JPH07311694A - Analyzing method for program execution path - Google Patents

Abstract

PURPOSE:To prevent defective generation at the time of program modification and improve the quality by managing the quality of a program product by paying attention to the program execution path. CONSTITUTION:An instruction execution network generating process part 14 generates new and old network tables 18a and 18b by using a source program 11 before modification, a source program 12 after modification, and a parameter 13 as input data. Then an executable path quantity calculating process part 15 generates new and old combination tables 19a and 19b on the basis of the new and old network tables 18a and 18b. Further, an execution path table generating process part 16 generate new and old execution path tables 20a and 20b on the basis of the new and old network tables 18a and 18b and new and old combination tables 19a and 19b. Lastly, an influence path information extracting process part 17 extracts and outputs influence path information 201 on the basis of the new and old execution path tables 20a and 20b.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a program execution path analysis method, and more particularly to a program execution path analysis method used to determine a target range of a program test.

[0002]

2. Description of the Related Art Conventionally, when a source program is changed while a created program is being debugged, a cross reference or the like has been used to investigate what kind of influence the change has. In addition, when performing a program test again on the changed source program, retest so that only the affected parts due to the program change are tested based on the coverage information set in each part of the source program. By deciding the test item, the efficiency of the re-test is improved (the invention described in Japanese Patent Laid-Open No. 1-229341).

[0003]

In the above-mentioned prior art, although it was possible to identify the affected part due to the change work based on the coverage information, the program execution path executed via the affected part (at the time of program execution) It was not possible to specify any one of the "process flows" that could be selected. Therefore, if the program execution path before and after the program change is different (for example,
When a branch statement is added / deleted, branch conditions are changed, etc.), it is easy to create defects that cannot be detected by the program test by the retest items determined based on the coverage information, and the produced program There is a problem that the quality control of the product cannot be performed sufficiently.

Therefore, an object of the present invention is to solve the above problems, to enable a program test focusing on the program execution path, and to provide a program execution path analysis method capable of sufficiently controlling the quality of a program product. To provide.

[0005]

[Means for Solving the Problems]

(1) In order to achieve the above object, the analysis method of the program execution path of the present invention, for the source program to be analyzed, selects the branch statement and the execution statement in the instruction statement constituting the source program, respectively, All the combinations of branch conditions in the branch statement of are obtained, and based on this, a program execution path table showing all program execution paths in the source program is created.

(2) Further, the source program before and after the change work associated with debugging (before the change: the first source program,
(After change: second source program), based on a second source program obtained by changing the first source program while creating a first program execution path table based on the first source program. A second program execution path table is created, and the first and second program execution path tables are compared and compared to extract all the program execution paths affected by the change.

[0007]

The operation based on the above configuration will be described.

(1) According to the program execution path analysis method of the present invention, for a source program to be analyzed, a branch statement and an execution statement in the instruction statements that make up the source program are selected, and the branch in each branch statement is selected. By obtaining all combinations of conditions and creating a program execution path table that represents all program execution paths in the source program based on the combinations, whether or not individual statement statements in the program are executed, Since it becomes possible to perform a program test by paying attention to whether or not each program execution path is adopted at the time of program execution, it is possible to sufficiently perform quality control of the produced program product.

(2) Further, the source program before and after the change work associated with debugging (before the change: the first source program,
(After change: second source program), based on a second source program obtained by changing the first source program while creating a first program execution path table based on the first source program. There is a change effect by creating a second program execution path table and comparing and comparing the first and second program execution path tables to extract all the program execution paths affected by the change. Since it is possible to determine the retest items for all program execution paths and perform the program test again, the defects created by the change work can be detected reliably and accurately, and the quality of the produced program product Can manage well.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the program execution path analysis method of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a diagram showing a schematic configuration of an embodiment of a system for realizing the program execution path analysis method of the present invention. In the figure, source program 1 before change
1, the post-change source program 12 and the parameter 13 as input data, the instruction execution network creation processing unit 14
Is the old network table 18a corresponding to the source program 11 before the change and the source program 12 after the change.
The new network table 18b corresponding to is created.
Next, using the old network table 18a and the new network table 18b as input data, the feasible path number calculation processing unit 15 stores the old combination table 19a and the changed source program 12 corresponding to the source program 11 before the change. The corresponding new combination table 19 is created. Further, using the old combination table 19a and the new combination table 19 as input data, the execution path table generation processing unit 16 corresponds to the old execution path table 20a and the changed source program 12 corresponding to the source program 11 before change. The new execution path table 20b is generated. Finally, using the old execution path table 20a and the new execution path table 20b as input data, the affected path information extraction processing unit 17 changes all the affected paths (the source program 11 before the change to the source program 12 after the change). The influence path information 201 indicating the information regarding the program execution path affected by the action is extracted.

FIG. 2 is a diagram showing an example of the pre-change source program in FIG. 1, and FIG. 3 is a diagram showing an example of the post-change source program in FIG. In both figures, 21 is an execution start point on the source program, and 22 is an execution end point. Reference numeral 23a in FIG. 2 shows the command statement before the change, and reference numeral 23b in FIG. 3 shows the command statement after the change. In this embodiment, the command statement with the sequence number “110” in FIG. "H;" is sequence number "106""108"
"110" three-step statement "IF X", "TH"
It has been changed to EN H; "and" ELSE Y; ".

FIG. 4 is a diagram showing an example of the parameters in FIG. In this embodiment, the sequence number “020” of the execution start point 21 and the sequence number “130” of the execution end point 22 on the source program are input as the parameter 13.

FIG. 5 is a diagram showing an example of a network table corresponding to FIG. 2, in which 64 is a command statement sequence number storage area, 65 is a command statement storage area, 61 is a control information storage area, and 62 is a selection. A possible number storage area, 63 is a pointer area. In the figure, a pointer area 63 having the number of entries corresponding to the selectable number set in the selectable number storage area 62 is secured, and each pointer area 63 has a next pointer (executed subsequent to the command statement). The sequence number of the statement to be written is stored. In the following description, the selected sentence and the repeated sentence are collectively referred to as a branch sentence.

FIGS. 6 to 8 are diagrams showing an example of the processing flow of the instruction execution network creation processing section in FIG. 1, which will be described with reference to these three figures. First, after the outer frame of the old network table 18a (new network table 18b) is generated, the pre-change source code provided as the input data is generated.
The sequence numbers and the contents of the instruction statements are set in the sequence number storage area 64 and the instruction statement storage area 65 for all the instruction statements constituting the source program 11 (the changed source program 12). At this time, when the corresponding imperative sentence is a selection sentence, DM is added at the end of each selection clause.
A YEND sentence is added (step 501). Next, for each imperative statement, the imperative statement is a selected statement (IF statement or CASE
In the case of a statement), the number of selected clauses
"2" (for sentences, etc.)
(Indicates two kinds of selections in case of continuous execution), “1” in the case of continuous sentence, and “0” in the case of execution end point, respectively, is set in the selectable number storage area 62 as the selectable number. At the same time, if the statement is the execution start point, the flag "S"
, The flag “E” for the execution end point, the flag “L” for the repeated statement, and the control information storage area 61.
(Step 502).

In the processing after step 503, the above-mentioned next pointer is obtained by sequentially analyzing from the beginning the command statements constituting the source program 11 before change (source program 12 after change) given as input data. Is set in the pointer area 63.

First, the old network is used as an analysis pointer.
Set the sequence number of the first command statement in the table 18a (new network table 18b) (step 5).
03). Then, the analysis pointer points to the end of the last statement of the old network table 18a (new network table 18b) (step 504 = YES), and all pointer areas in the network table 18 have been set. If (step 505 = YES),
The instruction execution network creation process ends (step 5).
04 → step 505 → END). Further, the analysis pointer points to the next of the last statement of the old network table 18a (new network table 18b) (step 504 = YES), and all pointer areas in the network table 18 have been set. Not (step 50)
If 5 = NO, the analysis is continued until all pointer areas are set (step 504 → step 50).
5 → step 503 → ……).

On the other hand, when the analysis pointer points within the network table 18 (step 504 = NO),
The statement is the execution end point (step 506 = YE).
S) or if the pointer area corresponding to the imperative sentence has already been set (step 507 = YES), the imperative sentence analysis is skipped and the immediately following imperative sentence is analyzed (step 506). → step 507 → step 508 → step 504 → ...). Further, the pointer area corresponding to the command statement has not been set (step 50).
In the case of 7 = NO), the pointer area setting process after step 509 is performed.

If the imperative sentence is a continuous sentence (step 509 = YES) and the imperative sentence is an unconditional branch sentence (step 510 = YES), the branch destination sequence is used as the next pointer. Set a number in the pointer area and branch to step 508 (step 509 →
Step 510 → Step 512 → Step 508 → ...
...), if the imperative statement is not an unconditional branch statement (step 510 = NO), the sequence number of the imperative statement immediately after the imperative statement is set and the procedure branches to step 508 (step 509 →). Step 510 → Step 511 →
Step 508 → ...). That is, when the analysis pointer is set in the instruction statement corresponding to the sequence number "120" in the network table in FIG.
After "130" is set as the next pointer (step 511), "130" is set as the analysis pointer and the analysis is continued (step 508).

If the imperative statement is not a continuous statement (step 509 = NO), and if the imperative statement is a repeat statement (step 520 = YES), the imperative statement immediately after the repeat statement and the repeat The sequence numbers of the end statement corresponding to the statement and the imperative statement immediately after the end statement are obtained (step 521), and the repetition statement and the next pointer of the end statement for the repetition statement are set in the respective pointer areas 63. Then, (step 522), the sequence number of the imperative sentence immediately after the end sentence for the repetitive sentence is set in the analysis pointer, and the analysis is continued (step 52).
3). That is, the command statement (DO WHI) corresponding to the sequence number "080" in the network table in FIG.
When the analysis pointer is set in (LE sentence), DO W
Sequence number “090” of the imperative statement immediately after the HILE statement, DO
Sequence number “10” of END statement corresponding to WHILE statement
0 ”, the sequence number of the statement immediately after the END statement“ 1 ”
10 "is obtained (step 521), and" 090 "and" 110 "are set as the next pointers to the DO WHILE statement, and E
As a next pointer for the ND sentence, "080" is set in each pointer area 63 (step 52).
2) The sequence number "110" of the imperative sentence immediately after the end sentence for the repetitive sentence is set in the analysis pointer,
The analysis is continued (step 523).

When the imperative sentence is neither a continuous sentence nor a repetitive sentence (step 509 = NO, step 520 = NO), when the imperative sentence is a selected sentence (step 530), the selectable number is first. By referring to the storage area 62, the number n of selected clauses of the selected sentence is calculated, and the work variable k
← n−1, and the sequence number S1 of the instruction statement described at the beginning of the first selection clause and the selection statement group (comprising the selection statement and a plurality of selection clauses at the branch destination). (The entire selection process), the sequence number S0 of the next statement is obtained (step 531). And the statement S1
, The sequence number S2 of the statement immediately after the statement S2, the sequence number S3 of the last statement of the selected section, and the sequence number S4 of the statement immediately after the statement S3 are obtained (step 532).
From "n- in the pointer area 63 of the selected sentence.
The sequence number S1 and the statement S
The sequence number S2 is set in the pointer area 63 of 1 and the sequence number S0 is set in the pointer area 63 of the statement S3 (DMYEND statement) (step 533).
After the above processing, if there are remaining selected clauses to be processed (step 534 = NO), the work variables k ← k−1, S1 ← S4 are set, and then step 532 is performed.
By branching to step 535 (step 535), the same processing is performed for the subsequent selected clauses. When the processing for all the selected clauses is completed (step 534 = Y
ES), after setting S0 as the next analysis pointer, the process is continued (step 536).

That is, when the analysis pointer is set to the instruction statement (IF statement) corresponding to the sequence number "020" in the network table in FIG. 5, first, n ← 2 and k ← 1, S1 ← “030”, S0 ← “0
50 "(step 531), S2 ←" 031 ", S3
← “031”, S4 ← “040” (step 53
From 2), "030" is set as the first next pointer of the IF statement of "020", "031" is set as the next pointer of the THEN statement of "030", and "050" is set as the next pointer of the DMYEND statement of "031". After setting (step 533), since k ≠ 0, S1 ←
As “040”, S4 ← “040”, k ← 0, the process branches to step 532 (step 534 = NO → step 535). Next, S2 ← “041”, S3 ← “04
Since 1 ”and S4 ←“ 050 ”(step 532),“ 040 ”is set as the second next pointer of the IF statement of“ 020 ”,“ 041 ”is set as the next pointer of the ELSE statement of“ 040 ”, and“ 041 ”is set. After setting “050” as the next pointer of the DMYEND sentence of “” (step 533) and k = 0, set “050” as the next analysis pointer and continue the process (step 534 = YES). → Step 536).

FIG. 9 is a diagram showing an example of the combination table corresponding to FIG. 2. 81 is an execution path name storage area, and 82 is an execution path name storage area.
Is a branch statement sequence number storage area (instruction statement in which a value of 2 or more is set in the selectable number storage area 62 in FIG. 5 = branch statement sequence number is stored), and 83 is a selection number storage area (each In the execution path of (1), it indicates which selection clause of the instruction statement corresponding to the sequence number stored in the branch statement sequence number storage area 82 is selected and executed). In the figure, the value stored in the selection number storage area 83 is the corresponding branch statement sequence number storage area 82.
It is an integer value that is 1 less than the number of branches of the instruction statement indicated by the branch statement sequence number stored in 1. and 0 or more. For example, when the number of branches of a certain branch statement = 3, the value stored in the selection number storage area 83 is “0”,
It is one of three values of "1" and "2", and each execution path shown in each line of the branch statement sequence number storage area 82 is one of "0", "1", and "2". The selected value of is set. That is, the value S stored in the selection number storage area 83 indicates that the (S + 1) th selected clause in the corresponding branch statement is executed.

Since the maximum number of branches in the source program of this embodiment is 2, the selection value S stored in the selection number storage area 83 is "0" or "1" in the combination table shown in FIG. , The selection value S = 0 indicates that the first selection clause is executed, and the selection value S = 1 indicates that the second selection clause is executed. In addition, the setting order of the values S stored in the selection number storage area 83 is as follows. First, the sequence number storage area corresponding to the execution path associated with the first branch statement (the branch statement with the sequence number “020” in FIG. 9). "0" or "1" is stored in 82, and then "0" is stored in the sequence number storage area 82 corresponding to the execution path associated with the second branch statement (the branch statement with the sequence number "050" in FIG. 9). Alternatively, "1" is stored and the order is "...

10 and 11 are diagrams showing an example of the processing flow of the executable path number calculation processing unit in FIG.
In both figures, the old network table 18a is first shown.
The total number of executable paths is obtained by multiplying all the values stored in the selectable number storage area 62 in the (new network table 18b). In the case of the source program before change shown in this embodiment, the total number of executable paths = 2 × 1 × ... × 1 × 2 × 1 × ... × 1 ×
2 × 1 × ... × 1 = 8 (step 710). Further, the sequence number and the number of instruction statements = branch statements whose value stored in the selectable number storage area 62 is 2 or more are obtained. In the present embodiment, the number of branch statements is 3, and the sequence numbers thereof are "020", "050", and "080", respectively (step 711). The execution path name storage area 81
The execution path name and the sequence number obtained in step 711 are respectively stored in the branch statement sequence number storage area 82 to generate the old combination table 19a (new combination table 19b) (steps 712 and 713). . Next, the selection value setting pointer is positioned at the first entry of the branch statement sequence number storage area 82 in the old combination table 19a (new combination table 19b) (step 714), and the selection value round setting target number P, current The number of branches Pi of the branch statement to be processed and the processing target entry C indicating the entry of the branch statement sequence number storage area 82 of the current processing target are defined and set, and P
← <total number of executable paths>, Pi ← 1, C ← 1 (step 715).

The process from step 716 is a process of storing the selected value in the selection number storage area 83. First, the selectable number storage area in the old network table 18a (new network table 18b) is stored. Referring to 62,
The selectable number of the branch statement indicated by C stored in the branch statement sequence number storage area 82 is calculated, and the value is set to the branch number Pi of the branch statement of the current processing target (step 7).
16). Then, the selection value repetition number R (used for repeatedly setting the same selection value) and Rs for R recovery are defined as R ← (P / Pi) and Rs ← (P / Pi) (step 717), the old combination table in which the selected value is set.
A row counter E ← 0 (step 718) indicating the number of rows of the table 19a (new combination table 19b) and a selected value S ← 0 (step 719) are set.

Next, it is judged whether or not the selection value repetition number R = 0. If R ≠ 0, the process branches to step 721, and if R = 0, the process branches to step 730 (step 720).
That is, when R ≠ 0 (step 720 =
No), the value of the selection value S is stored in the selection number storage area 83 indicated by the selection value setting pointer (step 721), and then the selection value setting pointer is modified to the next line to repeat the selection value R ← R. After setting -1, the row counter E ← E + 1 (step 722), the process returns to step 720. If R = 0 (step 720 = YE
S), C branch statement sequence number storage area 82
It is determined whether the setting of the selection value for is completed by whether the value of the row counter E has reached the total number of executable paths, and if E <(the total number of executable paths), the step 731, and if E = (total number of executable paths), branch to step 734 (step 73).
0).

When E <(total number of executable paths) in step 730 (step 730 = YES), the selection value S ← S + 1 is set (step 731) and the selection value repetition number R is recovered from Rs. After setting R ← Rs (step 732), it is determined whether or not the relationship between the selected value S and the number of branches Pi of the currently processed branch statement is “S <Pi” (step 733). This is a determination for deciding whether or not to return the selection value that has been set so far to the initial value “0” for the selection value unset path of the branch statement that is the current processing target. If <Pi ”(step 733 = YES), the process returns to step 720 to set the selection value S immediately after S ← S + 1 to the selection value unset path. If “S ≧ Pi” (step 733 = NO), step 7 is performed to set the initial value “0” as the selection value S in the selection value unset path.
Return to 19.

If E = (total number of executable paths) in step 730 (step 730 = NO), it is determined in step 734 whether Rs = 1.
It is determined whether or not the selection value S has been set for all paths of all entries in the branch statement sequence number storage area 82 of the old combination table 19a (new combination table 19b). That is, when Rs = 1 (step 7)
34 = YES), the execution path number calculation process is terminated assuming that the selection value S has been set for all the paths. Further, when Rs ≠ 1 (step 734 = N
In O), it is assumed that there is an entry for which the selection value S is not set among the entries in the branch statement sequence number storage area 82 and the branch statement sequence number storage area 82.
Since the selection value setting pointer is set for the next entry in C, C ← C + 1 (step 735), P ← Rs
After (Step 736), the process returns to Step 716.

By the above-described execution pass number calculation processing, the selection value S in the old combination table 19a shown in FIG. 9 is set in the selection number storage area 83 in the following order. In the case of the first entry (“020”) in the branch statement sequence number storage area 82, “0” is continuously set in PANDR001 to 004. “1” is continuously set in PANDR005 to 008. In the case of the second entry (“050”) in the branch statement sequence number storage area 82, “0” is continuously set in PANDR001 to 002. “1” is continuously set in PANDR003 to 004. “0” is continuously set in PANDR005 to 006. "1" is continuously set in PANDR007 to 008. In the case of the third entry (“050”) in the branch statement sequence number storage area 82 “0” is set in PANDR001. “1” is set in PANDR002. “0” is set in PANDR003. “1” is set in PANDR004. “0” is set in PANDR005. “1” is set in PANDR006. “0” is set in PANDR007. “1” is set in PANDR008. That is, in the execution path PANDR001, 020
, The first selection clause (THEN statement), the 050 selection statement, the first selection clause (THEN statement), and the 080 repetition statement, the first selection clause (with loop), To be executed.

FIG. 12 is a diagram showing an example of the execution path table corresponding to FIG. In the figure, 101 is an execution path name storage area (corresponding to the execution path name storage area 81 in FIG. 9), and 102 is an instruction statement sequence number in which a sequence number corresponding to an instruction statement executed in the execution path is stored. A storage area 103 is an instruction statement storage area in which execution instruction statements executed in the execution path are stored. The command statement sequence number storage area 102 and the command statement storage area 103 form a pair, and are generated in a sequence according to the execution order of the execution command statements. Further, each execution path corresponding to the execution path name in the execution path name storage area 101 is
The instructions to be executed or the execution order of the executed instructions are different from each other.

13 and 14 are diagrams showing an example of the processing flow of the execution path table generation processing section in FIG. The outline of the process is the old network table 18a.
Based on (new network table 18b) and old combination table 19a (new combination table 19b),
Old combination table 19a (New combination table 19a
For all execution paths set in b), after the instructions executed in the execution path are arranged in the execution order, the instructions executed in each execution path are the same including the execution order. The presence or absence of a certain other execution path is checked, and when there are two or more identical execution paths, only one execution path is left and all other identical execution paths are deleted.

In both figures, first, steps 910-9 are performed.
After performing necessary initialization in 13, the execution path name Pn is obtained by extracting one execution path name from the execution path name storage area 81 in the old combination table 19a (new combination table 19b). Is stored in the execution path name storage area 101 of the old execution path table 20a (new execution path table 20b) (step 914), and the old network table 18a (new network table 1) is generated.
8b) execution start point (step 91)
5). Then, the contents (for example, Sn and Cn) of the command statement sequence number storage area 64 and the command statement storage area 65 of the entry pointed to in the old network table 18a (new network table 18b) are changed to An instruction statement sequence number storage area 102 in the old execution path table 20a (new execution path table 20b).
And the instruction statement storage area 103, and in order to set the instruction statement (for example, Cn + 1) to be executed subsequently, the instruction statement sequence number storage area 102 and the instruction statement storage area 103 are modified by one entry ( Step 916).

Next, in the control information storage area 61 of the command statement Cn, it is determined whether the command statement Cn stored in the old execution path table 20a (new execution path table 20b) in step 916 is the execution end point. It is confirmed whether the flag E is stored or not, and if it is the execution end point, step 93
If not, the process branches to step 918 (step 917). When the flag E is not stored in the control information storage area 61 of the command statement Cn (step 917 = NO), it is set in step 918 by referring to the selectable number storage area 62 corresponding to the command statement Cn. It is confirmed whether or not the number of selected clauses is 2 or more, and when it is determined that the number is not 2 or more (step 918 = NO),
The next pointer of the statement Cn is used as the analysis pointer (step 919), and the process returns to step 916. If it is determined that the set number of selected clauses is 2 or more (step 918 = YES), the statement to be executed next to the statement Cn is the old network table 18a (new network). Since it is not always the statement of the next entry in the table 18b), the entry and sequence corresponding to the execution path currently being processed in the old combination table 19a (new combination table 19b). The selection value stored in the selection number storage area 83 at the intersection with the number Sn column is obtained (step 920), and the (S + 1) th next pointer of the statement Cn is used as the analysis pointer based on the selection value (step 920). 921) and the procedure returns to step 916.

When the flag E is stored in the control information storage area 61 of the command statement Cn (step 917 = NO),
That is, if it is determined that the processing of rearranging all the command statements executed in the execution path currently being processed in the execution order is completed, whether the same rearrangement processing is completed for all the execution paths. If it is determined that all the rearrangement processes have been completed, the process branches to step 934, and if not, the process branches to step 931 (step 930). That is,
When it is determined that the above-mentioned rearrangement processing has not been completed for all execution paths (step 930 = NO), the old combination table 19a (new combination) is set in order to perform the processing for the subsequent execution paths. The entry currently being processed in the table 19b) and the old execution path table 20a.
The entry currently being processed in the (new execution path table 20b) is modified by one entry each (step 931 and step 932), and then step 914.
Return to. Further, when it is determined that all the rearrangement processes have been completed (step 930 = YES), whether or not there is another execution path in which the executed instructions are the same, including the execution order, for each execution path. When there are two or more identical execution paths, only one execution path is left and the other identical execution paths are deleted (step 934), and the execution path table generation processing is completed. To do.

For example, in the execution path in the pre-change source program 11 shown in FIG. 2 of this embodiment, the execution path P
ANDR001 and PANDR002, execution path PA
The command statements executed by NDR005 and PANDR006 are exactly the same, including the order of execution, as shown below. PANDR001: 020 → 030 → 050 → 060 →
120 → 130 PANDR002: 020 → 030 → 050 → 060 →
120 → 130 PANDR005: 020 → 040 → 050 → 060 →
120 → 130 PANDR006: 020 → 040 → 050 → 060 →
120 → 130 Therefore, the execution paths PANDR002 and PANDR006 are deleted from the old execution path table 20a generated by the execution path table generation processing in this embodiment.

FIG. 15 is a diagram showing an example of the processing flow of the affected path information extraction processing section in FIG. In the figure,
First, the old execution path table 20a generated from the source program 11 before the change and the source program 12 after the change
All the new execution path tables 20b generated from the above are compared for each execution path, and if a matching execution path is detected, the old execution path table 20a and the new execution path table 20b are matched. The executed path is deleted (step 111). Then, regarding the execution paths remaining in the old execution path table 20a and the new execution path table 20b,
DELETED PATHS in the old execution path table 20a and ADDED PATH in the new execution path table 20b.
As S, the affected path information 201 is extracted and output (step 112).

FIG. 16 is a diagram showing an example of the affected path information extracted based on FIGS. 1 and 2. In this embodiment, four paths of the source program before change are deleted and the source program after change is deleted. Eight passes have been added. In addition,
“<” And “>” denoted by reference numeral 121 are loop symbols indicating loops 1 to n times.

[0039]

【The invention's effect】

(1) As described in detail above, according to the program execution path analysis method of the present invention, for the source program to be analyzed, the branch statement and the execution statement in the instruction statement constituting the source program are selected, Individual instruction statements in a program are executed by obtaining all combinations of branch conditions in each branch statement and creating a program execution path table that represents all program execution paths in the source program based on this combination. Whether or not each program execution path is adopted at the time of program execution, it is possible to carry out a program test, so that it is possible to perform sufficient quality control of the produced program product. The effect that it can be obtained.

(2) Further, the source program before and after the change work associated with debugging (before the change: the first source program,
(After change: second source program), based on a second source program obtained by changing the first source program while creating a first program execution path table based on the first source program. There is a change effect by creating a second program execution path table and comparing and comparing the first and second program execution path tables to extract all the program execution paths affected by the change. Since it is possible to determine the retest items for all program execution paths and perform the program test again, the defects created by the change work can be detected reliably and accurately, and the quality of the produced program product The effect is that management can be performed sufficiently.

[Brief description of drawings]

FIG. 1 is a diagram showing a schematic configuration of an embodiment of a system for realizing a program execution path analysis method of the present invention.

FIG. 2 is a diagram showing an example of a pre-change source program in FIG.

FIG. 3 is a diagram showing an example of a changed source program in FIG.

FIG. 4 is a diagram showing an example of parameters in FIG.

5 is a diagram showing an example of a network table corresponding to FIG.

FIG. 6 is a diagram (No. 1) showing an example of the processing flow of the instruction execution network creation processing unit in FIG. 1.

7 is a diagram (No. 2) showing an example of the processing flow of the instruction execution network creation processing unit in FIG. 1. FIG.

FIG. 8 is a diagram (No. 3) showing an example of the processing flow of the instruction execution network creation processing unit in FIG. 1.

9 is a diagram showing an example of a combination table corresponding to FIG.

FIG. 10 is a diagram (No. 1) showing an example of the processing flow of the executable path number calculation processing unit in FIG. 1;

FIG. 11 is a diagram (No. 2) showing an example of the processing flow of the executable path number calculation processing unit in FIG. 1;

12 is a diagram showing an example of an execution path table corresponding to FIG.

FIG. 13 is a diagram (No. 1) showing an example of the processing flow of the execution path table generation processing unit in FIG. 1.

FIG. 14 is a diagram (No. 2) showing an example of the processing flow of the execution path table generation processing unit in FIG. 1;

FIG. 15 is a diagram showing an example of a processing flow of an affected path information extraction processing unit in FIG.

FIG. 16 is a diagram showing an example of affected path information extracted based on FIGS. 1 and 2.

[Explanation of symbols]

 11 Source Program Before Change 12 Source Program After Change 13 Parameters 14 Instruction Execution Network Creation Processing Unit 15 Executable Path Number Calculation Processing Unit 16 Execution Path Table Generation Processing Unit 17 Impacted Path Information Extraction Processing Unit 18a , 18b Network table 19a, 19b Combination table 20a, 20b Execution path table 201 Influenced path information 21a, 21b Source program execution start point 22a, 22b Source program execution end point 23a, 32b Changed statement 61 Control information storage area 62 Selectable number storage area 63 Pointer area 64, 102 Command statement sequence number storage area 65, 103 Command statement storage area 81, 101 Execution path name storage area 82 Branch statement sequence number storage area 83 Selection number Storage area 121 Loop symbol

Claims (2)

[Claims] 1. A branch statement and an executable statement in an instruction statement constituting a source program are selected, all combinations of branch conditions in each branch statement are obtained, and based on this, all program execution paths in the source program. A method for analyzing a program execution path, characterized by creating a program execution path table representing 2. A second program execution path based on a second source program obtained by modifying the first source program while creating a first program execution path table based on the first source program. 2. A program execution path analysis according to claim 1, wherein a table is created, and all the program execution paths affected by the change are extracted by comparing and comparing the first and second program execution path tables. Method.