United States Patent
`
`[19]
`
`[11] Patent Number:
`
`4,893,232
`
`Shimaoka et a1.
`
`[45] Date of Patent:
`
`Jan. 9, 1990
`
`[54] DATA MANAGEMENT SYSTEM FOR
`EFFICIENTLY SEARCHING DATA
`ELEMENTS AND HAVING A FLEXIBLE
`REGISTRATION ORDER OF DATA
`ELEMENTS
`
`[75]
`
`Inventors:
`
`Kouiehi Shimaoka; Matoi Iiznka,
`both of Tokyo, Japan
`
`[73] Assignee:
`
`Kabnsbiki Kaisha Toshiba, Kawasaki,
`Japan
`
`[21] App]. No.: 235,617
`
`[22] Filed:
`
`Aug. 19, 1988
`
`Related US. Application Data
`
`[63]
`
`Continuation of Ser. No. 932,677, Nov. 20, 1986, aban-
`doned.
`
`Foreign Application Priority Data
`[30]
`Nov. 30, 1985 [JP]
`Japan ................................ 60-269633
`
`Int. cu ................................................ G06F 1/00
`[51]
`[52] us. or. .................................... 364/200; 364/300;
`364/2822; 364/2814
`[53] Field of Search ........................ 364/200, 300, 900
`
`[56]
`
`References Cited
`U.s. PATENT DOCUMENTS
`
`4,627,019 12/1986 Ng ....................................... 364/900
`4,774,661
`9/1988 Kumpati .............................. 364/300
`
`Primary Examiner—Raulfe B. Zache
`Attorney. Agent, or Firm—Finnegan, Henderson,
`Farabow, Garrett and Dunner
`
`[57]
`
`ABSTRACI‘
`
`In a data management system of this invention, data
`elements to be registered are managed by item informa-
`tion records each comprising a name of the data ele-
`ment, an item code corresponding to an item to which
`the data element belongs, and a serial number. The
`relation between item information records is managed
`by a relation information record expressed by a source
`item ID, an object item ID, an item code associated
`with a source corresponding to an item, and a relation
`ID indicating the relation between the source and ob-
`ject items. The relation ID is given by a r latiou table.
`When a search instruction is input, the it
`information
`record, the relation information record, and the relation
`table are referred to, thus obtaining item information to
`be searched. The item formation record and relation
`information record can be registered as needed.
`
`24 Claims, 8 Drawing Sheets
`
` DATA DiCTiONARY/
`
`DIRECTORY ACCESS SECTION
`
`IBM EX. 1009
`
`001
`
`001
`
`

`

`US. Patent
`
`Jan. 9, 1990
`
`Sheet 1 of8
`
`4,893,232
`
`FIG.I
`
`
`
`PRINTER
`
`DISPLAY
`
`9
`
`8
`
`FIG.2
`
`I4
`
`
` RELATION
`
`TABLE
`
`DATA DICTIONARY/
`
`DIRECTORY_ACCESS SECTION-
`
`'.
`34
`
`
`RELATION
`ITEIVI INFORMATION
`
`
`
`RECORD
`QgéflgATION
`RECORD
`
`
`
`SEARCH
`REGISTERING
`SECTION
`QEgITTO-ER'NG
`
`
`
`
`
`SECTION
`
`
`
`
`
`
`
`I0
`
`
`
`ITEM
`RELATION
`
`
`| NFORMATION
`INFORMATION
`
`
`F I LE
`FILE
`
`
`002
`
`002
`
`

`

`US. Patent
`
`J
`
`8cm
`
`4,893,232
`
`
`
`
`
`00,Elmma;w”:82.9:m70.2592.Bfimfl92m:_lmsazFfizflmS;
`
`
`
`
`
`
`
` xmoz.5W:>m<028mw.xwoz.SE:>m<_>=mn_
`
`<m.0_n_
`
`29.5%SEEZom..=<._.mo:
`
`moooSE:momDOm1“>m<ozoomm
`
`7_O_._.<n_mm:
`
`n:53.:FOMBO
`
` 20.24%“.In:SE:momDOm
`
`
`
`xm02_ZO_._.<4mm
`
`
`
`
`
`e.eo%0m0_“—
`
`axmoz.2053mmEgan.
`
`mooosub
`
`003
`
`003
`
`
`
`
`

`

`US. Patent
`
`Jan. 9, 1990
`
`Sheet 3 of 8
`
`4,893,232
`
`
`
`Quiz/Ewe2m:_OOOxw SEE.NO0Oxa._ZO_._.<§mO....Z_
`
`EE
`
`
`
`
`053.50 smE_ooom3_ZOFSZEOnZ.
`93.5mm_29.2582”_84.450smEmo0095g!_zo_p<s_mo.._z_
`
`
`
`
`
`ESE223mEEEE
`
`
`
`éofiflmo2053mmEEEE
`
`
`
`ESEzgzdmfigfig
`
`E55020.29EEEE
`
`
`
`
`
`
`
`3»
`
`mV
`
`9»
`
`av
`
`<0
`
`mm
`
`om
`
`on
`
`_n_
`
`_u_
`
`_n_
`
`_n_
`
`_u_
`
`_n_
`
`_u_
`
`_u_
`
`004
`
`004
`
`
`
`

`

`US. Patent
`
`Jan. 9, 1990
`
`Sheet 4 of 8
`
`4,893,232
`
`
`
`
`
`mIIIIIIIIIIIIIII
`IIIIIIIIIIIIIIII
`IIIIIIIIIIIIIII
`
`IEIIIIIIIIIIIIIII
`IIIIIIIIIIIIIII
`
`IIIIIIIIIIIIEIEIEEII
`
`IIIIIIIIIIEIEIIIIII
`
`IEEEIIIIIIIEEIEEIIIIII
`
`IEEIIIIIIIIIIIEEIIIII
`
`IEEIIIIIIIIIIIIIIII
`
`IIIEHEEEEEEIEEIEEEEEEII
`
`IIIIIIIIIIIIIII
`
`IEEIIEEIEEIEEIEEIIIEEIIIIII
`
`mIIIIIIIIIIEflIIIEfl
`
`IRE-IIIIIIIEEIIHEIIEEEEII
`
` IIEEIEEEIifl-IIIIIIEEIEEEEE
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`DD: DATA ITEM
`I
`IO: /0 INFORM.
`
`WS' WORK STATION
`' I/O INFORM.
`
`PRI PRINT OUTPUT
`I NFORM.
`
`PU. PHYSICAL PATH
`- BETWEEN NODEs
`
`LU' LOGICAL PATH
`‘ BETWEEN NODES
`
`FLI FILE
`
`JK IJOB FUNCTION
`
`JB 2 JOB
`
`AC lACCESS
`SPECIFIC.
`LCI LNK SPECIFIC.
`
`PG: PROGRAM
`
`JCS JOB STEP SEQ.
`
`AR 2 AREA
`
`ND: NODE
`
`; GRAPHIC DISF.’
`GS SEQ.
`LNI LINE SPECIFIC.
`
`VI ' TRANSACTION
`' VOUJME
`
`RR ilTEM CORRELATION V2 ZFILE VOLUME
`
`T, RECORD ITEM
`' CORRELATION
`
`DBI DATA BASE
`
`T2 I TREE CORRELATION
`
`DA? DB AREA
`
`RD: RECORD
`
`OP : PROCESSING
`OPERATION
`
`005
`
`005
`
`

`

`US. Patent
`
`Jan. 9, 1990
`
`8
`
`4,893,232
`
`
`
`006
`
`

`

`US. Patent
`
`Jan. 9, 1990
`
`Sheet 6 of8
`
`4,893,232
`
`FIG.8
`
`START
`
`SPECIFY SOURCE
`ITEM CODE
`
`SPECIFY OBJECT
`ITEM CODE
`
`REFER RELATION TABLE TO
`OBTAIN RELATION ID
`BETWEEN SOURCE AND
`O'JECT ITEM CODES.
`
`I2
`
`SI4
`
`SIG
`
`W
`
`
`
`
`READ OUT ONE ITEM
`INFORMATION RECORD TO
`OBTAIN SOURCE ITEM ID
`
`
`
`OT REGISTERED
`
`READ OUT ONE ITEM
`INFORMATION RECORD TO
`OBTAIN OBJECT ITEM II
`
`
`
`
`
`
`
`
`PRODUCE AND REGISTER
`RELATION INFORMATION
`
`RECORD
`
`
`COMPLETED WITH
`REGARD TO CURRENT
`SOURCE ITEM ID
`
`
`
`COMPLE
`D WI H
`REGARD TTOEALL SBURC:
`ITEM ID
`
`INFORM AN ERROR
`TO WORK STATION
`$34
`
`828
`
`-N
`E D
`
`30
`
`32
`
`007
`
`007
`
`

`

`US. Patent
`
`Jan. 9, 1990
`
`Sheet 7 of 3
`
`4,893,232
`
`FIG.9A
`
`344
`
`s45
`
`START
`
`SPECIFY SOURCE ITEM
`CONDITION
`
`SPECIFY OBJECT ITEM
`CONDITION
`
`| ||I I I I
`
`I
`
`| I I
`
`SPECIFY RELATION
`I_ ______________ _j
`
`ONLY RELATION
`SPECIFIED?
`N
`
`Y
`
`548
`
`ITEM ID
`PECIFIED?
`
`Y
`
`OBTAIN RELATION
`INFORMATION
`
`358
`
`
`
`
`COMPLETED?
`
`Y
`
`856
`
`G
`
`008
`
`
`OBTAIN ITEMI
`USING
`
`
`
`PRIMARY ITEM NDEX
`
`LIST UP ITEM IDS w
`
`854
`36*
`
`
`LIST UP SOURCE AND
`OBJECT ITEM IDS
`
`$60
`
`008
`
`

`

`US. Patent
`
`Jan. 9, 1990
`
`Sheet 8 of 8
`
`4,893,232
`
`FIGQB
`
`0
`
`OURCE-> OBJECT ?
`
`Y
`362
`
`
`
`
`OBTAIN RELATION
`OBTAIN RELATIONAL
`
`INFORMATION USING
`INFORMATION USING
`SECONDARY RELATION INDEX
`PRIMARY RELATION INDEX
`
`RELATIO
`376 N
`
`
`PECIFIED?
`
`
` DOES OBTAINED
`
`TION ID COINCIDE W
`
`SPECIFIED
`
`
`
`LIST UP ITEM INFORMATION
`
`S70
`
`872
`
`009
`
`009
`
`

`

`1
`
`4,893,232
`
`DATA MANAGEMENT SYSTEM FOR
`EFFICIENTLY SEARCHING DATA ELEMENTS
`AND HAVING A FLEXIBLE REGISTRATION
`ORDER OF DATA ELEMENTS
`
`This application is a continuation of application Ser.
`No. 932,677, filed Nov. 20, 1986, now abandoned.
`
`BACKGROUND OF THE INVENTION
`
`The present invention relates to a data management
`system suitable for creating and managing various data-
`bases.
`Conventionally, a data dictionary/directory (DD/D)
`is adopted as a database management system. The
`DD/D manages a data item appearing on a database as
`a core of a data system and its attribute, accumulates
`“metada ”, and generally performs system operations
`and maintenance support. However, since a complex
`“relation” between various data elements cannot be
`expressed by the DD/D, it is difficult to search data
`elements and items having a relation therebetween.
`As a means for expressing the “relation” between
`data elements, a relational database (RDB) using a rela-
`tion model has been proposed. However, in order to
`express the relation with the RDB, a “table” must be
`created, and matching processing during a search oper-
`ation takes a great deal of time. In addition, in order to
`update one record, in records must be updated.
`Alternatively, as a means for expressing the “rela-
`tion" between data elements, a COOASYL (an acro-
`nym for Conference For Data Systems Languages) type
`database is known. However, in a database of this type,
`a registration order of the data elements is fixed, and
`arbitrary data elements cannot be registered as needed,
`or data elements cannot be related after they are regis-
`tered.
`
`SUMMARY OF THE INVENTION
`
`The present invention has been made in consideration
`of the above situation and has as its object to provide a
`data management system which can efficiently search
`item information associated with a data element.
`The data management system comprises item infor-
`mation file means for storing a plurality of item informa-
`tion records, each showing a state for a data element
`belonging to an item, the item information file means
`beingaccessedbyanitemindexusingapartofthe
`stored item information; relation table storing means for
`storing a relation table in which information indicating
`a relation between items is registered; relation informa-
`tion file means for storing a plurality of relation infor-
`mation records, each showing a relation between source
`and object data elements, the relation information file
`means being accessed by a relation index using a part of
`the stored relation information, and
`search means for obtaining item information for a
`data element to be searched, by accessing the item infor-
`mation file means using the item index in order to obtain
`the item information for source and object data ele-
`ments to be searched, by accessing the relation table
`storing means in order to obtain the information indicat-
`ing the relation, and by accessing the relation informa-
`tion file means using the relation index in order to obtain
`the relation information, in accordance with an input
`search instruction.
`
`In addition to the above object, it is another object of
`the present invention to provide a data management
`
`10
`
`15
`
`25
`
`30
`
`35
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`system in which the registration order of data elements
`‘ is not fixed, a data element can be registered regardless
`of an item to which the data element should belong
`when data element registration is necessary, and data
`elements can be easily related as needed.
`The data management system further comprises:
`relation information record registering means for
`forming, in accordance with an input relation informa-
`tion record registering instruction, relation information
`records from yet unrelated item information records in
`items specified by the relation information record regis-
`tering instruction, and for registering the formed rela-
`tion information records in the relation information file
`means, and
`item information record registering means for form-
`ing, when a data element is registered, an item informa-
`tion record for the data element, and for registering the
`formed item information record in the item information
`file means.
`
`It is still another object of the present invention to
`provide a search method capable of efficiently manag~
`ing a database.
`The search method comprises registering a data ele-
`ment in the database and an item information, indicating
`a registered state of the data element apd generated
`upon registration, as an item information record in an
`item information file;
`relating yet-unrelated item information records to
`one another,
`respectively belonging to designated
`items, using a relation table indicating relations between
`items to produce a relation information record indicat-
`ing a relation between data elements, resulting in regis-
`tration of the relation information record in a relation
`information file;
`.
`c searching the item information to be searched, by
`referring to the relation table, the item information file,
`and the relation information file in accordance with an
`input search instruction.
`According to the data management system of the
`present invention, the registration order of data ele-
`ments will not be fixed, and design data and the relation
`between the design data can be desirably input like a
`design memo, ing, in the design and manufacture of a
`software system. Therefore, a flexible design procedure
`which can deal with clear design data first and then
`unclear design data can be supported. When a necessary
`data element is to be searched, a search operation is
`performed by utilizing a relation table between data
`elements in addition to the relation information record
`and the item information record, so that design contra-
`diction, lack of design data, design data which requires
`further decision, and the like can be inquired and veri-
`fied. In addition, high-grade design data can be reliably
`transmitted to and reutilized for lower-grade design.
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`FIG. 1 is a block diagram of a data management
`system according to an embodiment of the present in-
`vention;
`FIG. 2 is a function block diagram for explaining the
`operation of the system shown in FIG. 1;
`FIG. 3A shows a format of an item information re-
`cord, FIG. 3B shows a format of an item identification
`in the item information record shown in FIG. 3A, and
`FIG. 3C shows a format of a relation information re-
`cord;
`FIGS. 4A to 4D are examples of item information
`records;
`
`010
`
`010
`
`

`

`3
`FIGS. 5A to 5D are examples of relation information
`records;
`FIG. 6 is a relation table used in this embodiment;
`FIG. 7 is a diagram showing the connection relation
`between items shown in FIG. 6;
`FIG. 8 is a flow chart for explaining the operation for
`creating a relation information record in the embodi-
`ment shown in FIG. 1; and
`FIGS. 9A and 9B are flow charts for explaining the
`operation for searching item information associated
`with a data element in the embodiment shown in FIG.
`1.
`
`PREFERRED EMBODIMENT OF THE
`INVENTION
`
`4,893,232
`
`5
`
`4
`and a record search section 33 for executing search
`processing using files 10 and 12 and table 14.
`For example, when a currently processed data ele-
`ment is to be registered as a part of a database, an item
`to which the data element should belong is specified. As
`shown in FIG. 33, an item ID consisting of an item
`code corresponding to the specified item and a serial
`number is generated. The data element, when the regis-
`tration is specified, is stored in accordance with the item
`10 ID. At this time, the item information record shown in
`FIG. 3A is generated and is registered in file 10.
`FIGS. 4A to 4D show item information records reg-
`istered in file 10 as described above. FIG. 4A shows an
`item information record in which a data element name is
`15 “abc”, an item code is “JK”, an item ID is “JKOODI”
`and, thereafter, item detailed information is written. For
`example, if the item code is “JK”, the item detailed
`information indicates a job for every day or for once a
`month.
`
`A data management system according to an embodi-
`ment of the present invention will be described in detail
`with reference to the accompanying drawings.
`The arrangement of a system according to this em-
`bodiment will be described with reference to FIG. 1.
`The system of this embodiment comprises a CPU 2 for
`controlling the entire system and executing search pro-
`cessing, and the like, a main memory 4 for temporarily
`storing necessary data, a magnetic disk file 6 for storing
`data elements, item information records, relation infor-
`mation records, relation table, and the like, a display
`unit 8 for displaying a searched result, and a printer 9.
`The operation of the data management system will be
`described with reference to FIG. 2, showing the char-
`acteristic features of the present invention as function
`blocks.
`Referring to FIG. 2, the data management system
`comprises two index files, i.e., an item information file
`10 for storing item information records and a relation
`information file 12 for storing relation information re-
`cords, a DDD (design data dictionary) access section 30
`for accessing files 10 and 12, and a relation table 14
`showing the relation between items to which data ele-
`ments (in this embodiment, design data) belong.
`The item information record stored in item informa-
`tion file 10 has a format shown in FIG. 3A. The item
`information record consists of a primary item index, a
`secondary item index, and item detailed information. In
`the primary index, a data element name and an item
`code corresponding to an item to which the data ele-
`ment belongs are written. In the secondary item index,
`an item identification (ID) is written. The item ID con-
`sists of the item code and a serial number, as shown in
`FIG. 3B. The serial number indicates a registration
`number when a newly input data element is registered
`in an item indicated by the corresponding item code.
`FIG. 3C shows the format of the relation information
`record stored in relation information file 12. The rela-
`tion information record consists of a primary relation
`index and relation detailed information. The primary
`relation index consists of a source item ID, a relation
`identification (ID), an object item ID, and a source item
`code. The combination of the source item 1D and the
`source item code is referred to as a secondary relation
`index.
`DDD access section 30 comprises an item informa-
`tion record registering section 31 for registering the
`item information record in item information file 10, a
`relation information record registering section 32 for 65
`reading out the item information record from file 10,
`generating a relation information record using relation
`table 14, and registering it in relation information file 12,
`
`25
`
`30
`
`20
`
`FIGS. 48 to 4D show item information records in
`which data element names are “efg”, “lmn”, and “xyz”,
`item codes are “JK”, “WS”, respectively, and “WS”,
`and item IDs are “11(0002", “WSOODI”, and “WSOOOZ”,
`respectively.
`In the item information record showq in FIG. 43,
`since one data element belonging to item “1K" has
`already been shown, the new item ID is “JK0002”. This
`also applies to the item information records shown in
`FIGS. 4C and 4D.
`The registration order of the item information re-
`cords is not predetermined in accordance with item
`codes or the relation between item information records.
`Therefore,
`the item information record is registered
`when the item ID is generated, i.e., when a data element
`35 is to be registered.
`The item information records registered in file 10 as
`described above can be read out by a primary item
`index or a secondary item index. At this time, when item
`information is to be searched using the primary item
`40 index, if its item code is unknown but a data element
`name is known, it can be searched. In this case, the item
`ID, i.e., an item code and a serial number, is treated as
`one data. This also applies to the primary and secondary
`relation indexes.
`The procedure for generating a relation information
`record using the registered item information records
`will be explained with reference to FIG. 8. In this case,
`relation table 14 stores a table shown in FIG. 6. In this
`table, source items are shown using item codes in a
`50 longitudinal column, object items are shown using item
`codes in a horizontal column, and relation IDs showing
`the relation between items are shown in columns at
`intersections of these source and object items. For ex-
`ample, if the source item code is “JK” and the object
`55 item code is “WS”, a relation ID indicating the relation
`therebetween is represented by “RR”.
`Item codes iiDD”, Silo”, G‘WSDQ, I‘PRD’, I‘PU'!’ l‘LUSS’
`I‘FL”, I‘DBII’ ‘IDA”, ISRDII, “3K”, I‘J'B‘!’ “PG’-, 5‘AR”,
`and “ND” respectively indicate a data item, I/O infor-
`60 mation, work station I/O information, print output
`information, a physical path between nodes, a logical
`path between nodes, a file, a database, a DB area, a
`record, a job function, a job, a program, an area, and a
`node.
`svrzn’ “OP”, “AC",
`Relation IDS “RR”. cs'rlu’
`“LC”, “1C”, “GS", “LN”, “v1", and ssvzn [Espa-
`tively indicate an item correlation, a record item corre-
`lation, a tree correlation, a processing operation, access
`
`45
`
`011
`
`011
`
`

`

`4,893,232
`
`5
`specification, link specification, as job step sequence, a
`graphic display sequence, line specification, a transac-
`tion volume, and a file volume.
`FIG. 7 shows the connection relations on the basis of
`the relation table shown in FIG. 6.
`A source item code, e.g., “1K", is specified in step
`812 in FIG. 8, and an object item code, e.g., “WS”, is
`specified in step 814. Thereafter, registering section 32
`refers to relation table 14 in order to check the relation
`between the input source and object item codes in step
`516. As a result, as can be seen from FIG. 6, relation ID
`“RR” is obtained as their relation.
`For example, if a source item code, e.g, “RD” is
`specified in step 812, and an object item code, e.g.,
`“WS”, is specified in step 514, since no relation therebe-
`tween is present, no relation ID is obtained in step 816.
`Step Sls determines if the relation ID is obtained If
`no relation ID is obtained, the flow advances to step
`S34: and error processing for informing a specification
`error, and the like,
`is executed. If the relation ID is
`obtained, i.e., if Y in step S18, the flow advances to step
`820.
`
`In step 820, a primary item index is obtained from the
`specified source item code, and item information file 10
`is referred to in accordance therewith. Thus, an item ID
`which includes the source item code specified in step
`S12 and is not related is obtained from the referred item
`information record.
`Since “IR” is selected as the item code, item ID
`“JK0001” is obtained first from the item information
`record shown in FIG. 4A as a source item ID. At this
`
`time, if no corresponding item ID is present, Y is ob-
`tained in step 822, and the flow advances to step S34 to
`perform error processing. If the corresponding item
`information record is present, N is obtained in step 822,
`and the flow advances to step 824.
`Since “WS” is selected as the object item code, item
`ID “W50001” is obtained first from the item informa-
`tion record shown in FIG. 4C as the object item ID.
`After execution of step S24, the same determination
`as in step 822 is made. Ifan object item ID is present, N
`is obtained in step S26, and the flow advances to step
`828. If no object item ID is present, Y is obtained in step
`S26, and the flow advances to step S34.
`A relation information record shown in FIG. 5A is
`formed based on obtained source item ID “11(0001”,
`relation ID “RR", object item ID “WSOOOI”, object
`item code “K", and the item detailed information in
`accordance with the format shown in FIG. 3C, and is
`registered in relation information file 12.
`Step S30 then determines if an unprocessed object
`item ID having item code “WS” is present with regard
`to “11(0001” read out as a current object item ID. If the
`determination in step S30 is Y, step S32 is executed.
`However, if the determination is N in step S30, step S24
`is executed again.
`In this embodiment, since item ID “W50002” corre-
`sponding to the item information record shown in FIG.
`4D is present as an unprocessed object item 1D having
`object item code “W8", step 824 is executed.
`In step S24, the item information record shown in
`FIG. 4D is read out in the same manner as described
`above, and “W50002” is obtained as an object item ID.
`Thereafter, step 528 is executed through step 826,
`and the relation information record shown in FIG. SE is
`formed and registered in file 12. At this time, since no
`unprocessed item information record having “IK” as a
`
`6
`source item code is present, Y is obtained in step S30,
`and the flow advances to step S32.
`In step 832, the same determination as in step S30 is
`made. More specifically, step 832 determines if an un-
`processed item information record having “JK” as a
`source item code is present. If it is present, N is obtained
`in step 832, and the flow returns to step 820.
`In step 520, one of the remaining specified informa-
`tion records having source item code “1K" is read out.
`In this case, the item information record shown in FIG.
`4B is read out An object item information record to be
`related to this record is read out in step S24. In this case,
`the item information record shown in FIG. 4C is read
`out. In step 528, the two item information records are
`related in the same manner as described above. As a
`result, the relation information record shown in FIG.
`5C is formed, and is registered in relation information
`file 12.
`After relating of the item information record shown
`in FIG. 4C is completed, step S30 determines if an un-
`processed object item information record is present. In
`this case, since the unprocessed item information record
`shown in FIG. 4D is present, step S24 is again executed
`to read out the item information record shown in FIG.
`4D.
`In step 330, the item information records shown in
`FIGS. 43 and 4D are related. As a result, the relation
`information record shown in FIG. 5D is formed, and is
`stored in file 12.
`Thereafter, step S32 determines if an unprocessed
`source item information record is present. Since no
`unprocessed item information record is present, Y is
`obtained in step 832, and relating processing ends.
`With the above processing, the item information re-
`cords shown in FIGS. 4A to 4D are registered in file 10,
`and the relation information records shown in FIGS.
`5A to 5D are registered in file 12.
`The above-mentioned relating processing can be de-
`sirably performed using relation table 14 as needed.
`Search processing for obtaining item information for
`searching a necessary data element will be described in
`detail with reference to FIGS. 9A and 9B.
`The search processing can be classified into (1) pro-
`cessing for obtaining object information from source
`information and a relation, (2) processing for obtaining
`source information from object information and a rela-
`tion, and (3) processing for obtaining source and object
`information from a relation.
`Processing (1) can be further classified as (l-a) search
`processing using a specific source item and a specific
`relation, (l-b) search processing using a specific source
`item and an arbitrary relation, (1-c) search processing
`using a specific source data element, and a specific rela-
`tion, and (l-d) search processing using a specific source
`data element and an arbitrary relation.
`Similarly, processing (2) can be further classified to
`(2-a) search processing using a specific object item and
`a specific relation, (2-b) search processing using a spe—
`cific object item and an arbitrary relation, (2-c) search
`processing using a specific object data element and a
`specific relation, and (2-d) search processing using a
`specific object data element and an arbitrary relation.
`The search processing using the arbitrary source
`information and the specific relation and the search
`processing using the arbitrary object information and
`the specific relation are the same as processing (3).
`The above-mentioned search processing will be ex-
`plained in detail with reference to FIGS. 9A to 9B.
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`45
`
`55
`
`65
`
`012
`
`012
`
`

`

`4,893,232
`
`7
`First, in step 840 (steps S42 to S46), search conditions
`are input, and a search instruction is generated. More
`specifically, in step S42, a source item condition is input.
`If the input item condition is not an item code but a
`source data element, it is determined whether the item
`ID of the corresponding data element has been ob-
`tained. If no item ID is obtained, an item ID corre-
`sponding to a specified data element is determined from
`file 10.
`In step S44, an object item condition is input. If the
`input item condition is not an item code but an object
`data element, it is determined whether the item ID of
`the data element has been obtained. If no item ID is
`obtained, an item ID corresponding to the specified
`data element is determined from file 10.
`In step S46, a relation is specified. In this case, it is
`determined whether a relation ID is specified. If no
`relation ID is specified, a relation ID is determined from
`relation table 14.
`In step S40, step S42 is executed for search processing
`(1) and is passed for other search processing. Step S44 is
`executed for search processing (2) and is passed for
`other processing. Step S46 is executed for processing
`(l-a), (1-b), (2-a), (2-b), and (3) and is passed for other
`processing.
`After the search conditions are determined and the
`search instruction is generated, in step S48 determines if
`the search processing using only the relation, i.e., search
`processing (3) is to be executed. If Y in step S48, the
`flow advances to step SS8. However, if N in step S48,
`step 550 is executed.
`Step 850 determines if an item ID is already specified,
`i.e., if a data element is specified. If N in step 550, the
`flow advances to step 352, and item information is ex-
`tracted using a primary item index from the input item
`code. Thereafter, an item ID is extracted from the read-
`out item information and is entered into a list in step
`$54. (This step is abbreviated as “LIST UP ITEM IDS”
`in FIG. 9A). If the data element is specified, Y is deter-
`mined in step 550, and step 854 is directly executed.
`Thereafter, step 856 is executed to determine if item
`IDs of all the data elements belonging to the specified
`item are obtained. If Y in step 356, the flow advances to
`step 562. However, if N in step 856, the flow returns to
`step 852. When the data element is specified, step 854 is
`executed once. When the item is specified, step SS4 may
`be executed a number of times.
`With the above processing, in, e.g., search processing
`(l-a) in which “1K" is specified as the source item, item
`IDs “JKOOOI” and "JK0002” are obtained. In search
`
`processing (2-a) in which “WS” is specified as the ob-
`ject item, item IDs “WSOOfll” and “WSOOOZ” are ob-
`tained.
`After the item IDs are obtained, it is determined in
`step 862 ifsearch processing (1) or (2) is to be executed.
`In the case of search processing (1), Y is determined in
`step 862 and step 564 is then executed.
`In the case of search processing (2), N is obtained in
`step 862 and step S76 is executed.
`In step 364, file 12 is accessed using the primary
`relation index from the previously obtained source item
`ID, e.g., “JK0001” and relation ID, e.g., “RR” obtained
`in step S46, thus reading out a relation information
`record. For example, the relation information record
`shown in FIG. 5A is first read out, and “WSOODI” is
`obtained as an object item ID. The obtained object item
`IDs are entered into a list in step S“. Thereafter, the
`completion of processing is checked in step 868. If Y in
`
`8
`step 568, step S70 is executed. However, if N in step
`568, step 564 is executed again, and a next object item
`ID, e.g., “WSOOOZ” is obtained. In this manner, object
`item IDs “WSW!” and “WSOOOZ” are obtained in
`correspondence with source item ID “JKOOOI”.
`After detection of the object item IDs, file 10 is ac-
`cessed using the secondary item index in accordance
`with the obtained object item IDs, thus reading out an
`item information record. For example, the item infor-
`mation record shown in FIG. 4C is read out in accor-
`dance with object item ID “WSOOOI”. The readout item
`information records are entered into a list in step S72.
`Thereafter, in step $74, the completion of the process-
`ing is checked. If Y in step S74, the search processing
`ends. If N in step S74, step S70 is executed again and the
`item information record shown in FIG. 4D is obtained
`with reference to object item ID “W50002”.
`In this manner, search processing (1) ends. In the case
`of search processing (1-b), i.e., when a relation is not
`specified, in step S64, file 12 can be accessed using only
`the item ID without using a relation ID when search
`processing is performed using the primary relation in-
`dex.
`In the case of search processing (l-c), i.e., when a
`data element, not an item, is specified, Y 9 determined
`in step 550.
`When the source item information is to be obtained
`from the object item ID, the processing is performed
`using the object
`item IDs, e.g.,
`“WSOOOI” and
`“WSDOOI” obtained in step 554.
`In step 876, the relation information record is read
`out using the secondary relation index in accordance
`with the object item ID. The relation information re-
`cord shown in FIG. 5A is read out in accordance with
`object item ID “W50001”.
`Step S78 then determines if a relation is specified. If Y
`in step 878, step 580 is executed. If N in step S78, step
`882 is executed. Step $80 determines if the relation ID
`in the relation information record read out in step S76
`coincides with a specified relation ID. If “RR” was
`specified as the relation ID, since the relation ID shown
`in FIG. 5C is “RR”, Y is obtained in step 580, and step
`582 is executed. If N in step 580, i.e., if the specified
`relation ID does not coincide with the relation ID in the
`readout relation information record, the readout rela-
`tion information record is ignored, and step S76 is exe-
`cuted again.
`In step 582, source item IDs are entered into a list.
`For example, source item ID “JK0001” corresponds to
`object item ID “WSDDOI”.
`Thereafter,
`the completion of the processing is
`checked in step SS4. If Y in step 584, step S70 is exe-
`cuted. If N in step SS4, step S76 is executed again.
`In addition, source item ID “JK0002” is obtained in
`correspondence with source item ID “WSOOOI”, and
`source item IDs “JK0001” and “JKOOO2” are obtained
`in correspondence with item ID “WSOOOZ”, in the same
`manner as described above.
`The same processing is performed after step S70. In
`this manner, search processing (2) is executed. In the
`case of search processing (2-a), i.e., if a relation is speci-
`fied, N is determined in step S78. If an object data ele—
`ment is specified, Y is obtained in step 850.
`If Y in step S48, i.e., if only the relation is specified,
`relation information is read out using a primary relation
`index in accordance with the relation ID. In step 560,
`source item IDs and object item IDs are listed up from
`the readout informa