Method for deriving data mappings and data aliases5710917Abstract This invention describes a computer application system and method for automatically deriving data mappings by processing stored data mappings. Derived data mappings are system generated data mappings. The system comprises a plurality of stored data mappings, a data mapping report generator, and a data mapping tool. Heretofore, data mappings were created by human analysis of data from two or more sources to determine the relationship between data fields. This time consuming process has been eliminated by the present invention. The derived data mappings may be stored for later use or provided to other system programs. The derivation may be performed at various levels of abstraction. Derived data mappings that should not be used are also identified. Claims What is claimed is: Description DESCRIPTION
TABLE 1A
__________________________________________________________________________
STORED MAPPING DATA FOR THE EXAMPLE IN FIG. 5
Date Group
Data Field
File Name
Name Name Data Type
Length
Description
__________________________________________________________________________
EEADDR EENUM NUMERIC
6 Employee Number
EENAM EENAML
ALPHA 10 Employee Last Name
EENAMF
ALPHA 8 Employee First Name
EENAMM
ALPHA 1 Employee Middle Initial
EEPHON
NUMERIC
5 Employee Telephone
EEEML ALPHANUM
20 Employee E-Mail ID
EEHOM EEHOMR
ALPHANUM
25 Employee Home Street
EEHOMC
ALPHA 15 Employee Home City
EEHOMA
ALPHA 2 Employee Home State
EEHOMZ
NUMERIC
5 Employee Home Zip
EETEL NUMERIC
10 Employee Home Telephone
EEPOSN EENUMB
NUMERIC
7 Employee Number
EELNAM
ALPHA 15 Employee Last Name
EEMIDD
UC ALPHA
1 Employee Middle Initial
EEFNAM
ALPHA 8 Employee First Name
EEORGN
NUMERIC
2 Organization Code
EEDIVN
NUMERIC
3 Division Code
EEDEPT
NUMERIC
2 Department Code
EESECT
NUMERIC
1 Section Code
EESUPR
NUMERIC
7 Supervisor's Number
__________________________________________________________________________
It often becomes necessary to move data from one database to another or to one set of files within a single database to a different set of files. Such changes can be necessitated in several common circumstances. For example, a corporation might restructure its internal organization in such a way that employees are no longer classified by organization, division, department and section, but only by using the organization, division and department categories. Another similar situation would be when changes in the U.S. Postal Service classification system might require that five digit zip codes be converted to nine digit zip codes. Further, growth in the size of the organization may require that the field width of the employee identification number be increased in size. Yet another example might be changes in federal, state and local tax regimes requiring that state and local taxes be withheld where only federal taxes had earlier been withheld. Changes in technology might require that the employee directory also contain the facsimile and beeper numbers of employees. In yet a different circumstance, some systems within an applications suite may store the same data as other systems but the data stored within the two systems may not be identical because of failure to update one of the databases. Changes in storage format may also be necessitated by changes in the computation algorithm used by the search and update engine or by a requesting system. Changes in storage format may also come about due to upgradation of a system within an application suite, or of an entire applications suite or of the entire computing environment itself. FIG. 6 illustrates an exemplary mapping from one storage format to another. In FIG. 6 some fields from the files EEWITHLD 411, EEPAYCK 410 and EEADDR 403 are mapped to a new field PAYROLL 601. Likewise some fields from the files EEADDR 403 and EEPOSN 404 are mapped to a new file DIRECTRY 602. Thus, the new file PAYROLL 601 comprises an employee identification number field 603, a social security number field 604, a home address data group 605, an annual pay rate field 606, number of payments per year field 607, a filing status field 608, an allowances field 609 and a year-to-date withholding data group 610. The home address data group 605 is further comprised of a street address field 611, a city name field 612, a state name field 613 and a nine digit zip code field 614. The year-to-date withholding field 610 is further comprised of a federal income tax withholding field 615, a FICA withholding field 616, a medicare withholding field 617, a state tax withholding field 618, and a local tax withholding tax field 619. The mapping of the data in files EEWITHLD 411, EEPAYCK 410 and EEADDR 403 to PAYROLL 601 is shown by the directional arrows 637 through 648. Some of these directional arrows carry the letter "T" on them denoting that the process of mapping involves translation from one code to another. For example, the EEPAY field 525 is mapped to the annual pay rate field 606 by the mapping relationship 638. This may, for example, result from the use of different bases in the payroll information database. The EEPAY field 525 contained a value corresponding to the pay received by an employee in one pay period whereas the Annual Pay Rate field 606 contains the annual compensation of the employee. The mapping 639 of the EEPERD 526 to the Number of Payments Per Year 607 provides another example of when a translation may be required to map the data from an old storage format to another. The EEPERD 526 may have stored the data in the form of alphabetic code such as 'M' for monthly, 'S' for semi-monthly, 'B' for biweekly, 'W' for weekly, etc. In contrast, the Number of Payments Per Year field 607 may contain a numeric value indicating the number of pay periods per year. Another kind of translation is illustrated by mapping relationship 644 which maps the EENUM field 501 of file EEADDR 403 to the Employee ID field 603 of PAYROLL file 601. Such a translation may result from an increase in the size of the firm or a conversion of one employee number to a different employee number due to the acquisition of an organizational group. Sometimes data fields are also concatenated in addition to being translated. This is shown, for example, by mapping relationship 657 which maps the EEDEPT field 520 and EESECT field 521 of EEPOSN file 404 to the department location code 633 of the DIRECTRY file 602. FIG. 6 also shows mapping relationships 649 through 657 detailing the relationship between the old files EEADDR 403 and EEPOSN file 404 and the new DIRECTRY file 602. As shown in FIG. 6 the DIRECTRY file 602 is comprised of multiple data fields 620, 622, 623, 624 and 627 and data groups 621, 625 and 626 which are further comprised of data fields 628 through 636. As can be seen from the above example, even a simple data mapping from four old files to two new files is a very complex task and consequently very tedious and somewhat error prone. The task is further complicated by the fact that multiple files may contain the same or similar information in different fields and sometimes in different internal representation formats. FIG. 7 shows the various high-level steps involved in migrating data from one storage format to another. The process starts with the analysis of the prior data organization as shown in step 701. This is followed by the data map that identifies the application system, the data group and the data field of the prior data storage format and relates it to the new data organization to be created, as shown by step 702. The next step after the preparation of the data map is the design construction, testing and debugging of computer programming instructions to effectuate the data migration. This is shown by step 703 of FIG. 7. After the computer code has been debugged and tested, it is used to translate and reorganize the data into the desired new storage format as shown by step 704. Finally, the new data organization is tested to make sure that it satisfies the design objectives. This is shown in step 705 of FIG. 7. As can be seen from FIG. 6, the data mapping process is an inherently complex one. Further, data mapping may be done on various data bases within a computer system at differing times. While the maps of the data stored within various databases is usually retained within the computer system, prior systems have not attempted to reuse stored data maps to automatically derive newer data maps. An example illustrating the derivation of new data maps from prior data maps is illustrated in FIG. 8. If a prior data mapping effort had equated to the field POTYPE in file A to the field POSTAT in file B, and on a different occasion, the field POSTAT in file B had been translated to the field POFLAG in file C, and field POIND in file D had been equated the field POSTAT in file B, the three different mappings can be depicted as shown in FIG. 8. The mapping of file A 801 to file B 802, file B 802 to file C 803, and file D 804 to file B 802 is shown by mapping relationships 805, 806 and 807, respectively. As can be seen in FIG. 8, mapping relationship 805 essentially uses an identity algorithm as shown by the algorithm box 809. Likewise, the mapping relationship 806 essentially consists of an identity algorithm 812. However, the mapping relationship 807 consists of a translation algorithm 815. In the example illustrated, if the value of the POSTAT field is one in file B, it is translated to a POFLAG value of four in file C. Prior data mapping techniques have required that any new data mapping, even if between the same fields of the same files, would have required the manual creation of new data maps. However, as illustrated in FIG. 8, one can automatically derive a mapping relationship 817 to translate the POTYPE field of file A to the POFLAG of file C. This automatically derivable mapping relationship is shown by the broken double-sided arrow line 817. Similarly, the mapping relationships between the POTYPE field of file A and the POIND field of file D is also automatically derivable as shown by the broken double-sided arrow 818. In a similar manner, the relationship between POIND field of file D and the POFLAG field of file C is also automatically derivable as shown by the broken double-sided arrow 819. It is to be noted that the derivation of a mapping relationship can be bi-directional unless one of the mappings involved concatenation or combination of data fields or elements in a manner that makes it impossible to undue the transformation. Some illustrative uses of data map derivation are shown in FIG. 9. A map derivation tool 904 takes as its input prior data maps 901, 902 and 903 and derives a new data map 905. In the example illustrated, box 901 shows a pre-existing data map from file A to file C and box 903 shows a pre-existing data map from file D to file B. The map derivation tool 904 is also capable of accepting an earlier derived data map such as the one illustrated from file C to file D in box 902. The map derivation tool 904 can derive a data map from file A to file B as shown in box 905. The derived data map can serve several uses as shown in FIG. 9. The derived data map can be an input to a bridge code generating tool 906. Such a bridge code generation tool creates specific computer programming instructions to bridge data from System A 910 to System B 911. The derived data map may also be used as an input to standardized data bridging tool 907 which can also accomplish the task of bridging data from System A 910 to System B 911. Derived data maps may also be fed into a report generator for the use of end users. The maps may also be used to create data aliases and to cross reference various data fields, as shown in box 908. The present invention also automatically derives data aliases. Data aliases are data maps that identify differing fields as containing the same substantive information. Thus, in the example shown in FIG. 8, the POTYPE field of file A is an alias for field POSTAT in file B, the field POFLAG in file C and the field POIND in file D. Table 2 shows the data alaises for each of the four data field names used in the example accompanying FIG. 8.
TABLE 2
______________________________________
DATA ALIASES
DATA FIELD DATA ALIASES
______________________________________
A. POTYPE B. POSTAT, C. POFLAG, D. POIND
B. POSTAT A. POTYPE, C. POFLAG, D. POIND
C. POFLAG A. POTYPE, B. POSTAT, D. POIND
D. POIND A. POTYPE, B. POSTAT, C. POFLAG
______________________________________
A high level description of the processing of deriving data maps from other stored data maps is illustrated in FIG. 10. The process starts as shown by circle 1001 by taking data maps stored within the computer system as shown in box 1002 and generating two sets of transient maps 1003 and 1004. Transient maps 1003 consists of the original data maps concatenated with the reversed data maps. Transient maps 1004 consists of the reversed data maps concatenated with the original data maps. Transient maps 1003 and 1004 are fed into a map derivation tool 1005 that generates derived maps by forward chaining of transient map 1003 and reverse chaining of transient maps 1004 as shown in boxes 1006 and 1007, respectively. The derived maps generated by the map derivation tool can then be stored with the pre-existing stored maps 1002. If storage of the derived data maps is not desired, the derived maps are discarded as shown in circle 1010. The derived maps generated by this process can be used as inputs to a requirements document generator 1011 or a data movement tool 1012 or to a code generator 1013. After one or more of these applications terminates, the process ends as shown by circle 1014. It should be noted that the map derivation tool 1006 performs the forward chaining only on the original data maps of the transient map set 1003. Likewise, the map generation tool 1005 performs the reverse chaining procedure 1007 only on the reversed data maps of the transient maps set 1004. The map derivation process is shown in greater detail in algorithmic form in FIG. 11. The process consists of a first pass 1102 and a second pass 1103 on two sets of transient maps. The first pass 1102 is further comprised of the steps of reading in the original data maps listing which contains the system names, the data group names and the data field names for source and target data as shown in step 1105. Next the source and the target data are transposed to create reversed maps shown in step 1106. Next, for the target field of each of the original data maps, the remaining field-to-field maps for instances where the source field is the same as the target field, whenever this search is successful in finding such a source field, the located map is termed to be an alias for the original data map, as shown in step 1107. All such aliases are listed as a source field for the target field of the original data map and also as an additional search argument for potential use as a target, as shown in step 1108. First pass 1102 by forward chaining, generates one set of derived maps. The second pass 1103 generates another set of derived maps based upon the reverse chaining of the map list. This is also illustrated in FIG. 11. The second pass 1003 is very similar to the steps involved in the first pass. It starts by reading in the original data map listings which detail the system names, the data group names and the data field names for both source and target data. This is shown in step 1109. Next, one creates reversed maps by transposing the source and the target data, as shown in step 1110. Thereafter, one reads in the original data maps again as shown in steps 1111. The concatenation of the reversed maps set and the original data maps produces a second set of transient maps set. Using this transient maps set as input, the map derivation tool searches for all aliases of the target field for each map of the reversed data map set. This is done by searching the remaining field-to-field maps for instances where the source field is the same as the target field for which the search is being done. This is shown in step 1112. If the search is successful in locating any aliases for the target field of each of the reversed data maps, these aliases are added as a source field for the target field and also as additional search arguments for potential use as a target, as shown in step 1113. After the second pass, the operation of the map derivation tool ends as shown by circle 1103. The map derivation process will be better understood by considering an illustrative example. Consider a situation where four Application Systems, CUTCHEX, APBILL, REQUEST and BILLR consist of 29 data fields organized into nine data groups as shown in Table 3 below.
TABLE 3
______________________________________
DATA ORGANIZATION EXAMPLE
Application System
Data Group
Date Field
______________________________________
CUTCHEX Account A
B
C
Check D
E
F
G
Bill H
I
J
APBILL Z7ACCT K
L
M
Z7XMPO N
O
P
Q
Z6BCDF R
S
T
REQUEST MIRQ001 U
V
W
MIRQ002 X
Y
Z
BILLR INVOICE AA
BB
CC
______________________________________
Data maps are constructed in the first instance to link various fields together based upon their business definition. Data maps also describe any transformation that takes place when specific values of one field are copied or modified or combined before storage in another field. As already explained, stored mapping data in a system may combine prior manual mapping efforts with prior derived data maps. An illustrative mapping is detailed in Table 4 and depicted in FIG. 12.
TABLE 4
__________________________________________________________________________
STORED MAPPING DATA
From .fwdarw.
To
S. Application Application Data Aggregate
No.
System
Data Group
Data Field
System
Data Group
Field
Algorithm
d
__________________________________________________________________________
1 APBILL
Z6BCDF
R .fwdarw.
CUTCHEX
BILL H Substring
Y
2 APBILL
Z6BCDF
R .fwdarw.
CUTCHEX
BILL I Substring
Y
3 APBILL
Z6BCDF
R .fwdarw.
CUTCHEX
BILL J Substring
Y
4 CUTCHEX
ACCOUNT
A .fwdarw.
APBILL
Z7ACCT
K Equate
N
5 CUTCHEX
ACCOUNT
B .fwdarw.
APBILL
Z7ACCT
M Equate
N
6 CUTCHEX
ACCOUNT
C .fwdarw.
APBILL
Z7XMPO
N Concatena
N
te
7 CUTCHEX
CHECK D .fwdarw.
APBILL
Z7XMPO
N Concetena
N
te
8 CUTCHEX
CHECK F .fwdarw.
APBILL
Z7XMPO
O Substring
N
9 CUTCHEX
CHECK F .fwdarw.
APBILL
Z7XMPO
P Substring
N
10 REQUEST
MIRQ001
U .fwdarw.
APBILL
Z7ACCT
K Equate
N
11 REQUEST
MIRQ001
U .fwdarw.
BILLR INVOICE
BB Equate
N
12 REQUEST
MIRQ001
V .fwdarw.
APBILL
Z7ACCT
L Equate
N
13 REQUEST
MIRQ001
V .fwdarw.
BILLR INVOICE
AA Equate
N
14 REQUEST
MIRQ001
W .fwdarw.
APBILL
Z7ACCT
M Equate
N
15 REQUEST
MIRQ002
X .fwdarw.
APBILL
Z6BCDF
R Concatena
Y
te
16 REQUEST
MIRQ002
Y .fwdarw.
APBILL
Z6BCDF
R Concatena
Y
te
17 REQUEST
MIRQ002
Z .fwdarw.
APBILL
Z6BCDF
R Concetena
Y
te
__________________________________________________________________________
Thus, field R of Data Group Z6BCDF of Application System APBILL is mapped to Fields H, I and G of Data Group BILL of Application System CUTCHEX using a concatenation algorithm. Data Fields A and B of Data Group ACCOUNT of Application System CUTCHEX are equated to Data Fields K and M of Data Group Z7ACCT of Application System APBILL, respectively. Data Field C of Data Group ACCOUNT and Data Field D of Data Group CHECK, both of Application System CUTCHEX, are concatenated to Data Field N of Data Group Z7XMPO of Application System APBILL. Data Field F of Data Group CHECK of Application System CUTCHEX is subdivided to Data Fields O and P of Data Group Z7XMPO of Application System APBILL. Data Fields U, V and W of Data Group MIRQ001 of Application System REQUEST are equated to Data Fields K, L and M of Data Group Z7ACCT of Application System APBILL, respectively. Data Fields U and V of Data Group MIRQ001 of Application System REQUEST are also mapped to Data Fields BB and AA of Data Group INVOICE of Application System BILLR. Finally, Data Fields X, Y and Z of Data Group MIRQ002 of Application System REQUEST are concatenated to map to Data Field R of Data Group Z6BCDF of Application System APBILL. Starting with the stored mapping data one can derive maps at three different levels: at the System Level, at the Data Group Level and at the Data Field Level. Even though the mapping of one data field to another is usually directional, the derivation can be both in the forward direction as well as in the reverse direction. The derivation process is thus nondirectional. System Level Derived Maps The previous mapping example will produce the following distinct system level maps, both forward and reversed:
______________________________________
First Pass Second Pass
______________________________________
Original Section Reversed Section
APBILL .fwdarw.
CUTCHEX CUTCHEX .fwdarw.
APBILL
CUTCHEX .fwdarw.
APBILL APBILL .fwdarw.
CUTCHEX
REQUEST .fwdarw.
APBILL APBILL .fwdarw.
REQUEST
REQUEST .fwdarw.
BILLR BILLR .fwdarw.
REQUEST
Reversed Section Original Section
CUTCHEX .fwdarw.
APBILL APBILL .fwdarw.
CUTCHEX
APBILL .fwdarw.
CUTCHEX CUTCHEX .fwdarw.
APBILL
APBILL .fwdarw.
REQUEST REQUEST .fwdarw.
APBILL
BILLR .fwdarw.
REQUEST REQUEST .fwdarw.
BILLR
______________________________________
Since the first pass and the second pass are very similar in structure, the forward chaining and the reverse chaining processes may be executed recursively using the same code. During the first pass, the original section of the stored map is processed and all subsequent system level maps are scanned for potential derivations. Duplicative derivations are then removed using database return codes. In this example, the first pass would produce the following derivation:
______________________________________
Derivation Comments
______________________________________
CUTCHEX .fwdarw.
REQUEST CUTCHEX is mapped to APBILL in
the original section which is mapped to
REQUEST in the reversed section.
REQUEST .fwdarw.
CUTCHEX REQUEST is mapped to APBILL in
the original section which is mapped to
CUTCHEX in the reversed section.
______________________________________
During the second pass, the reversed section precedes the original section. For each entry in the reversed section, all subsequent system level maps are scanned for potential derivations. As before, duplicative derivations are removed using database written codes. This process results in the production of the following derivations:
______________________________________
Derivation Comments
______________________________________
APBILL .fwdarw.
BILLR APBILL is mapped to REQUEST in the
reversed section which is mapped to
BILLR in the original section.
BILLR .fwdarw.
APBILL BILLR is mapped to REQUEST in the
reversed section which is mapped to
APBILL in the original section.
______________________________________
Data Group Level Derived Maps The data group level mapping derivations are produced using the same technique just described. From the original mapping example, the following distinct set of data group maps, forward and reversed, may be produced and is shown below:
______________________________________
From .fwdarw.
To
Application Application
System Data Group System Data Group
______________________________________
Original
APBILL Z6BCDF .fwdarw.
CUTCHEX BILL
CUTCHEX ACCOUNT .fwdarw.
APBILL Z7ACCT
CUTCHEX ACCOUNT .fwdarw.
APBILL Z7XMPO
CUTCHEX CHECK .fwdarw.
APBILL Z7XMPO
REQUEST MIRQ001 .fwdarw.
APBILL Z7ACCT
REQUEST MIRQ001 .fwdarw.
BILLR INVOICE
REQUEST MIRQ002 .fwdarw.
APBILL Z6BCDF
Reversed
CUTCHEX BILL .fwdarw.
APBILL Z68CDF
APBILL Z7ACCT .fwdarw.
CUTCHEX ACCOUNT
APBILL Z7XMPO .fwdarw.
CUTCHEX ACCOUNT
APBILL Z7XMPO .fwdarw.
CUTCHEX CHECK
APBILL Z7ACCT .fwdarw.
REQUEST MIRQ001
BILLR INVOICE .fwdarw.
REQUEST MIRQ001
APBILL Z6BCDF .fwdarw.
REQUEST MIRQ002
______________________________________
The two-pass map derivation process is executed to produce the following derivations:
______________________________________
From .fwdarw.
To
Ser. Application Application
No. System Data Group System Data Group
______________________________________
1 APBILL .fwdarw.
Z7ACCT .fwdarw.
BILLR INVOICE
2 BILLR .fwdarw.
INVOICE .fwdarw.
APBILL Z7ACCT
3 CUTCHEX .fwdarw.
ACCOUNT .fwdarw.
BILLR INVOICE
4 CUTCHEX .fwdarw.
ACCOUNT .fwdarw.
CUTCHEX CHECK
5 CUTCHEX .fwdarw.
ACCOUNT .fwdarw.
REQUEST MIRQ001
6 CUTCHEX .fwdarw.
BILL .fwdarw.
CUTCHEX BILL
7 CUTCHEX .fwdarw.
BILL .fwdarw.
REQUEST MIRQ002
8 CUTCHEX .fwdarw.
CHECK .fwdarw.
CUTCHEX ACCOUNT
9 CUTCHEX .fwdarw.
MIRQ001 .fwdarw.
CUTCHEX ACCOUNT
10 REQUEST .fwdarw.
MIRQ002 .fwdarw.
CUTCHEX BILL
11 REQUEST .fwdarw.
MIRQ002 .fwdarw.
REQUEST MIRQ002
______________________________________
Data Field Level Derived Maps As before, the same two-pass procedure is executed, but against all segments of the entire stored map. In this instance, unlike in the two previous higher-level derivations, the aggregation or disaggregation of various data fields must now be taken into account. The original mappings and the reversed mappings may be tabulated as shown below:
__________________________________________________________________________
From .fwdarw.
To
S. Application Application Data Aggregate
No.
System
Data Group
Data Field
System
Data Group
Field
Algorithm
d
__________________________________________________________________________
1 APBILL
Z6BCDF
R .fwdarw.
CUTCHEX
Bill H Substring
Y
2 APBILL
Z6BCDF
R .fwdarw.
CUTCHEX
Bill I Substring
Y
3 APBILL
Z6BCDF
R .fwdarw.
CUTCHEX
Bill J Substring
Y
4 CUTCHEX
ACCOUNT
A .fwdarw.
APBILL
Z7ACCT
K Equate
N
5 CUTCHEX
ACCOUNT
B .fwdarw.
APBILL
Z7ACCT
M Equate
N
6 CUTCHEX
ACCOUNT
C .fwdarw.
APBILL
Z7XMPO
N Concatenate
N
7 CUTCHEX
CHECK D .fwdarw.
APBILL
Z7XMPO
N Concatenate
N
8 CUTCHEX
CHECK F .fwdarw.
APBILL
Z7XMPO
O Substring
N
9 CUTCHEX
CHECK F .fwdarw.
APBILL
Z7XMPO
P Substring
N
10 REQUEST
MIRQ001
U .fwdarw.
APBILL
Z7ACCT
K Equate
N
11 REQUEST
MIRQ001
U .fwdarw.
BILLR INVOICE
BB Equate
N
12 REQUEST
MIRQ001
V .fwdarw.
APBILL
Z7ACCT
L Equate
N
13 REQUEST
MIRQ001
V .fwdarw.
BILLR INVOICE
AA Equate
N
14 REQUEST
MIRQ001
W .fwdarw.
APBILL
Z7ACCT
M Equate
N
15 REQUEST
MIRQ002
X .fwdarw.
APBILL
Z6BCDF
R Concatenate
Y
16 REQUEST
MIRQ002
Y .fwdarw.
APBILL
Z6BCDF
R Concatenate
Y
17 REQUEST
MIRQ002
Z .fwdarw.
APBILL
Z6BCDF
R Concatenate
Y
18 CUTCHEX
BILL H .fwdarw.
APBILL
Z6BCDF
R Concatenate
Y
19 CUTCHEX
BILL I .fwdarw.
APBILL
Z6BCDF
R Concatenate
Y
20 CUTCHEX
BILL J .fwdarw.
APBILL
Z6BCDF
R Concatenate
Y
21 APBILL
Z7ACCT
K .fwdarw.
CUTCHEX
ACCOUNT
A Equate
N
22 APBILL
Z7ACCT
M .fwdarw.
CUTCHEX
ACCOUNT
B Equate
N
23 APBILL
Z7XMPO
N .fwdarw.
CUTCHEX
ACCOUNT
C Substring
N
24 APBILL
Z7XMPO
N .fwdarw.
CUTCHEX
CHECK D Substring
N
25 APBILL
Z7XMPO
O .fwdarw.
CUTCHEX
CHECK F Concatenate
N
26 APBILL
Z7XMPO
P .fwdarw.
CUTCHEX
CHECK F Concatenate
N
27 APBILL
Z7ACCT
K .fwdarw.
REQUEST
MIRQ001
U Equate
N
28 BILLR INVOICE
BB .fwdarw.
REQUEST
MIRQ001
U Equate
N
29 APBILL
Z7ACCT
L .fwdarw.
REQUEST
MIRQ001
V Equate
N
30 BILLR INVOICE
AA .fwdarw.
REQUEST
MIRQ001
V Equate
N
31 APBILL
Z7ACCT
M .fwdarw.
REQUEST
MIRQ001
W Equate
N
32 APBILL
Z6BCDF
R .fwdarw.
REQUEST
MIRQ002
X Substring
Y
33 APBILL
Z6BCDF
R .fwdarw.
REQUEST
MIRQ002
Y Substring
Y
34 APBILL
Z6BCDF
R .fwdarw.
REQUEST
MIRQ002
Z Substring
Y
__________________________________________________________________________
It should be noted that the above list of stored mappings comprises of both the original maps (shown as serial numbers 1 through 17) and the reversed maps (shown as serial numbers 18 through 34). As before, duplicative derivations produced during the derivation process may be eliminated using database return codes. The following data field level maps would be derived in the above example:
__________________________________________________________________________
From .fwdarw.
To
S. Application Application Data
No.
System
Data Group
Data Field
System
Data Group
Field
__________________________________________________________________________
1 APBILL
Z7ACCT
K .fwdarw.
BILLR INVOICE
BB
2 APBILL
Z7ACCT
L .fwdarw.
BILLR INVOICE
AA
3 BILLR INVOICE
AA .fwdarw.
APBILL
Z7ACCT
L
4 BILLR INVOICE
BB .fwdarw.
APBILL
Z7ACCT
K
5 CUTCHEX
ACCOUNT
A .fwdarw.
BILLR INVOICE
BB
6 CUTCHEX
ACCOUNT
A .fwdarw.
REQUEST
MIRQ001
U
7 CUTCHEX
ACCOUNT
B .fwdarw.
REQUEST
MIRQ001
W
8 CUTCHEX
ACCOUNT
C .fwdarw.
CUTCHEX
CHECK D
9 CUTCHEX
BILL H .fwdarw.
CUTCHEX
BILL I
10 CUTCHEX
BILL H .fwdarw.
CUTCHEX
BILL J
11 CUTCHEX
BILL H .fwdarw.
REQUEST
MIRQ002
X
12 CUTCHEX
BILL H .fwdarw.
REQUEST
MIRQ002
Y
13 CUTCHEX
BILL H .fwdarw.
REQUEST
MIRQ002
Z
14 CUTCHEX
BILL I .fwdarw.
CUTCHEX
BILL H
15 CUTCHEX
BILL I .fwdarw.
CUTCHEX
BILL J
16 CUTCHEX
BILL I .fwdarw.
REQUEST
MIRQ002
X
17 CUTCHEX
BILL I .fwdarw.
REQUEST
MIRQ002
Y
18 CUTCHEX
BILL I .fwdarw.
REQUEST
MIRQ002
Z
19 CUTCHEX
BILL J .fwdarw.
CUTCHEX
BILL H
20 CUTCHEX
BILL J .fwdarw.
CUTCHEX
BILL I
21 CUTCHEX
BILL J .fwdarw.
REQUEST
MIRQ002
X
22 CUTCHEX
BILL J .fwdarw.
REQUEST
MIRQ002
Y
23 CUTCHEX
BILL J .fwdarw.
REQUEST
MIRQ002
Z
24 CUTCHEX
CHECK D .fwdarw.
CUTCHEX
ACCOUNT
C
25 REQUEST
MIRQ001
U .fwdarw.
CUTCHEX
ACCOUNT
A
26 REQUEST
MIRQ001
W .fwdarw.
CUTCHEX
ACCOUNT
B
27 REQUEST
MIRQ002
X .fwdarw.
CUTCHEX
BILL H
28 REQUEST
MIRQ002
X .fwdarw.
CUTCHEX
BILL I
29 REQUEST
MIRQ002
X .fwdarw.
CUTCHEX
BILL J
30 REQUEST
MIRQ002
X .fwdarw.
REQUEST
MIRQ002
Y
31 REQUEST
MIRQ002
X .fwdarw.
REQUEST
MIRQ002
Z
32 REQUEST
MIRQ002
Y .fwdarw.
CUTCHEX
BILL H
33 REQUEST
MIRQ002
Y .fwdarw.
CUTCHEX
BILL I
34 REQUEST
MIRQ002
Y .fwdarw.
CUTCHEX
BILL J
35 REQUEST
MIRQ002
Y .fwdarw.
REQUEST
MIRQ002
X
36 REQUEST
MIRQ002
Y .fwdarw.
REQUEST
MIRQ002
Z
37 REQUEST
MIRQ002
Z .fwdarw.
CUTCHEX
BILL H
38 REQUEST
MIRQ002
Z .fwdarw.
CUTCHEX
BILL I
39 REQUEST
MIRQ002
Z .fwdarw.
CUTCHEX
BILL J
40 REQUEST
MIRQ002
Z .fwdarw.
REQUEST
MIRQ002
X
41 REQUEST
MIRQ002
Z .fwdarw.
REQUEST
MIRQ002
Y
__________________________________________________________________________
Consider derived maps 8 and 24. Although the fields CUTCHEX. ACCOUNT. C and CUTCHEX. CHECK. D were concatenated to field APBILL. Z7XMPO. N, the field APBILL. Z7XMPO. N was not considered to be an aggregation by business persons because it is a field whose components are never mentioned separately. Consequently, a derivation that considers fields CUTCHEX. ACCOUNT. C and CUTCHEX. CHECK. D to be equivalent in meaning would be acceptable to a business user. Consider also derived maps 9-23 and 27-41. These derivations resulted from a linkage between fields X, Y and Z to fields H, I and J, respectively through field R. Since field R is composed of three distinct pieces of information which can be separately described and used, it is considered to be an aggregation in business usage. Consequently, the resulting derivations are not worthwhile and should be discarded. Examples of some common mapping algorithms are shown in Table 5.
TABLE 5
__________________________________________________________________________
COMMON PROCESSING RULES
S. No.
Routine Name
Argument
Comments
__________________________________________________________________________
1 EQ A straight move from the CPS field to the DRM
field
2 NA No action required in the extract programs.
This data is available from another source.
3 DATE1INC
convert from
Use the common date conversion routine that
dec 7, yymmdd
exists in the OSIF include data set. The
name of the routine (file) is DATE1INC. This
source code is included in the extract
routine source with %INCLUDE.
4 DATE2INC
convert from
Use the common date conversion routine that
dec 8, yymmdd
exists in the OSIF include dataset. The name
of the routine (file) is DATE2INC. This
source code is included in the extract
routine source with %INCLUDE.
5 DATE3INC
char 6,
Use the common date conversion routine that
yymmdd exists in the OSIF include dataset. The name
of the routine (file) is DATE3INC. This
source code is included in the extract
routine source with %INCLUDE.
6 DATE4INC
char 8,
Use the common date conversion routine that
mm/dd/yy
exists in the OSIF include dataset. The name
of the routine (file) is DATE4INC. This
source code is included in the extract
routine source with %INCLUDE.
7 TIME1INC
fixed dec 7,
Use the common time conversion routine that
hhmmss exists in the OSIF include dataset. The name
of the routine (file) is TIME2INC. This
source code is included in the extract
routine source with %INCLUDE.
8 TIME2INC
convert from
Use the common time conversion routine that
hh:mm:ss
exists in the OSIF include data set. The
name of the routine (file) is TIME2INC. This
source code is included in the extract
routine source with %INCLUDE.
9 SUBSTR x,y Take the substring beginning in position x
for a length of y.
10 BIT1YON If the source field has a value of 1, then
make the target field = `Y`, if the source
field has a value of 0, then make the target
field = `N`.
11 BIT1YONBB If the source field has a value of 1, then
make the target field = `Y`, if the source
field has a value of 0, then make the target
field = `N`, if the source field has a value
of spaces or null, then make the target field
= `N`.
12 REPGRP This field is a repeating group on the source
file. It will probably have a suffix number
on the source field name. Each of these
source field names will be used to create a
new output record.
13 MS X,Y The source filed is moved into the BDW field
starting at position `x` for a length of `y`.
14 CONSTANT
X The data that will be moved into the BDW
field is a constant value `x`.
15 RCDEXINC Use the common date conversion routine that
exists in the OSIF include dataset. The name
of the routine (file) is RCDEXINC.
16 COND.sub.-- EQ.sub.-- 01
If there is no XAE00237 records for a
particular ORDER.sub.-- LINE (XAE00221) then use the
XAE00221 data to populate POLNACCT otherwise,
use the XAE00237 data to populate POLNACCT.
__________________________________________________________________________
Although a preferred embodiment of the method and apparatus of the present invention has been illustrated in the accompanying drawings and described in the preceding detailed description, it is to be understood that the invention is not limited to the embodiment disclosed, but is capable of numerous rearrangements, modifications and substitutions without departing from the spirit of the invention as set forth and defined by the claims following.
|
Same subclass Same class Consider this |
||||||||||