JP2006277662A - Conversion rule correction program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a conversion rule correction program capable of improving the efficiency of correction work corresponding to a change in a data format made to be a conversion object about definition information of a data format conversion rule. <P>SOLUTION: In the conversion rule correction program, a computer executes the correction of definition information of a conversion rule between a first data format and a second data format on the basis of a change in the definition information of the first data format. The correction program has an analysis procedure for calculating a formal difference of the definition information between before and after changing the first data format and deriving logical change information about the change in the definition information of the first data format on the basis of the difference (S12) and a first correcting procedure for correcting the definition information of the conversion rule on the basis of the logical change information derived in the analysis procedure (S14). <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a conversion rule correction program, and more particularly, to a conversion rule correction program that causes a computer to execute correction of definition information of a data format conversion rule.

  In a company, a large number of computer systems are operating, and each system generally has its own database. When it is necessary to perform data linkage between a plurality of systems, it is inefficient to create a linkage tool for each combination of systems. Therefore, conventionally, there is a method of defining an intermediary or common database (hereinafter referred to as “integrated DB”), and attempting data linkage between a plurality of systems via the integrated DB.

  FIG. 1 is a conceptual diagram for explaining data linkage between systems via an integrated DB. As shown in FIG. 1, the following format conversion is performed between each system (system A, system B, and system C) and the integrated DB in order to absorb the difference in the respective data formats.

  First, when data managed in each system is input to the integrated DB, the input API 501 is used to absorb the difference between the data format of each system and the integrated DB. It is converted into a data format in a common integrated DB. On the other hand, when data managed in the integrated DB is output to each system, the output API 502 is used to absorb the difference between the data format of each system and the integrated DB. It is converted into a data format for each system.

  According to this method, each system only needs to be aware of the input / output interface with the integrated DB, which is more efficient than a case where a linkage tool is created for each combination of systems.

  However, when a specification change occurs in the data format of the integrated DB, not only the integrated DB itself but also an input API or an output API (hereinafter referred to as “input / output API”) that links data items related to the specification change. For all of the above, complicated operations such as coding, compiling and linking are required. Therefore, the expansion of the integrated DB once introduced is very expensive.

  Therefore, in recent years, XML (eXtensible Markup Language) technology is often used for data format conversion. XML data can be easily converted in data format using an XSLT (XSL Transformations) file. XML is suitable for expressing a data format of a database. Therefore, by expressing the contents of the database and integrated DB of each system in XML, the XSLT file can be used as a data format conversion function instead of the input / output API.

  The XSLT file is a simple text file and does not require compilation. Therefore, by using the XSLT file, it is possible to reduce the number of maintenance steps for changing the specification of the integrated DB.

Patent Document 1 describes a technique for improving the efficiency of editing structured data such as an XML document.
Japanese Patent Laid-Open No. 2003-345781

  However, even when XSLT is used, modification of the XSLT file related to the specification change of the integrated DB is still required. Conventionally, this operation has been performed manually. The amount of correction increases as the number of systems increases, and the risk of correction errors increases. After all, there was a problem that it was unavoidable to link many systems with the integrated DB.

  The present invention has been made in view of the above points, and is a conversion rule that can improve the efficiency of correction work according to a change in the data format to be converted with respect to the definition information of the data format conversion rule. The purpose is to provide a fix.

  Therefore, in order to solve the above-mentioned problem, the present invention executes the correction of the definition information of the conversion rule between the first data format and the second data format on the computer based on the change of the definition information of the first data format. A conversion rule correction program for obtaining a formal difference in definition information before and after the change of the first data format, and a logical change relating to a change in the definition information of the first data format based on the difference An analysis procedure for deriving information, and a first correction procedure for correcting the definition information of the conversion rule based on the logical change information derived in the analysis procedure.

  In such a conversion rule correction program, it is possible to automatically correct the definition information of the conversion rule for processing the data format based on the change of the definition information of the data format. Therefore, it is possible to improve the efficiency of the modification of the conversion rule according to the change of the data format definition information.

  According to the present invention, it is possible to provide a conversion rule correction program that can improve the efficiency of the correction work according to the change of the data format to be converted with respect to the definition information of the data format conversion rule.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 2 is a diagram illustrating a hardware configuration example of the format conversion file correction apparatus according to the embodiment of the present invention. Here, the format conversion file describes definition information of conversion rules used to absorb the difference in data format between the integrated DB and each system linked to the integrated DB (hereinafter referred to as “linked system”). 1 and corresponds to the positioning of the input API 1501 or the output API 1502 in FIG.

  The integrated DB refers to an intermediary or common database for performing data linkage between a plurality of systems. In the present embodiment, an XSLT (XSL Transformations) file is described as a specific example of the format conversion file, but the format conversion file is not necessarily limited to the XSLT file. However, when the data format of each linkage system and integrated DB is expressed in the XML format, it is desirable to define the format conversion file by XSLT having high affinity with XML. Hereinafter, the format conversion file functioning as the input API 501 is referred to as an input format conversion file, and the format conversion file functioning as the output API 502 is referred to as an output format conversion file.

  The format conversion file correction device 10 of FIG. 2 includes a drive device 100, an auxiliary storage device 102, a memory device 103, an arithmetic processing device 104, a display device 105, and an input device that are mutually connected by a bus B. 106.

  A program for realizing processing in the format conversion file correction device 10 is provided by a recording medium 101 such as a CD-ROM. When the recording medium 101 on which the program is recorded is set in the drive device 100, the program is installed from the recording medium 101 to the auxiliary storage device 102 via the drive device 100. The auxiliary storage device 102 stores the installed program and also stores necessary files and data.

  The memory device 103 reads the program from the auxiliary storage device 102 and stores it when there is an instruction to start the program. The arithmetic processing unit 104 executes functions related to the format conversion file correction device 10 in accordance with a program stored in the memory device 103. The display device 105 displays a GUI (Graphical User Interface) or the like by a program. The input device 106 includes a keyboard and a mouse, and is used to receive various operation instructions.

  FIG. 3 is a diagram illustrating a functional configuration example of the format conversion file correction device according to the embodiment of the present invention. 3, the format conversion file correction device 10 includes a format definition file difference analysis unit 11, a format conversion file correction unit 12, a test case correction unit 13, a test execution unit 14, and the like as a format conversion file correction function. Each of these units is realized by a program installed in the format conversion file correction device 10.

  FIG. 4 is a flowchart for explaining the processing outline of the format conversion file correction apparatus. The function of each part in FIG. 3 will be described along the processing procedure of FIG.

  It is assumed that there is a specification change in the data format of the integrated DB and the format definition file of the integrated DB is modified. Here, the format definition file refers to a file in which a data format (schema information) is defined. In the present embodiment, it is assumed that the data format is defined by XML.

  In response to an instruction from the user (menu selection, icon click, etc.), the format definition file difference analysis unit 11 displays the new integrated DB format specification screen 21 on the display device 106, and operates the corrected format definition file. Upload to the area (S11).

  FIG. 5 is a diagram showing a display example of a new integrated DB format designation screen. In FIG. 5, the new integrated DB format designation screen 21 is composed of a modification table selection area 211, a format definition file designation area 212, a reference button 213, a modification button 214, and the like.

  The correction table selection area 211 is an area for selecting, for example, a table to be corrected in the data format from the tables constituting the integrated DB, and the table name is set for each table constituting the integrated DB. It is displayed. For example, it can be seen that the product master association includes a product addition table, a product update table, a product search table, and the like.

  The format definition file specification area 212 is a text input area for specifying a corrected format definition file for the table selected in the correction table selection area 211. The user may input the file name of the corrected format definition file by directly inputting a character string in the format definition file specification area 212, or by clicking the browse button 213 to display a file dialog. The corrected format definition file may be selected from the file dialog.

  The correction button 214 is a button for instructing upload of the corrected format definition file to the work area for the table selected in the correction table selection area 211. Therefore, when the correction button 214 is clicked, the format definition file is uploaded.

  When the format definition file is uploaded, the format definition file difference analysis unit 11 reads the uploaded format definition file (hereinafter referred to as “new format definition file”) and the conventional integrated DB format definition file 31 (hereinafter referred to as “old format definition file”). A formal difference from the “format definition file” is obtained, and logical change information of both is derived based on the difference (S12).

  Here, the logical change information of the format definition may not be uniquely determined. For example, it may not be possible to determine whether a data item has been newly added or the name of an existing data item has been changed. Therefore, the format definition file difference analysis unit 11 identifies all possible change patterns and outputs the change information of each change pattern as the change pattern information 34. However, it is possible to estimate the certainty of each change pattern from the similarity of the data item names before and after the change. Therefore, the format definition file difference analysis unit 11 calculates the certainty (hereinafter referred to as “determinism”) for each change pattern and outputs it as the definiteness information 35.

  Subsequently, the format conversion file correction unit 12 automatically corrects all the format conversion files 32 for all the change patterns using the change pattern information 34 and the certainty degree information 35, and converts the correction result into the format conversion file correction result. The information 36 is output (S14). Here, all format conversion files 32 refer to all format conversion files 32 defined for each cooperative system.

  Subsequently, the test case correction unit 13 automatically corrects the test case information 33 using the change pattern information 34 and the determination degree information 35. The test case information 33 refers to information for verifying whether the data format conversion by the corrected format conversion file is performed correctly when the format conversion file is corrected.

  FIG. 6 is a diagram for explaining the test case information. In FIG. 6, the upper part shows the test on the input side (conversion from each cooperative system to the integrated DB), and the lower part shows the test on the output side (conversion from the integrated DB to each cooperative system).

  In the test on the input side, the test input data 331 in the data format specific to each cooperative system is converted into the input format conversion file 32a, and the conversion result is compared with the verification data 332 in the integrated DB data format prepared in advance. Is done. If they match, the validity of the input format conversion file 32a is verified.

  In the test on the output side, the initial test data 333 in the data format of the integrated DB is converted into the output format conversion file 32b, and the conversion result is compared with the verification data 334 in the data format of each linked system prepared in advance. Is done. If the two match, the validity of the output format conversion file 32b is verified.

  Here, the test input data 331, the verification data 332, the test initial data 333 and the verification data 334 correspond to the test case information 33. However, the test case information 33 that needs to be corrected in accordance with the change in the data format of the integrated DB is limited to the verification data 332 and the test initial data 333. That is, the input test data 331 on the input side and the verification data 334 on the output side are information that depends on the data format of each cooperative system, and are not affected by the change in the data format of the integrated DB.

  Note that the test case information 33 is required for each linkage system. Therefore, in step S16, the verification data 332 and the test initial data 333 for all the linked systems are automatically corrected, and the correction result is output as the test case correction result information 37.

  Subsequently, the format conversion file correction unit 12 displays the correction results of the format conversion file 32 and the test case information 33, that is, the format conversion file correction result information 36 and the test case correction result information 37 on the correction content confirmation screen 22. (S17).

  FIG. 7 is a diagram illustrating a display example of the correction content confirmation screen. In FIG. 7, the correction content confirmation screen 22 includes a format conversion file selection area 221, a test case selection area 222, a correction result display area 223, another correction result display area 224, a button 225, a button 226, a button 227, a button 228, and the like. Has been.

  The format conversion file selection area 221 displays a list of corrected format conversion files, and is an area for selecting a format conversion file to be viewed from the list. The test case selection area 222 displays a list of corrected test case information, and is an area for selecting test case information to be viewed from the list.

  The correction result display area 223 is an area in which the format conversion file or the test case information selected in the format conversion file selection area 221 or the test case selection area 222 is displayed. The separate correction result display area 224 is an area in which, when an arbitrary tag in the correction result displayed in the correction result display area 223 is clicked, another correction result is displayed for the block that the tag configures. That is, the separate correction result display area 224 is an area in which correction results based on other change patterns are displayed for the block selected in the correction result display area 223 when the change pattern is not uniquely determined.

  Immediately after the correction content confirmation screen 22 is displayed, a correction result based on the change pattern with the highest degree of definiteness is displayed.

  The button 228 is a button for executing a test using the format conversion file correction result information 36 and the test case correction result information 37. The buttons 225 and 226 and 227 will be described later.

  Thus, the user can view all the format conversion files and the test case information after correction on the correction content confirmation screen 22.

  When the user is satisfied with the correction result (yes in S18) and the button 228 is clicked (S19), the test execution unit 14 determines that the format conversion file correction result information 36 and the test case correction result information 37 are based on FIG. The test as described in (1) is automatically executed (S20). The test execution unit 14 displays the test result on the test result display screen 24 (S21).

  When the user is satisfied with the test result (yes in S22) and clicks the save button on the test result display screen 24 (S23), the test execution unit 14 normalizes the format conversion file correction result information 36 and the test case correction result information 37. The format conversion file 32 or the test case information 33 is saved (S24).

  On the other hand, when the user is not satisfied with the correction result in step S18 (no in S18) and clicks an arbitrary tag in the correction result displayed in the correction result display area 223, the format conversion file correction unit 12 or the test case correction unit 13 displays other correction results (correction results based on other change patterns) relating to blocks (hereinafter referred to as “selected blocks”) formed by the tag in the separate correction result display area 224 (S25). In addition, when the previous candidate button 229a or the next candidate button 229b in the different correction result display area 224 is clicked, the format conversion file correction unit 12 or the test case correction unit 13 corrects the selected block based on another change pattern. Display the result.

  When an appropriate correction result exists in the correction results displayed in the separate correction result display area 224 (Yes in S26) and the user clicks the button 225 (S27), the format conversion file correction unit 12 or the test case correction unit 13 displays the format conversion file correction result information 36 (corrected format conversion file) related to the correction result displayed in the separate correction result display area 224 in the correction result display area 223 (S28).

  At this time, if there are a plurality of format conversion files having the same correction result for the block displayed in the separate correction result display area 224, the format conversion file correction unit 12 or the test case correction unit 13 The file closest to the format conversion file displayed in the correction result display area 223 is automatically selected and displayed.

  On the other hand, there is no appropriate information in the format conversion file correction result information 36 based on all the change patterns (that is, the automatically corrected correction result) (no in S26), and the user can check the correction content confirmation screen 22 ( When the button 226 (edit correction content candidate) or the button 227 (create new correction content candidate) in FIG. 7) is clicked (S29), the format conversion file correction unit 12 displays the correction content candidate edit screen 23 (S30). .

  FIG. 8 is a diagram illustrating a display example of the modification content candidate edit screen. In FIG. 8, the modification content candidate editing screen 23 includes an application pattern input area 231, a modification result input area 232, a priority order input area 233, a confirmation button 234, and the like.

  The application pattern input area 231 is an area for inputting information (hereinafter referred to as “application pattern”) for specifying a location to which correction is to be applied. In addition to character strings that completely match, the application target can be specified abstractly using regular expressions or the like.

  The correction result input area 232 is an area for inputting a correction result to be applied to a portion that matches the application pattern. The correction result can be described in an abstract manner by using a part of the application pattern as a variable. For example, in the figure, {1} which is the description 2312 in the correction result means that the character string existing in the {1: *} portion which is the description 2311 in the application pattern is inserted as it is.

  The priority order input area 233 is an area for inputting the priority order of correction contents including a combination of the application pattern and the correction result. The priority order of the correction contents is that a number of correction contents can be registered using the correction contents candidate editing screen 23, and any correction contents are given priority when a correction portion to which a plurality of application patterns apply is detected. It is a parameter for specifying

  When the confirmation button 234 is clicked on the correction content candidate editing screen 23, the format conversion file correction unit 12 registers the correction content input on the correction content candidate editing screen 23 in the correction content candidate master 38.

  FIG. 9 is a diagram illustrating a configuration example of a correction content candidate master. In FIG. 9, the correction content candidate master 38 includes items such as a correction content candidate ID, a type, an application pattern, a correction result, and a priority order for each correction content. The correction content candidate ID is an item in which information for identifying each correction content is registered. The correction content candidate ID is automatically numbered. The type is an item in which a target type to which correction content is applied (an input format conversion file, an output conversion file, test initial data, verification data, or the like) is registered. The application pattern, the correction result, and the priority order are items in which information input to the correction content candidate edit screen 23 is registered as it is.

  Further, the format conversion file correction unit 12 or the test case correction unit 13 corrects the format conversion file correction result information 36 or the test case correction result information 37 based on the correction contents registered in the correction content candidate master 32 ( S31), the correction result is displayed on the correction content confirmation screen 22 (S17). Subsequent operations are as described above.

  Next, a format definition file difference analysis process (S12) by the format definition file difference analysis unit 11, a format conversion file correction process (S14) by the format conversion file correction unit 12, and a test case correction process (S16) by the test case correction unit 13 ) Will be described in detail.

  First, details of the format definition file difference analysis processing will be described. Here, a case where the format definition file is changed as shown in FIG. 10 is taken as an example.

  FIG. 10 is a diagram illustrating an example of the format definition file before and after the change of the integrated DB in the embodiment of the present invention. For reference, FIG. 11 shows an example of specific XML data of the integrated DB based on the format definition file before and after the change.

  In FIG. 10, a format definition file 311 indicates before change, and a format definition file 312 indicates after change. Hereinafter, the pre-change format definition file 311 and the post-change format definition file 312 are referred to.

  The pre-change format definition file 311 will be described. In the description 311-1, the description 311-2, and the description 311-3, an employee_id element, a name element, and a furigana element are defined. Here, the term “element” follows XML and is associated with a data item on the integrated DB.

  Description 311-4 defines an employee element. It is also defined that the employee_id element, name element, furigana element, and english_info element are child elements of the employee element.

  In description 311-5, an english_info element is defined. It is also defined that the name element is a child element of the english_info element.

  That is, the format definition in the pre-change format definition file 311 forms a tree structure shown in the tree 311t.

  Next, the post-change format definition file 312 will be described. In the description 312-1, the description 312-2, the description 312-3, and the description 312-4, an employee_id element, a name element, a furigana element, and an english_name element are defined.

  In the description 312-5, the name_info element is defined. The name element, furigana element, and english_name element are also defined as child elements of the name_info element.

  In the description 312-6, an employee element is defined. Further, it is also defined that the employee_id element and the name_info element are child elements of the employee element.

  That is, the format definition in the post-change format definition file 312 constitutes a tree structure shown in the tree 312t.

  To explain with reference to the tree 311t and the tree 312t, the post-change format definition file 312 adds a name_info element to the pre-change format definition file 311 (a1), and changes the name element and furigana element under the employee element. This is equivalent to moving the element under the name_info element (m1, m2), renaming the name element under the english_info element to the english_name element, moving it under the name_info element (n1), and further deleting the english_info element (d1).

  The format definition file difference analysis processing will be described taking the above change as an example. FIG. 12 is a flowchart for explaining the format definition file difference analysis processing.

  It is difficult for a computer program to immediately derive the logical changes described with reference to the tree 311t and the tree 312t. Therefore, first, the formal (simple) difference between the format definition files before and after the change is analyzed, and the analysis result is output as simple difference information (S121).

  FIG. 13 is a diagram illustrating an example of simple difference information. In FIG. 13, the simple difference information 41 indicates an element type, change contents, and the like for each change element. The changed element is a path name of an element that has changed regardless of whether the format definition file 311 before change or the format definition file 312 after change. The element type is used to identify a tag and an attribute because an attribute or the like can be defined for the tag in XML. The change content is information indicating the type of change. Thus, the formal difference means a difference between the two derived by simply comparing the description contents before and after the change of the format definition file. In the present embodiment, a new addition or deletion of a change element is made. It is judged either. That is, in the formal difference analysis, changes that require logical judgment, such as movement of a change element and change (rename) of an element name, are not determined.

  In this way, all elements detected as differences are extracted as simple difference information 41.

  Subsequently, a loop process for each record (change element) in the simple difference information 41 is executed. That is, first, one of the unresolved change elements is selected as a processing target (S122). However, here, the change content “deleted” is preferentially selected. The changed element selected here is hereinafter referred to as “current element”. An unresolved change element refers to a change element whose change content has not been finalized. In the initial state, all change elements are treated as unresolved change elements.

  Subsequently, possible logical changes for the current element are identified (S123). Here, logical changes include not only “new addition” or “deletion” but also a certain guess for judging the contents such as “movement” and “element name change”. Changes that require logical analysis.

  That is, the logical change content cannot always be uniquely derived from the simple difference information 41. For example, even if the changed content is “deleted”, a plurality of changed contents are conceivable, such as when the element is really deleted, moved, or the element name is changed. So here are all the possible changes. Details of this process will be described later.

  Subsequently, it is determined whether or not the change content of the current element is unique (S124). If the change content of the current element is unique (Yes in S124), information regarding the change content is output as change pattern information 34, assuming that the current element has been resolved (S125).

  On the other hand, when the change content of the current element is not unique, the current element is postponed as an unresolved change element. That is, the change contents of the current element may be narrowed down by resolving the change contents of the other change elements.

  Following step S125 or no in step S124, it is determined whether or not the change elements whose change contents are uniquely determined have been exhausted (S126). If not exhausted, the processing from S122 to S125 is repeated. If it runs out, the process proceeds to step S130.

  In step S131, one unresolved change element is selected. The changed element selected here is hereinafter referred to as “current element”. Subsequently, possible logical changes for the current element are identified (S132). This process is the same as step S123. However, there is a possibility that the change contents that are identified in step S132 when the change contents of other change elements are resolved are less than the previous time.

  Subsequently, one change content is selected from the possible change contents for the current element (S133). The change content selected here is referred to as the current change content. Subsequently, it is determined whether or not the current element is the last change element (S134). If the current change element is not the last change element (no in S134), the step of changing the current element is assumed to be the current change content. S130 is recursively executed (S135). That is, for example, if the change content for the current element is “add new”, each change is recursively selected one by one from the change content that can be taken by all other unresolved change elements. As a combination of element changes, a consistent (change pattern) is generated.

  On the other hand, if the current element is the last change element (Yes in S134), it is assumed that one consistent combination (change pattern) has been completed, and the change pattern is output as the change pattern information 34 (S136). . At this time, the uniquely determined change contents solved in step S125 are also added to the combination to form one change pattern.

Subsequently, it is determined whether or not all possible changes regarding the current element that has become the root of the recursion process have been processed (S137). If all the changes are not processed, one of the unprocessed changes is set as the current change, and the processes after step S134 are repeated. By repeating this process for all the changes of the current element that has become the root of the recursive process, all combinations of the changes without contradiction are output to the change pattern information 34. FIG. It is a figure which shows the example of a structure. In FIG. 14, the change pattern information 34 includes items such as a pattern ID, a change element, an element type, change contents, a movement source, and a movement destination for each change element.

  The pattern ID is an ID for identifying each change pattern. That is, one change pattern is constituted by the change contents for one or more change elements having the same pattern ID. In FIG. 14, at least pattern IDs “1” and “2” are displayed.

  The changed elements, element types, and changed contents are the same as those in the simple difference information 41 (FIG. 13). However, the change contents here are specified not only for new addition or deletion, but also for logical change such as movement or element name change.

  The movement source is the path name of the parent element that belonged to the change element whose change content is “move” or “delete” before the change. The move destination is the path name of the parent element that has newly belonged to the change element whose change contents are “Add new” or “Move”, or the new element name for the change element of “Change element name”. is there.

  When the processes for all the change patterns are completed (S137), the degree of determination is calculated for each change pattern and is output as the degree of determination information 35 (S138).

  Subsequently, in step S123 or S132, a process for identifying a logical change content that may be possible for each changed element, that is, only “new addition” and “deletion” were represented in the simple difference information 41. Details of processing for subdividing the contents of change into “new addition”, “move”, “deletion”, “type change”, “element name change”, and the like will be described. The type change means a type change between an element surrounded by tags and an attribute defined in the tag.

  FIG. 15 is a flowchart for explaining a process for identifying a logical change content that is possible for one change element.

  First, it is determined whether the change content of the current element is “deletion” or “new addition” (S123a). Here, the current element is a change element selected in step S122 or S131 of FIG. When it is determined that the change content is “new addition”, “new addition” is secured as the possibility of the change content of the current element (S123b). That is, there is always a possibility of new addition. However, if the current element is already set as a movement destination or the like by analysis of another changed element, it is handled that there is no possibility of “new addition”. Here, the possibility of moving, changing the type, changing the element name, etc. is not considered. Regarding these changes, since the change element whose change content is “deletion” always exists as a pair in the simple difference information 41, the change content is identified as the change content regarding the change element “deletion” as described later. is there.

  On the other hand, if it is determined in step S123a that the change content of the current element is “deletion”, there is one change element (hereinafter referred to as “new addition element”) whose change content is “new addition” in the simple difference information 41. Selected (S123c).

  Subsequently, it is determined whether or not the element names of the current element and the newly added element match (S123d). If they match, it is determined whether or not the element type (whether it is a tag or an attribute) matches (S123e). If the element types are the same, a new addition is added as a possibility of changing the current element. “Move” with the element as the move destination is secured (S123f).

  On the other hand, if the element types do not match, “type change” is secured as a possibility of the change contents of the current element (S123g). In addition, when the parent element of the current element is different from the newly added element, “move” with the newly added element as the movement destination is also secured. That is, “type change” and “movement” are treated as being performed at the same time.

  If the element names of the current element and the newly added element do not match (no in S123d), the possibility of “change of element name” is investigated as the change contents of the current element (S123h). Details of this processing will be described later.

  Following step S123f, S123g, or S123h, the process proceeds to step S123i, where it is determined whether or not the processing has been completed for all the new additional elements in the simple difference information 41. The subsequent processing is also performed for the remaining newly added elements.

  If the process has been completed for all the newly added elements, “delete” is secured as the possibility of changing the current element (S123j), and the process ends. Note that “deletion” is ensured because the possibility of “deletion” cannot be completely denied.

  Next, a process for investigating the possibility of element name change in step S123h will be described. FIG. 16 is a flowchart for explaining a process for investigating the possibility of changing an element name.

  First, the similarity between the element name of the current element and the element name of the newly added element is calculated (S123h-1). Here, the similarity is a scale indicating the similarity between two character strings, and is obtained by the following formula as an example.

Similarity = number of matched characters ÷ (number of matched characters + number of different characters)
For example, consider “employee_name_information” and “emp_kana_name_info”. The number of matching characters in both character strings is 12 characters in total, including 3 characters “emp” and 9 characters “_name_info”. The number of different characters is a total of 12 characters including “royee” and “_kana” having the larger number of characters, that is, “royee” and 5 characters.

Therefore, the similarity is
12 / (12 + 12) = 0.5
It becomes.

  Subsequently, it is determined whether or not the similarity is equal to or greater than a predetermined threshold (S123h-2). If the similarity is smaller than a predetermined threshold, the possibility of “changing the element name” is denied.

  On the other hand, if the degree of similarity is equal to or greater than a predetermined threshold, “change element name” with the element name of the newly added element as the element name after the change is secured as the possibility of changing the current element (S123h-3). . Subsequently, it is determined whether or not “change” is accompanied as the change content (S123h-4). Whether or not “moving” is accompanied may be determined by determining the difference between the parent element of the current element and the newly added element. When it is determined that “move” is accompanied, “move” with the parent element of the newly added element as the move destination is secured as a possibility of changing the current element (S123h-5).

  Next, the determination degree calculation process in step S138 of FIG. 12 will be described. To explain again, “determinism” refers to the certainty of each change pattern, and is an example of a scale for ranking a plurality of change patterns.

  In this embodiment, each change element is weighted for each change content candidate, and the sum for each change pattern is taken for the weight. Furthermore, the value obtained by dividing the sum of the weights of the respective change patterns by the sum of the weights of all the change patterns is set as the determinism. Therefore, the problem is how to weight each change element candidate for the change contents. Hereinafter, an example in the present embodiment will be described.

First, weighting is determined according to the contents of change. For example,
Define a weighting parameter value between 0 and 1 so that Move> Change Type> Change Element Name> Add New> Delete. This magnitude relationship may be determined according to the possibility of occurrence of each change content. In the present embodiment, for example, the weight for “deletion” is minimized from the viewpoint that it is unlikely that an existing data item will be deleted along with the function expansion. The weighting parameters for each change will be described as a_move (movement), a_type (type change), a_name (element name change), a_add (new addition), and a_del (deletion).

  Regarding the element name change, the above-mentioned similarity is taken into consideration. Here, the similarity is b.

  Regarding movement, some amount of movement is considered. That is, the smaller the amount of movement, the higher the determinism is estimated. Here, the amount of movement refers to the amount of movement in the data structure (tree structure), not the number of lines moved in the format definition file. For example, the movement amount is calculated by setting the movement amount in the upper and lower one layer as 1, and the reciprocal thereof is set as c.

  As for the movement, it is estimated that the higher the similarity between the names of the parent elements before and after the movement, the higher the determinism. Therefore, considering the similarity of the parent element names, this parameter is set to d.

  Furthermore, regarding the movement, whether or not a plurality of elements have moved to the same name element is also considered, and when a plurality of elements have moved to the same name element, a new additional attribute (XML) is added to the destination element. When there is no attribute added to the tag in (), the determinism is lowered. In other words, if the type defined as an element before the change is changed to the attribute of a certain element, there should be a newly added attribute in the destination element. However, the absence of a newly added attribute means that the source element has been deleted, and as described above, there is a low possibility that it will be deleted during function expansion. Let this parameter be p_dup. The value of p_dup is appropriately determined from 0 to 1, and the value of p_dup is 1 when an attribute is added to the tag of the destination element.

  Considering whether or not the element name change and movement, or the type change and movement occur at the same time, the determinism is lowered when at least one of them occurs at the same time. However, if the element name before the change is similar to the element name of the parent element that is the move destination, or if the element name of the parent element before the move is similar to the element name after the change, the degree of determinism is lowered Make it smaller. In this case, the deterministic averages of the element name change and the movement are averaged and further multiplied by a penalty parameter p_multi (0 to 1). If f is the greater of the similarity between the element name before the change and the element name of the parent element that is the move destination, and the similarity between the element name of the parent element before the move and the element name after the change, the penalty parameter Can be obtained by the following equation.

p_multi = δ + (1-δ) · f
Here, δ is an adjustment parameter having a range of 0-1.

  From the above, the degree of determination (s_rec) for each change content is calculated by the following equation, for example.

Movement: s_rec = a_move × (c + γ ・ d) × p_dup
Here, γ is an adjustment parameter.

Change type: s_rec = a_type
Change element name: s_rec = a_name x b
Add: s_rec = a_add
Delete: s_rec = a_del
Therefore, the determinism (s_el) for one change element is
If multiple changes for one change element do not occur simultaneously,
s_el = s_rec
If multiple changes occur simultaneously for one change element,
s_el = (Σs_rec / number of changes) x p_multi
Is calculated as

  For reference, for example, when a_move = 1, a_type = 0.8, a_name = 0.3, a_add = 0.3, a_del = 0, γ = 0.5, δ = 0.3, the determinism (s_el) for some of the changed elements Try to calculate.

For the addition of “employee / name_info”,
Determinism: s_el = a_add = 0.3
It becomes.

For moving “employee / name” below “employee / name_info”,
Reciprocal of travel distance: c = 1/1 = 1
Similarity of element name of parent element: d = 0
Than,
Determinism: s_el = a_move x (c + γd) / (1 + γ) = 0.67
It becomes.

Regarding the change of the element name from “employee / english_info / name” to “english_name” and the movement of “employee / english_info / name” to “employee / name_info”
similarity between name and english_name: b = 4 / (4 + 8) = 0.33,
Determinism of element name change: s_rec = a_name × b = 0.1
Movement amount from “employee / english_info” to “employee / name_info”: 2 (reciprocal: c = 0.5)
Determinism of movement: s_rec = a_move × (c + γd) / (1 + γ) = 0.47
Here, since movement and element name change occur simultaneously, p_multi is calculated.

Similarity between the element name “english_info” of the parent element before movement and the element name “english_name” after the change:
8 / (8 + 5) = 0.62
Similarity between the element name “name” before the change and the element name “name_info” of the new element:
4 / (4 + 5) = 0.44
f = (larger of 0.62 and 0.44) = 0.62
p_multi = δ + (1 − δ) f = 0.73
Therefore,
Determinism: s_el = (0.1 + 0.47) / 2 × 0.73 = 0.21
It becomes.

About deletion of "english_info"
Determinism: s_el = a_del = 0
It becomes.

  When the determinism (s_el) is calculated for each change element of the change pattern information 34 (FIG. 14) in this way and further the determinism is calculated for each change pattern, for example, as shown in FIG. FIG. 17 is a diagram illustrating the degree of determination calculated for the change element and the change pattern.

The determinism of each change element of the change pattern with pattern ID 1 (hereinafter referred to as “change pattern 1”, and other change patterns follow the same naming convention) is 0.30, 0.67 in order from the top. , 0.67, 0.21, and 0 are calculated. Since the definite degree (s_pat) of the change pattern 1 is the sum of the deterministic degrees of the change elements belonging to the change pattern 1,
s_pat = 0.30 + 0.67 + 0.67 + 0.21 + 0 = 1.85
It becomes.

  Similarly, the definiteness of change pattern 2 is calculated as 1.67.

  Therefore, it can be seen that the change pattern 1 has a higher determinism than the change pattern 2. Note that the degree of determination for each change pattern is output as the degree of determination information 35.

  By the way, the definiteness of each change pattern has been described above as the definiteness of the change pattern divided by the sum of the determinism of all the change patterns (that is, the ratio), but here, pattern 1 and pattern 2 Since it is only displayed, it was calculated as an absolute value for convenience.

  This is the end of the description of the format definition file difference analysis processing by the format definition file difference analysis unit 11. As described above, by this processing, the change pattern information 34 that is information indicating a change pattern that is likely to be changed in the format definition file and the deterministic information 35 that is the determinism for each change pattern are output.

  Next, details of the format conversion file correction process (FIG. 4: S14) by the format conversion file correction unit 12 will be described. That is, it is a process of automatically correcting all the format conversion files 32 for each linked system using the change pattern information 34 and the certainty information 35. In order to facilitate the explanation, here, a description will be given of correction of a format conversion file with a predetermined linkage system (hereinafter referred to as “linkage system A”) as an example.

  FIG. 18 is a diagram for explaining the outline of the format conversion file correction process for the linkage system A.

  In FIG. 18, a tree 511, a tree 311t, and a tree 312t indicate the data format of the linkage system A, the data format of the integrated DB before the change, and the data format of the integrated DB after the change, respectively.

  The arrow between the tree 511 and the tree 311t in the upper part of the figure indicates the format conversion file between the linkage system A and the integrated DB before the change. That is, an arrow 321a indicates an output format conversion file, and an arrow 321b indicates an input format conversion file.

  Similarly, an arrow between the tree 511 and the tree 312t in the lower part of the figure indicates a format conversion file between the cooperation system A and the integrated DB after change.

  Therefore, in the automatic correction processing of the format conversion file, the output format conversion file indicated by the arrow 322a based on the output format conversion file indicated by the arrow 321a (hereinafter referred to as “pre-correction output format conversion file 321a”). (Hereinafter, referred to as “corrected output format conversion file”) is generated, and an arrow 322b is generated based on an input format conversion file (hereinafter referred to as “pre-correction input format conversion file 321b”) indicated by an arrow 321b. Is generated (hereinafter referred to as “corrected input format conversion file 322b”).

  As is clear from the tree 511, the data format of the cooperation system A is assumed to be composed of an EMP element, an EMP_NAME element, an EMP_KANA element, and an EMP_ROMAN element as child elements of the EMP element. The EMP element has an EMP_NO attribute.

  FIG. 19 is a diagram illustrating a definition example of an output format conversion file before and after correction. 19, in the output format conversion file 321a before correction, the employee element is the EMP element, the employee / employee_id element is the EMP_NO attribute of the EMP element, the employee / name element is the EMP / EMP_NAME element, and the employee / furigana element is The EMP / EMP_KANA element defines that the employee / english_info / name element is converted to an EMP / EMP_ROMAN element.

  On the other hand, in the post-correction output format conversion file 322a, the employee element is an EMP element, the employee / employee_id element is the EMP_NO attribute of the EMP element, the employee / name_info / name element is the EMP / EMP_NAME element, and the employee / name_name / name_g It is defined that an element is converted into an EMP / EMP_KANA element, and an employee / name_info / english_name element is converted into an EMP / EMP_ROMAN element.

  In addition, the part enclosed with the broken line in a figure shows a mutually different part before and behind correction.

  FIG. 20 is a diagram illustrating a definition example of an input format conversion file before and after correction. In FIG. 20, the pre-correction format conversion file 321b includes an EMP element in an employee element, an EMP_NO attribute in an employee / employee_id element, an EMP / EMP_NAME element in an employee / name element, and an EMP / EMP_KANA element in an employee / furigan element. , EMP / EMP_ROMAN element is defined to be converted into an employee / english_info / name element, respectively.

  On the other hand, in the post-correction input format conversion file 322b, the EMP element is an employee element, the EMP_NO attribute is an employee / employee_id element, the EMP / EMP_NAME element is an employee / name_info / name element, and the EMP / EMP_KANAin element is an employee / emplome / emploe It is defined in the / furigana element that the EMP / EMP_ROMAN element is converted into an employee / name_info / english_name element, respectively.

  In addition, the part enclosed with the broken line in a figure shows a mutually different part before and behind correction. Further, in FIGS. 19 and 20, the reference numerals attached to the respective rows are used in the following description.

  FIG. 21 is a flowchart for explaining the format conversion file correction process. In FIG. 21, it is generically referred to as a “format conversion file” regardless of input or output.

  First, the format conversion file correction unit 12 selects one format conversion file to be processed from the format conversion file 32 (S141). The format conversion file selected here is hereinafter referred to as a “current format conversion file”.

  Subsequently, the format conversion file correction unit 12 selects one change pattern from the change pattern information 34 (FIG. 14) (S142). The change pattern selected here is hereinafter referred to as “current change pattern”.

  Subsequently, the format conversion file correction unit 12 corrects the current format conversion file based on the current change pattern, and outputs the format conversion file as the correction result as the format conversion file correction result information 36 (S143).

  Subsequently, the format conversion file correction unit 12 determines whether the correction of the current format conversion file based on all the change patterns has been completed (S144). When the correction based on all the change patterns has not been made, the processes after step S142 are repeated. When the correction based on all the change patterns has been completed, the format conversion file correction unit 12 determines whether or not all the format conversion files 32 have been corrected (S145). If there is an uncorrected format conversion file 32, the processes after step S141 are repeated. When the processing has been completed for all the format conversion files 32, the format conversion file correction processing ends.

  As is clear from the above processing procedure, all the format conversion files are automatically corrected based on all the change patterns.

  Next, details of the format conversion file correction process based on the change pattern in step S143 will be described. FIG. 22 is a flowchart for explaining the format conversion file correction processing based on the change pattern.

  First, one change element is extracted from the change elements constituting the current change pattern in the order (new addition, element name change, type change, move, delete order) according to the change contents (S143a). Subsequently, the current format conversion file is corrected according to the change contents of the extracted change element (S143b).

  Subsequently, it is determined whether or not the processing has been completed for all the changed elements (S143c), and if there are unprocessed changed elements, the processes after step S143a are repeated. When the processing has been completed for all the changed elements, the correction processing based on the correction content registered in the correction content candidate master 38 (FIG. 9) is performed on the current format conversion file (S143d). That is, it is checked whether or not there is a part that matches the application pattern in the current format conversion file. If there is a part that matches, the correction result is applied to the part. When a plurality of contradictory correction contents can be applied, the correction contents with the highest priority are applied.

  The corrected current format conversion file is output as format conversion file correction result information 36 (S143e).

  Next, correction of the format conversion file according to the change content in step S143b will be described for each change content.

  The case of new addition will be described. FIG. 23 is a diagram for explaining correction contents to be applied to the format conversion file when the change contents are newly added. FIG. 23 shows a modification example based on the first line of the change pattern information 34 (FIG. 14), that is, according to the new addition of the name_info element below the employee element. In addition, the code | symbol attached | subjected to each line in the specific example of FIG. 23 respond | corresponds to the thing in FIG. 19 or FIG. 20, and is the same also in subsequent figures.

  As shown in (A) in the figure, as a modification to the pre-modification output format conversion file 321a, an xsl: template element targeting the added element (name_info) is added (description 322a-25, description 322a-31).

  Further, as shown in (B) in the figure, as a modification to 321b in the pre-modification input format conversion file, a parent element (employee) to be added is searched, and the parent element (description 321b-5-5) is searched. The start tag and the end tag of the element (nameinfo) to be added are added to the description 321b-12) (descriptions 322b-7 and 11).

  The case of changing the element name will be described. FIG. 24 is a diagram for explaining the correction content to be applied to the format conversion file when the change content is an element name change. FIG. 24 shows a modification example in accordance with the change of the element name (name) of the employee / english_info / name element to English_name, based on the fourth line of the change pattern information 34 (FIG. 14).

  As shown in (A) in the figure, as a modification to the output format conversion file 321a before modification, first, the <xsl: value-of select = “expression”> tag or the <xsl: template match = “pattern”> Like an expression or pattern in a tag or the like, a character string for specifying an element to be subjected to XSLT processing is searched for a character string that matches the element name (name) of the change source. Subsequently, the searched character string is replaced with the element name (english_name) of the change destination (description 322a-29). However, even if the element names match, if the paths (employee / english_info) do not match, they will not be replaced.

  Further, as shown in (B) in the figure, as a modification to the pre-modification input format conversion file 321b, a tag name that matches the element name (name) before the modification is searched, and the retrieved tag name is changed. It is replaced with the previous element name (english_info) (description 322b-30, description 322b-32). However, even if the element names match, if the paths (employee / english_info) do not match, they will not be replaced.

  The case of movement will be described. FIG. 25 is a diagram for explaining correction contents to be applied to the output format conversion file when the change contents are movement. FIG. 25 shows a modification example in accordance with the movement of the employee / name element below employee / name_info, based on the second line of the change pattern information 34 (FIG. 14).

  First, a character string for specifying an element to be subjected to XSLT processing is searched for a character string that matches the element name (name) of the element to be moved (description 321a-9) (however, the movement source) Only those that match the path of (employee) are targeted). Subsequently, xsl: apply-templates that is the target parent element (name_info) as a sibling element (element having the same parent element (EMP)) of the element to which the searched character string belongs (description 321a-9). Elements are generated (description 322a-9). However, it is not generated if the element already exists. Subsequently, the element (description 322a-9) to which the first searched character string belongs is moved as a child element of the xsl: template element (description 322a-25) that is the target parent element (name_info). (Description 322a-26).

  FIG. 26 is a diagram for explaining correction contents to be applied to the input format conversion file when the change contents are movement. FIG. 26 also shows a modification example based on the second line of the change pattern information 34 (FIG. 14), that is, according to the movement of the employee / name element below employee / name_info.

  First, a tag whose tag name is the element name (name) of the element to be moved is searched (description 321b-20). Subsequently, the xsl: apply-templates element that calls the template (description 321b-19) to which the retrieved tag belongs is retrieved (description 321b-7), and the retrieved element (description 321b-7) is the destination. (Name_info: description 322b-7 to description 322b-11) (description 322b-8).

  The movement of other elements such as employee / furigana is also processed in the same manner as in FIG. 25 and FIG.

  The case of deletion will be described. FIG. 27 is a diagram for explaining correction contents to be applied to the format conversion file when the change contents are deletion. FIG. 27 shows a modification example based on the sixth line of the change pattern information 34 (FIG. 14), that is, according to the deletion of the employee / english_info element.

  As shown in (A) in the figure, the template element (description 321a-27, description 321a-31) for the element (english_info) to be deleted is deleted as a modification to the output format conversion file 321a before modification. Further, the xsl: apply-templates element (description 321a-11) referring to the template is deleted.

  Further, as shown in (B) in the figure, as a modification to the pre-modification input format conversion file 321b, a tag (description 321b-) having the element name (english_info) of the element to be deleted as a tag name is used. 9, description 321b-11) is deleted.

  This is the end of the description of the format conversion file correction process by the format conversion file correction unit 12. As described above, by this processing, all the existing format conversion files 32 are corrected based on all the change patterns, and output as format conversion file correction result information 36.

  Next, the details of the test case correction process (FIG. 4: S16) by the test case correction unit 13 will be described. That is, this is a process of automatically correcting all the test case information 33 for each cooperative system using the change pattern information 34 and the certainty degree information 35. In order to facilitate the explanation, the correction of the test case information 33 related to the cooperation system A will be described as an example.

  FIG. 28 is a diagram showing test case information used for testing an input format conversion file. In FIG. 28, test input data 331 is test input data based on the data format of the linkage system A (FIG. 18: tree 511). The pre-correction verification data 332a is verification data based on the data format of the integrated DB before the change (FIG. 18: tree 311t). The post-correction verification data 332b is verification data based on the data format of the integrated DB after the change.

  The test of the input format conversion file 32a is to cause the input format conversion file 32a to convert the test input data 331 and compare the information registered in the integrated DB as the conversion result with the verification data. Is. Here, since the data depending on the data format of the integrated DB is verification data, the correction target of the test case information 33 used for testing the input format conversion file is limited to the verification data. . Therefore, the process of correcting the pre-correction verification data 332a and generating the post-correction verification data 332b is one of the test case information correction processes.

  FIG. 29 is a diagram showing test case information used for testing the output format conversion file. In FIG. 29, initial test data 333a before correction is initial data for testing based on the data format of the integrated DB before change. The post-correction test initial data 333b is test initial data based on the data format of the integrated DB after the change. The verification data 334 is test verification data based on the data format of the cooperation system A.

  The test of the output format conversion file 32b causes the output format conversion file 32b to convert the initial test data stored in the integrated DB, and the information output as the conversion result and the verification data 334 are It is to compare. Here, since the data depending on the data format of the integrated DB is the initial data for testing, the test target information 33 used for testing the output format conversion file is limited to the initial data for testing. Is done. Therefore, the process of correcting the initial data for testing 333a before correction to generate the initial data for testing 333b after correction is another test case information correction process.

  FIG. 30 is a flowchart for explaining the test case correction process. In FIG. 30, the verification data or the initial test data is generically referred to as “test case information 33”.

  First, the test case correction unit 13 selects one test case information to be processed from the test case information 33 (S161). The test case information selected here is hereinafter referred to as “current test case information”.

  Subsequently, the test case correction unit 13 selects one change pattern from the change pattern information 34 (FIG. 14) (S162). The change pattern selected here is hereinafter referred to as “current change pattern”.

  Subsequently, the test case correction unit 13 corrects the current test case information based on the current change pattern, and outputs the test case information as the correction result as the test case correction result information 37 (S163).

  Subsequently, the test case correction unit 13 determines whether or not the current test case information has been corrected based on all the change patterns (S164). When the correction based on all the change patterns has not been made, the processes after step S162 are repeated. When the correction based on all the change patterns has been completed, the test case correction unit 13 determines whether or not all the test case information 33 has been corrected (S165). If there is uncorrected test case information 33, the processes after step S161 are repeated. When the process is completed for all the test case information 33, the test case correction process ends.

  As is clear from the above processing procedure, all the test case information 33 is automatically corrected based on all the change patterns.

  Next, details of the test case information correction process based on the change pattern in step S163 will be described. FIG. 31 is a flowchart for explaining test case information correction processing based on a change pattern.

  First, one change element is extracted from the change elements constituting the current change pattern in the order corresponding to the change contents (new addition, element name change, type change, move, delete order) (S163a). Subsequently, the current test case information is corrected according to the change contents of the extracted change element (S163b).

  Subsequently, it is determined whether or not the processing has been completed for all the changed elements (S163c). If there is an unprocessed changed element, the processes after step S163a are repeated. When the process has been completed for all the changed elements, the modification process based on the modification contents registered in the modification content candidate master 38 (FIG. 9) is performed on the current test case information (S163d). That is, it is checked whether there is a part that matches the application pattern in the current test case information. If there is a part that matches, the correction result is applied to the part. When a plurality of contradictory correction contents can be applied, the correction contents with the highest priority are applied.

  The corrected current test case information is output as test case correction result information 37 (S163e).

  Subsequently, the correction of the test case information according to the change content in step S163b will be described. FIG. 32 is a diagram for explaining the correction contents to be applied to the test case information according to the change contents. 28 and 29, the pre-correction verification data 332a (FIG. 28) and the pre-correction test initial data 333a (FIG. 29) may be exactly the same. In this case, the corrected verification data 332b (FIG. 28) and the corrected test initial data 333b are exactly the same. Therefore, only an example of correcting the pre-correction verification data 332a based on the change pattern 1 of the change pattern information 34 (FIG. 14) will be described here.

  Information to be corrected in correcting the test case information 33 is based on the data format of the integrated DB. The change pattern information 34 is information indicating a change pattern related to a change in the data format of the integrated DB. Therefore, the test case information 33 may be corrected as indicated by the change pattern information 34.

  As for the new addition (S51), the target element (name_info) is added in the element (employee) to be added (description 332b-1, description 332b-7).

  Regarding the element name change (S52), the element name (name) to be changed is replaced with the element name (english_name) to be changed (description 332b-6).

  As for the movement (S53), the target element (employee / name, employee / furigana, employee / english_info / name) is moved to the element to be moved (employee / name_info) (description 332b-4, 5, 6).

  For deletion (S54), the target element (employee / english_info) is deleted (description 332a-5, 7).

  This is the end of the description of the test case correction process by the test case correction unit 13. As described above, by this processing, all existing test case information 33 is corrected based on all change patterns, and is output as test case correction result information 37.

  As described above, according to the format conversion file correction device 10 in the embodiment of the present invention, based on the format definition file before and after the change of the integrated DB, the change content is automatically estimated, and the estimated change content is changed. Based on this, the existing format conversion file 32 and the test case information 33 can be automatically corrected. Further, the test can be automatically executed using the corrected format conversion file and the test case information. Therefore, it is possible to reduce the work man-hours associated with the specification change of the integrated DB, and it is also possible to prevent bugs from being manually mixed.

  Further, when there are a plurality of change patterns, the user is given a room for selection by browsing the change patterns, so that more appropriate correction can be performed. At that time, since the degree of determination is calculated and presented to the user for each change pattern, the user can obtain an effective determination criterion in selecting the change pattern.

  In addition, by reducing the work man-hour associated with the specification change of the integrated DB, the user can expand the function of the integrated DB with peace of mind.

  Further, since maintenance work of the integrated DB is reduced, the integrated DB can be introduced as a general package. It is possible to introduce a linkage between systems that are already supported, and the integrated DB can be upgraded by replacing it with a common latest version, not for each customer.

  As mentioned above, although the Example of this invention was explained in full detail, this invention is not limited to the specific embodiment which concerns, In the range of the summary of this invention described in the claim, various deformation | transformation * It can be changed.

Regarding the above description, the following items are further disclosed.
(Supplementary note 1) A conversion rule correction program for causing a computer to correct definition information of conversion rules between the first data format and the second data format based on a change in definition information of a first data format. ,
An analysis procedure for obtaining a formal difference between the definition information before and after the change of the first data format and deriving logical change information related to the change of the definition information of the first data format based on the difference;
A conversion rule correction program comprising: a first correction procedure for correcting definition information of the conversion rule based on the logical change information derived in the analysis procedure.
(Supplementary Note 2) The analysis procedure derives a plurality of patterns that can be taken as the logical change information based on the difference,
The conversion rule correction program according to appendix 1, wherein the first correction procedure corrects the definition information of the conversion rule based on each of the plurality of patterns.
(Supplementary Note 3) For each change element constituting the logical change information, a deterministic degree corresponding to the change content for each change element is calculated, and the definition information of the first data format is calculated based on the definite degree. The change rule correction program according to appendix 1 or 2, further comprising a determinism calculation procedure for calculating a determinism with respect to logical change information related to the change.
(Supplementary Note 4) Based on the logical change information derived in the analysis procedure, data based on the first data format before the change is corrected to data based on the first data format after the change. The conversion rule correction program according to any one of appendices 1 to 3, wherein the conversion rule correction program has two correction procedures.
(Additional remark 5) It has a verification procedure which verifies the definition information of the said conversion rule corrected in said 1st correction procedure using the data corrected in said 2nd correction procedure. The conversion rule correction program described.
(Additional remark 6) It has the correction result display procedure which presents the correction result in said 1st correction procedure, and displays the screen for inputting the correction with respect to the presented correction result on a display apparatus. 5. The conversion rule correction program according to any one of 5 above.
(Additional remark 7) The said analysis procedure analyzes the said logical change information for every change element which comprises the said difference, and combines the said logical change information for every said change element of said 1st data format. The change rule correction program according to any one of appendices 1 to 6, wherein logical change information related to a change in definition information is derived.
(Supplementary Note 8) In the first correction procedure, the correction according to the change contents of each change element constituting the logical change information related to the change of the definition information of the first data format is added to the definition information of the conversion rule. The change rule correction program according to any one of appendices 1 to 7, characterized in that it is applied.
(Additional remark 9) Said 2nd correction procedure is said 1st before change before the change according to the change content for every change element which comprises the logical change information regarding the change of the definition information of said 1st data format The change rule correction program according to appendix 4 or 5, which is applied to data based on a data format.
(Supplementary Note 10) A conversion rule correction device that corrects definition information of a conversion rule between the first data format and the second data format based on a change in definition information of a first data format,
Obtaining a formal difference between definition information before and after the change of the first data format, and analyzing means for deriving logical change information related to the change of the definition information of the first data format based on the difference;
A conversion rule correction apparatus comprising: first correction means for correcting the definition information of the conversion rule based on the logical change information derived by the analysis means.
(Additional remark 11) It is the conversion rule correction method which corrects the definition information of the conversion rule of said 1st data format and a 2nd data format using a computer based on the change of the definition information of a 1st data format. And
An analysis procedure for obtaining a formal difference between the definition information before and after the change of the first data format and deriving logical change information related to the change of the definition information of the first data format based on the difference;
A conversion rule correction method comprising: a first correction procedure for correcting the definition information of the conversion rule based on the logical change information derived in the analysis procedure.
(Additional remark 12) The computer-readable recording medium which recorded the change rule correction program as described in any one of additional remark 1 thru | or 9.

It is a conceptual diagram for demonstrating the data cooperation between systems via integrated DB. It is a figure which shows the hardware structural example of the format conversion file correction apparatus in embodiment of this invention. It is a figure which shows the function structural example of the format conversion file correction apparatus in embodiment of this invention. It is a flowchart for demonstrating the process outline | summary of a format conversion file correction apparatus. It is a figure which shows the example of a display of a new integrated DB format designation | designated screen. It is a figure for demonstrating test case information. It is a figure which shows the example of a display of a correction content confirmation screen. It is a figure which shows the example of a display of a correction content candidate edit screen. It is a figure which shows the structural example of a correction content candidate master. It is a figure which shows the example of the format definition file before and behind change of integrated DB in embodiment of this invention. It is an example of the concrete XML data of integrated DB based on the format definition file before and behind change. It is a flowchart for demonstrating a format definition file difference analysis process. It is a figure which shows the example of simple difference information. It is a figure which shows the structural example of change pattern information. It is a flowchart for demonstrating the process for identifying the logical change content which is possible about one change element. It is a flowchart for demonstrating the investigation process of the possibility of element name change. It is a figure which shows the fixed degree calculated about the change element and the change pattern. It is a figure for demonstrating the outline | summary of the correction process of the format conversion file about the cooperation system A. FIG. It is a figure which shows the example of a definition of the output format conversion file before and behind correction. It is a figure which shows the example of a definition of the input format conversion file before and behind correction. It is a flowchart for demonstrating a format conversion file correction process. It is a flowchart for demonstrating the correction process of the format conversion file based on a change pattern. It is a figure for demonstrating the correction content performed to a format conversion file when a change content is newly added. It is a figure for demonstrating the correction content performed to a format conversion file when a change content is element name change. It is a figure for demonstrating the correction content performed to the format conversion file for output when a change content is a movement. It is a figure for demonstrating the correction content performed to the format conversion file for input when a change content is a movement. It is a figure for demonstrating the correction content performed to a format conversion file when a change content is deletion. It is a figure which shows the test case information used for the test of the format conversion file for input. It is a figure which shows the test case information used for the test of the format conversion file for an output. It is a flowchart for demonstrating a test case correction process. It is a flowchart for demonstrating the correction process of the test case information based on a change pattern. It is a figure for demonstrating the correction content performed to test case information according to a change content.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Format conversion file correction apparatus 11 Format definition file difference analysis part 12 Format conversion file correction part 13 Test case correction part 14 Test execution part 21 New integrated DB format specification screen 22 Correction content confirmation screen 23 Correction content candidate edit screen 24 Test result display Screen 31 Integrated DB format definition file 32 Format conversion file 33 Test case information 34 Change pattern information 35 Determinism information 36 Format conversion file correction result information 37 Test case correction result information 38 Correction content candidate master 39 Test result information 100 Drive device 101 Recording Medium 102 Auxiliary storage device 103 Memory device 104 Arithmetic processing device 105 Display device 106 Input device B Bus

Claims (5)

A conversion rule correction program for causing a computer to correct the definition information of the conversion rule between the first data format and the second data format based on the change of the definition information of the first data format,
An analysis procedure for obtaining a formal difference between the definition information before and after the change of the first data format and deriving logical change information related to the change of the definition information of the first data format based on the difference;
A conversion rule correction program comprising: a first correction procedure for correcting definition information of the conversion rule based on the logical change information derived in the analysis procedure. The analysis procedure derives a plurality of patterns that can be taken as the logical change information based on the difference,
The conversion rule correction program according to claim 1, wherein the first correction procedure corrects the definition information of the conversion rule based on each of the plurality of patterns. For each change element that constitutes the logical change information, a deterministic degree corresponding to the change content for each change element is calculated, and a logical for changing the definition information of the first data format based on the definite degree The change rule correction program according to claim 1 or 2, further comprising a determinism calculation procedure for calculating a determinism with respect to various change information. A second correction procedure for correcting data based on the first data format before the change to data based on the first data format after the change based on the logical change information derived in the analysis procedure. The conversion rule correction program according to any one of claims 1 to 3, further comprising: 5. The conversion according to claim 4, further comprising: a verification procedure for verifying the definition information of the conversion rule corrected in the first correction procedure using the data corrected in the second correction procedure. Rule correction program.

Images