A. van Deursen and T. Kuipers. Rapid system understanding: Two COBOL case studies. In Sixth International Workshop on Program Comprehension; IWPC'98, pages 90--98. IEEE Computer Society, 1998. 22  Home/Search   Document Not in Database   Summary   Related Articles   Check

....with little builtin functionality. Likewise, if a procedure is called by many others (high fan in) it is likely to be some sort of utility routine, dealing with error handling or logging. The procedures with both low fan in and low fan out are the ones that are likely to contain business logic [8]. Yet another starting point can be the screens used in the system [19] The screen sequence can be identified, together with the key strokes leading to each subsequent sreen. Such screen sequences are very close to use cases, telling what actions an end user performs. Moreover, following the ....

A. van Deursen and T. Kuipers. Rapid system understanding: Two COBOL case studies. In Sixth International Workshop on Program Comprehension; IWPC'98, pages 90--98. IEEE Computer Society, 1998. 22

.... Others may include the classification of variables, for example according to the type they belong to [7] or according to whether a variable is used to represent data obtained from persistent data stores [4] Our own experience with selecting domain related data and functionality is described in [6]. In this paper, we will use two guidelines, one for selecting data elements and one for selecting programs. These helped to find objects in our Mortgage case study, and we expect them to work well for other systems as well. First, in COBOL systems the persistent data stores (following the ....

....data from the database, and to format it for our cluster tools. The source code was from Mortgage, a 100.000 LOC COBOL system from the banking area. It uses VSAM files for storing data. The toolset used for the generation of data, and the architecture of those tools is described in more detail in [6]. The Mortgage system is described in more detail in [6, 25] For our cluster experiments we used S PLUS, a statistical analysis package from MathSoft. The cluster algorithms described in [10] are implemented as part of SPLUS. 1 All experiments were performed on a SGI O2 workstation. 4.3 ....

[Article contains additional citation context not shown here]

DEURSEN, A. V., AND KUIPERS, T. Rapid system understanding: Two COBOL case studies. In International Workshop on Program Comprehension (1998), IEEE Computer Society, pp. 90--97.

....All example analyses described are performed on Mortgage, a relation administration subsystem of a large mortgage software system currently in production at various banks. It is a 100.000 LOC COBOL system and uses VSAM files for storing data. The Mortgage system is described in more detail in [6, 9]. 2. TYPE INFERENCE FOR COBOL COBOL programs consist of a procedure division, containing the executable statements, and a data division, containing declarations for all variables used. An example containing typical variable declarations is given in Figure 1. Line 6 contains a declaration of ....

....lack of space. Filtering This picture may not be as insightful as we might hope. A way to decrease the complexity of this picture is by filtering out data before performing the concept analysis. A selection of relevant programs from all programs in a COBOL system can be made as described in [6]. COBOL systems typically contain a number of programs that implement low level utilities such as file I O, error handling and memory management. These programs can in general be left out of the analysis, particularly when we are only interested in the general structure of the system. Filtering ....

[Article contains additional citation context not shown here]

A. van Deursen and T. Kuipers. Rapid system understanding: Two COBOL case studies. In Sixth International Workshop on Program Comprehension; IWPC'98, pages 90--98. IEEE Computer Society, 1998.

....All example analyses described are performed on Mortgage, a relation administration subsystem of a large mortgage software system currently in production at various banks. It is a 100.000 LOC COBOL system and uses VSAM files for storing data. The Mortgage system is described in more detail in [6, 9]. 2. Type inference for COBOL COBOL programs consist of a procedure division, containing the executable statements, and a data division, containing dec1 larations for all variables used. Some typical variable declarations are shown in Figure 1. In line 6, a variable STREET is declared. Its ....

....in Figure 3. We will not show the list of items and features because it (obviously) will not fit in this paper. Filtering This picture may not be as insightful as we might hope. A way to decrease the complexity of this picture is by filtering out data before performing the concept analysis. In [6] we have described how a selection of relevant programs from all programs in a COBOL system can be made. COBOL systems typically contain a number of programs that implement low level utilities such as file I O, error handling and memory management. These programs can in general be left out of the ....

[Article contains additional citation context not shown here]

A. van Deursen and T. Kuipers. Rapid system understanding: Two COBOL case studies. In Sixth International Workshop on Program Comprehension; IWPC'98, pages 90--98. IEEE Computer Society, 1998. 9

....programs may be control modules, not containing any interesting functionality. Filtering out such programs, for example based on fan in fanout metrics, is necessary in order to arrive at meaningful cluster or concept analysis results. For our case study, we used selection criteria discussed in [5, 7]: the items are the fields of persistent data records, and the programs are restricted to those with a low fan in and fan out. 13 top P4 c2 ZIPCD NUMBER NUMBER EXT P2 c4 CITY P3 c3 STREET PREFIX NAME P1 c1 TITLE INITIAL bot Figure 6: Lattice for the concepts of Table 2 The ....

A. van Deursen and T. Kuipers. Rapid system understanding: Two COBOL case studies. In Sixth International Workshop on Program Comprehension; IWPC'98, pages 90--98. IEEE Computer Society, 1998.

....system understanding, i.e. at assisting the re engineer in understanding what components (modules, databases, screens, copybooks, the system consists of, and how these are related to each other. Figure 1 shows the extractor query viewer approach used in most reverse engineering tool sets [37, 16, 20]. It can be used to extract all sorts of facts from the legacy sources into a database. This database, in turn, can be queried, and relations of interest can be visualized. In the extractor phase, syntactic analysis will help to unravel the structure of the legacy code. This requires the ....

....in Section 3. If no parser is available, and if the required fact extraction is sufficiently simple, lexical analysis methods may be used: these are less accurate, but simpler, faster to develop and to run, and tolerant with respect to unknown language constructs or errors in the source code [38, 20]. In the querying phase, operations on the repository include relational queries, restriction of relations to names matching regular expressions, taking transitive closures of, for example call relations, lifting call relations from procedures to files, etc. A crucial aspect of querying is ....

[Article contains additional citation context not shown here]

A. van Deursen and T. Kuipers. Rapid system understanding: Two COBOL case studies. In S. Tilley and G. Visaggio, editors, Sixth International Workshop on Program Comprehension; IWPC'98, pages 90--98. IEEE Computer Society, 1998.

....Section 6 is present in Mortgage, and whether the proposed solution, subtyping, is adequate. For most of the variables, pollution is not an issue, i.e. subtyping can be safely replaced by type equivalence. However, all forms of pollution as discussed 2 This system was also used as case study in [21, 7]. in Section 6 do occur in Mortgage. Typical cases include the use of a single MOVE statement to initialise many different variables, the use of alpha numeric string variables to represent various types of strings, and the use of sections that use global variables to simulate formal parameters ....

....the usefulness of type inferencing, and to assess the validity of the choices made. In particular, we want to apply our technique to other COBOL systems and collect quantitative data on the inferred types. ffl We are working on applying type inferencing to component extraction, following [13, 7]. ffl Extension to new languages, most notably Fortran and IBM 370 assembler. ffl Visualisation of the inferred equivalence and subtype relations, the typed literal and enumerations types on the level of COBOL programs as well as visualisation of (the usage of) system level types in complete ....

A. van Deursen and T. Kuipers. Rapid system understanding: Two COBOL case studies. In Sixth International Workshop on Program Comprehension; IWPC'98, pages 90--98. IEEE Computer Society, 1998.

.... Others may include the classification of variables, for example according to the type they belong to [7] or according to whether a variable is used to represent data obtained from persistent data stores [4] Our own experience with selecting domain related data and functionality is described in [6]. In this paper, we will use two guidelines, one for selecting data elements and one for selecting programs. These helped to find objects in our Mortgage case study, and we expect them to work well for other systems too. First, in Cobol systems the persistent data stores (following the ....

....data from the database, and to format it for our cluster tools. The source code was from Mortgage, a 100.000 LOC Cobol system from the banking area. It uses VSAM files for storing data. The toolset used for the generation of data, and the architecture of those tools is described in more detail in [6]. The Mortgage system is described in more detail in [6, 25] For our cluster experiments we used S PLUS, a statistical analysis package from MathSoft. The cluster algorithms described in [10] are implemented as part of S PLUS. 1 All experiments were performed on a SGI O2 workstation. ....

[Article contains additional citation context not shown here]

DEURSEN, A. V., AND KUIPERS, T. Rapid system understanding: Two COBOL case studies. In International Workshop on Program Comprehension (1998), IEEE Computer Society, pp. 90--97.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC