 from the XML schema file SA. Here, “AnalogIC”, “PartsName”, “InputOffsetVoltage”, and “unit” are extracted as items to be converted in the XML schema file SA. Therefore, the Xpath of these items and the item type are read from the XML schema file SA.

  The schema file analysis unit 202 analyzes from L101 of FIG. 8 that the Xpath is “/ AnalogIC” and the item type is “element” for the item “AnalogIC”. Similarly, from L102, for the item “PartsName”, it is analyzed that Xpath is “/ AnalogIC / PartsName” and the item type is “element”. Further, from L103, for the item “InputOffsetVoltage”, it is analyzed that Xpath is “/ AnalogIC / InputOffsetVoltage” and the item type is “element”. Further, the schema file analysis unit 202 indicates that the item “unit” is an “attribute” included in the item “InputOffsetVoltage” and its Xpath is “/ AnalogIC / InputOffsetVoltage / @ unit” based on the descriptions of L104 to L105. To analyze. Similarly, for the XML schema file SB of FIG. 9, the items to be converted are analyzed.

  The analysis results of the XML schema file SA and XML schema file SB are shown in FIG. The “correspondence” column in FIG. 10 shows the classifications of the conversion correspondences 1 to 3 extracted by the ontology analysis unit 201, the items to be converted for each conversion correspondence, the XML schema file name from which the items are extracted, and the type of the item , Xpath. The analysis result is output to the conversion pattern determination means 203.

(2-3) Determination of Conversion Pattern The conversion pattern determination unit 203 determines a conversion pattern between the XML document A and the XML document B based on the analysis result of the ontology file and the analysis result of the schema file. That is, based on the item to be converted extracted by the ontology analyzing unit 201, the type of each item analyzed by the schema file analyzing unit 202, and the Xpath, the conversion indicating from which type of item to which type of item is converted. Determine the pattern.

  A conversion pattern determination method by the conversion pattern determination unit 203 will be described with reference to FIGS. FIG. 10 shows the analysis results of the XML schema file SA and the XML schema file SB. FIG. 11 is an explanatory diagram showing a method for determining the conversion pattern of FIG. The conversion correspondence 1 in FIG. 10 means {“AnalogIC” and the value possessed by “Type” are the same value} from the analysis result of the ontology file 107 in FIG. Here, it is analyzed from the XML schema file SA and the XML schema file SB that the item “AnalogIC” is “element”, the item “Type” is “element”, and the “TypeContent” is “element value”. Therefore, the conversion pattern determination unit 203 determines that the conversion correspondence 1 is a conversion pattern of “element” → “element value” as shown in FIG.

  Similarly, conversion correspondence 2 in FIG. 10 means {“PartsName” and “Name” have the same value}. Here, the item “PartsName” is “element”, and the item “Name” is “element”. Therefore, the conversion pattern determination unit 203 determines that the conversion correspondence 2 is a conversion pattern of “element” → “element”.

  Also, the conversion correspondence 3 means {“InputOffsetVoltage” and “unit” takes “V” and “V_IO” has the same value}. Here, the item “InputOffsetVoltage” is “element”, the item “unit” is “attribute”, the “V” is “attribute value” of the item “unit”, and the item “V_IO” is “element”. Therefore, the conversion pattern determining means 203 determines that the conversion correspondence 3 is a conversion pattern of “element” + “attribute” + “attribute value” → “element”. The conversion pattern determination unit 203 outputs the determined conversion pattern to the template description unit 204.

(2-4) Output of XSLT file The template description means 204 searches the template DB 205 based on the conversion pattern determined by the conversion pattern determination means 203, and the template description means 204 searches the XSLT file 108 based on the searched template. Is added and output.

(2-4-1) Template DB
The template DB 205 stores, for each conversion pattern, a template that defines a conversion pattern between XML documents in a predetermined style sheet language. Here, XSLT is used as the style sheet language. FIG. 12 shows an example of a template stored in the template DB 205. FIG. 12 shows a list of templates for each conversion pattern in the ontology file shown in FIG. The “No” column and the “conversion pattern” column in FIG. 12 correspond to the conversion patterns in FIG. Also, a description example of the template corresponding to the “template” column is shown by XSLT. Note that characters such as “A”, “B”, “C”, “D”, and “E” in the “template” column and “conversion pattern” column correspond to each other. In addition, the value of Xpath of A is substituted for “/ Xpath / A” of <xsl: template match = “/ Xpath / A“> of No1.

(2-4-2) Template when data format of item value is different Examples of templates include the following templates. When the conversion pattern between XML documents is, for example, a conversion pattern from “element A” to “element B”, data of possible values even when the value of “element A” and the value of “element B” are the same The format may be different. Therefore, a template for converting the data format of values is stored in the template DB 205.

  Specifically, for example, the case where the item “Day” of the XML document A corresponds to the item “Date” of the XML document B will be described as an example. Here, “Day” and “Date” are “elements”. At the time of conversion from the XML document A to the XML document B, the value of the element “Day” has a data format in which the date is YYYYMMDD. On the other hand, the value of the element “Date” has a data format of YYMMDD. Even if the items represent the same date, the data format of the date value is different. Therefore, it is necessary to change the data format of the value when converting the element “Day” → the element “Date”. An example of a template for converting the data format of values is shown in FIG. In FIG. 13, the data format of the value of the XML document A is shown in the “XML document A” column, and the data format of the value of the XML document B is shown in the “XML document B” column. Here, it is assumed that the ontology file 107 is described in advance as “element“ Day ”having a data format of YYYYMMDD and element“ Date ”having a data format of YYMMDD are equivalent”. First, upon receiving a conversion pattern “element day” → “element date”, the template description unit 204 selects a template according to the conversion pattern. Further, a template is selected to convert the data format of the values of “element day” and “element date”. Here, the template description means 204 selects “Delete the first two characters” shown in FIG. 21 as a template for converting the data format of the value. Then, the template description unit 204 describes the XSLT file based on the selected “element A” → “element B” template and “delete the first two characters” template.

(2-4-3) Template having priority information Further, a template in which conversion priority information is set is stored in the template DB 205, and the template description unit 204 selects a template based on the priority. Or an XSLT file may be described.

(2-4-4) Creation of XSLT File The template description unit 204 receives the conversion pattern from the conversion pattern determination unit 203 and starts creating the XSLT file 108. Here, creation of the XSLT file 108 for converting the XML document A to the XML document B and the XSLT file 108 for converting the XML document B to the XML document A are started. First, when the template description unit 204 receives the conversion pattern, the template DB 205 determines whether the conversion direction is from the XML document A to the XML document B or the conversion direction from the XML document B to the XML document A. Search for templates in The template description unit 204 describes a template searched using the conversion pattern as a key in the XSLT file 108. The template description unit 204 outputs the created XSLT file 108 when the description in the XSLT file 108 is completed for all items to be converted.

  As a result of the above processing, an XSLT file 108 that converts XML document A to XML document B and an XSLT file 108 that converts XML document B to XML document A are output.

(2-4-5) Creation of XSLT File for Conversion from XML Document A to XML Document B Next, a method for creating the XSLT file 108 for converting XML document A to XML document B will be described with reference to FIG. A more specific description will be given by taking the conversion pattern of FIG. 11 as an example. FIG. 14 is an example of the XSLT file to be created. The template description unit 204 starts creating the XSLT file 108 including the XSLT header information (L501 to L502) shown in FIG. Note that the XSLT file 108 in FIG. 14 is the XSLT file 108 at the time of conversion from the XML document A to the XML document B. The template description unit 204 describes the XSLT description (parts L503 to L504) in the XSLT file in order to search for a template corresponding to the accepted conversion pattern.

  Further, in the case of conversion correspondence 1, the template description unit 204 searches the template DB 205 for a template corresponding to the conversion pattern “element” → “element value”. Here, as a conversion pattern “element” → “element value” from FIG. 7 templates are searched. And the template description means 204 is No. 7 template is added to the XSLT file (L505 to L506). Similarly, as a template for a conversion pattern of “element” → “element” corresponding to conversion 2, No. 1 in FIG. Search for one template. Then, it is added to the XSLT file of FIG. 13 (L507 to L508). For conversion correspondence 3, since the conversion pattern is “element” + “attribute” + “attribute value” → “element”, No. 9 templates are searched and added to the XSLT file (L509 to L510). When the description of the template is completed for all the conversion correspondences extracted by the ontology analysis means 201, the template description means 204 outputs the XSLT file 108.

(2-4-6) Creation of XSLT File for Conversion from XML Document B to XML Document A Next, the XSLT file 108 in the conversion from the XML document B to the XML document A is created in the same manner. The conversion pattern determination unit 203 determines a conversion pattern using the items to be converted, the item types, and the Xpath data of each item in FIG. Here, conversion of conversion correspondence 1 is “element value” → “element” conversion pattern, conversion of conversion correspondence 2 is “element” → “element” conversion pattern, and conversion of conversion correspondence 3 is “element” → “element” "+" Attribute "+" attribute value "conversion pattern. The conversion pattern determination unit 203 instructs the template description unit 204 to create an XSLT file. The template description unit 204 searches the template DB 205 for a template of the corresponding conversion pattern. The conversion pattern of “element value” → “element” is No. in FIG. 3 template, the conversion pattern from “element” to “element” is No. No. 1 has a conversion pattern “No.” → “Element” + “Attribute” + “Attribute Value”. Ten templates are searched. These templates are described in the XSLT file. FIG. 15 shows the XSLT file after the description. In FIG. 15, first, a declaration part (L801 to L802) of XSLT and a description (L803 to L804) for calling a template are performed. Then, from No. L803 to L806 in FIG. 3 corresponding to the template No. 3 from L808 to L809. 1 corresponds to No. 1 from L810 to L811. An XSLT template corresponding to 10 is described.

  Note that the conversion pattern determination unit 203 may instruct which of the XSLT files 108 to create in the conversion from the XML document A to the XML document B or the conversion from the XML document B to the XML document A. Alternatively, both XSLT files 108 may be described alternately and the XSLT file 108 may be created at the same time.

(2-4-7) XSLT file description method Here, the template description means 204 describes a template in the XSLT file 108, which is obtained by organizing the hierarchical structure of items and correcting a plurality of XSLT templates into one. It is preferable. For example, the template description unit 204 describes the template corrected as follows in the XSLT file 108. FIG. 16 is an example of an XSLT file created without modifying a template, and FIG. 17 is an example of an XSLT file created using a modified template. The XSLT file 108 in FIG. 16 describes a template of a conversion pattern in which the Xpath is AAA / BBB, CCC, AAA / DDD for each item to be converted. In this case, the item <AAA> is output twice in the portions of L301 and L302, and the description in the XSLT file increases. Therefore, a plurality of lower items are nested in the upper items, that is, hierarchized. For example, the templates are sorted in the order of the output items, such as AAA / BBB, AAA / DDD, and CCC. Then, as shown in FIG. 17, items included under the same item are grouped into one upper item and described in the XSLT file (102). As a result, the item <AAA> is output only once in the L401 portion, and the phenomenon of being output twice can be avoided.

  This will be described more specifically with reference to FIGS. FIG. 18 shows the Xpath for each item to be converted when the data structure is converted from the XML document A to the XML document B. The left column shows the Xpath of the XML document A, and the right column shows the Xpath of the XSLT file. FIG. 19 shows the XSLT file that is created.

  Here, as shown in FIG. 18, all items to be converted in the XML document A are converted to items below “Semiconductor” in the XML document B. Therefore, first, a template for converting the item of the XML document A into the item “Semiconductor” is described in the XSLT file (L601 to L602). That is, as shown in FIG. 19, the item “AnalogIC” is described in the XSLT file 108 by omitting the item “Semiconductor” (L603 to L604). Similarly, the items “PartsName” and “InputOffsetVoltage” are also described in the XSLT file 108 by omitting the item “Semiconductor” as indicated by L605 to L606 and L607 to L608, respectively.

  In the above description, hierarchization is performed based on the Xpath for each item to be converted in the XSLT file. However, hierarchization may be performed based on the Xpath of the XML document A. This will be described more specifically with reference to FIGS. FIG. 20 shows the XSLT file to be created.

As shown in FIG. 18, the items to be converted in the XML document A are all items below the item “Analog IC”. Therefore, a description for calling an item below the item “AnalogIC” is made in the template of the item “AnalogIC”. That is, as shown in FIG. 20, the following description is made when the template (L701 to L702) of the item “AnalogIC” is described in the XSLT file. A description (L703) for calling another template is made immediately before </ xsl: template> (L702) when the template of the item “AnalogIC” ends.
[processing]
FIG. 21 is a flowchart showing the flow of processing performed by the XML document conversion apparatus 100.

  Step S11: The ontology analysis means 201 reads the ontology file 107.

  Step S12: The ontology analysis means 201 extracts items to be converted from the ontology file 107.

  Step S13: The ontology analysis means 201 analyzes the correspondence between the extracted items to be converted.

  Step S14: The schema file analysis unit 202 reads the schema file of the XML document.

  Steps S15 and S16: The schema file analysis unit 202 analyzes item information such as the Xpath of the item to be converted and the item type from the read schema file.

  Step S17: The conversion pattern determination unit 203 determines a conversion pattern between the XML document A and the XML document B based on the analysis result of the ontology file and the analysis result of the schema file.

  Step S18: The template description unit 204 searches the template DB 205 to be added to the XSLT file 108 based on the conversion pattern determined by the conversion pattern determination unit 203.

  Steps S19 and S20: The template description unit 204 adds and outputs the XSLT file 108 based on the searched template.

  With the above processing, the XSLT file 108 used for conversion from the XML document A to the XML document B and the XSLT file 108 used for conversion from the XML document B to the XML document A are output from the XSLT file creation apparatus 101.

Note that the schema file analysis unit 202 may perform the schema file analysis process from step S4 to step S6 before the ontology analysis unit 201 performs the analysis process of the ontology file 107 from step S1 to step S3. In this case, first, the schema file analysis unit 202 extracts and analyzes all items in the schema file. Next, the ontology analysis unit 201 extracts items to be converted. Then, an item to be converted is extracted from all items extracted from the schema file, and its conversion pattern is determined. For the output of the XSLT file 108 from the conversion pattern, the same processing as described above is performed.
[effect]
By using a correspondence file that describes the correspondence between items between structured documents, items can be automatically and efficiently extracted from the schema file. Also, by analyzing the correspondence between items by the correspondence file analysis means and analyzing the hierarchical structure and type between items by the schema file analysis means, the conversion pattern between XML documents can be determined for each item. . Therefore, a style sheet can be created automatically, efficiently and easily based on the determined conversion pattern.

  The style sheet is created based on the analysis result of the correspondence file and the schema file without conversion from the schema file to the model of the correspondence file. Therefore, it is possible to prevent a lack of item information or a change in the meaning of information associated with the conversion.

<Other embodiment examples>
A program that causes a computer to execute the above-described processing and a computer-readable recording medium that records the program are included in the present invention. Here, examples of the recording medium include a computer readable / writable flexible disk, hard disk, semiconductor memory, CD-ROM, DVD, magneto-optical disk (MO), and others.

  It can be used in a metadata production system of a broadcasting station where the metadata data structure changes depending on the stage of shooting, editing, broadcasting, etc., and particularly suitable for a metadata data structure conversion device in the system. is there.

Explanatory drawing which shows conversion of the data structure between the XML documents which concern on the example of 1st Embodiment. 1 is a block diagram showing a configuration of an XSLT file creation apparatus 101. FIG. Four functions executed by the XSLT file creation apparatus 101. An example of conversion pattern variations. Conversion pattern No. 1 to No. 3 shows an example of conversion of XML document 3. An example in which the correspondence of these items is described in an ontology based on the conversion pattern of FIG. An example of ontology file 107. An example of an XML schema file SA. An example of an XML schema file SB. An example of the analysis result of XML schema file SA and SB. Explanatory drawing which shows the determination method of the conversion pattern of FIG. An example of the template memorize | stored in template DB205. An example of a template for converting the data format of a value. An example of the XSLT file created. An example of the XSLT file after description. An example of an XSLT file created without modifying a template. An example of an XSLT file created with a modified template An example of an Xpath for each item to be converted when the data structure is converted from the XML document A to the XML document B. XSLT file to be created. XSLT file to be created. 6 is a flowchart showing a flow of processing performed by the XML document conversion apparatus 100.

Explanation of symbols

100: XML document conversion apparatus 101: XSLT file creation apparatus 102: XSLT parser 107: ontology file 108: XSLT file 201: ontology analysis means 202: schema file analysis means 203: conversion pattern determination means 204: template description means 205: template DB

Claims (11)

A style sheet creation apparatus for converting a data structure between first and second structured documents having different data structures,
Schema file analyzing means for analyzing the first and second schema files describing the structures of the first and second structured documents,
Correspondence file analysis means for reading a correspondence file describing the association between the first and second structured documents for items which are information units included in each structured document, and analyzing the correspondence of the items; ,
Conversion pattern determining means for determining a conversion pattern from the first structured document to the second structured document based on the analysis result of the correspondence file and the analysis result of the schema file;
Style sheet output means for outputting a style sheet for converting a data structure between different structured documents of the first and second structured documents based on the conversion pattern;
A style sheet creating apparatus comprising:   The schema file analysis unit extracts items whose correspondence relationship between the first and second structured documents is analyzed from the first and second schema files by the correspondence file analysis unit, and extracts the extracted items. The style sheet creating apparatus according to claim 1, wherein information is analyzed.   The style sheet creating apparatus according to claim 2, wherein the item information is a type of the item and / or a hierarchical structure of the item.   The style sheet creating apparatus according to claim 3, wherein the item type includes an element, an attribute, an element value, and / or an attribute value.   The style sheet creation apparatus according to claim 1, wherein the correspondence file is an ontology file described by OWL (Web Ontology Language). The schema file analysis unit extracts items whose correspondence between the first and second structured documents is analyzed by the correspondence file analysis unit from the first and second schema files, and analyzes the information. And
The style sheet creation apparatus according to claim 1, wherein the conversion pattern determination unit determines a conversion pattern for each item by replacing a correspondence relationship of the extracted items with a correspondence relationship of item information.   The style sheet creation apparatus according to claim 1, wherein the conversion pattern determination unit further determines a conversion pattern from the second structured document to the first structured document. The style sheet output means includes
Conversion pattern storage means for storing, for each conversion pattern, a template that defines a conversion pattern from the item of the first structured document to the item of the second structured document in a predetermined style sheet language;
A template description unit that extracts a template corresponding to each conversion pattern determined by the conversion pattern determination unit from the conversion pattern storage unit, and describes the conversion pattern in the style sheet language in the style sheet based on the extracted template; ,
The style sheet creating apparatus according to claim 1, comprising:   9. The style sheet creation apparatus according to claim 8, wherein the conversion pattern storage unit stores a template defining a conversion pattern of item information.   2. The structured document according to claim 1, wherein the structured document is an XML (eXtensible Markup Language) document, the schema file is an XML schema file, and the style sheet is an XSLT (XML Stylesheet Language Transformations) file. Style sheet creation device. A stylesheet creation method for converting a data structure between first and second structured documents having different data structures,
A schema file analyzing step of analyzing first and second schema files describing the structure of each of the first and second structured documents;
A correspondence file analysis step of reading a correspondence file describing the association between the first and second structured documents for items which are information units included in each structured document, and analyzing the correspondence of the items; ,
A conversion pattern determining step for determining a conversion pattern from the first structured document to the second structured document based on the analysis result of the correspondence file and the analysis result of the schema file;
A style sheet output step of outputting a style sheet for converting a data structure between different structured documents of the first and second structured documents based on the conversion pattern;
A method for creating a style sheet, comprising:

Images