Created by James Hurrell, last modified on Jul 31, 2020
1.1.4-funcrel
Resolved Issues
Customer Ticket Id | Details |
---|
27872 | Rule Rationale update for the QR (rule id - 1002208): "All Read and Write statements in Programs must use the return code IOSTAT (FORTRAN)" |
Other Updates
Details |
---|
After upgrade "Add Version" would not throw error " Dependency dmtcoetechnoua:2.0 is missing: library dmtcoediscovererfortran" |
Fortran Analyzer no longer supports Oracle or SQL server as storage service. |
Rules
Rule Id | New Rule | Details |
---|
1002088 | FALSE | Weight for "Avoid too many Statements (STMT) in a Subroutine (FORTRAN)" has been changed from 0 to 1 |
1002090 | FALSE | Weight for "Avoid too many Statements (STMT) in a Program (FORTRAN)" has been changed from 0 to 1 |
1002092 | FALSE | Weight for "Avoid too many Statements (STMT) in a Function (FORTRAN)" has been changed from 0 to 1 |
1002094 | FALSE | Weight for "Avoid too many Statements (STMT) in a Module (FORTRAN)" has been changed from 0 to 1 |
1002110 | FALSE | Weight for "Avoid Programs with High Average size of Statements (AVGS) (FORTRAN)" has been changed from 0 to 1 |
1002112 | FALSE | Weight for "Avoid Subroutines with High Average size of Statements (AVGS) (FORTRAN)" has been changed from 0 to 1 |
1002114 | FALSE | Weight for "Avoid Functions with High Average size of Statements (AVGS) (FORTRAN)" has been changed from 0 to 1 |
1002116 | FALSE | Weight for "Avoid Modules with High Average size of Statements (AVGS) (FORTRAN)" has been changed from 0 to 1 |
1.1.3-funcrel
Resolved Issues
Customer Ticket Id | Details |
---|
25153 | Avoid untested Return Code variables after ALLOCATE in Programs (FORTRAN) - returns false positive when no ALLOCATE is present in the code |
25225 | Missing security violations for “Avoid untested Return Code variables after Read or Write in Subroutines (FORTRAN)”. |
25487 | Fortran function objects are not always created in Metamodel |
24044 | Rule 1002142 -Unavoid un-deallocated variables or arrays in Fortran-90 Modules (FORTRAN) -description needs update |
25198 | Associated Value details should be clear for a violation - Fortran |
1.1.2-funcrel
Resolved Issues
Customer Ticket Id | Details |
---|
23872 | Fortran Analyzer 1.1.1-funcrel creates over 500 false positives for rule Avoid Untested Return Code variables after ALLOCATE in Subroutines |
Other Updates
Details |
---|
Incorrect objects created for Fortran Stat Variable |
1.1.1-funcrel
Updates
Rename "Extension" to Fortran "Analyzer"
Fortran Extension is now known as Fortran Analyzer.
Deprecate FortranIfConstruct, FortranSelectCase, FortranDoConstruct
Objects of type FortranIfConstruct, FortranSelectCase, FortranDoConstruct will no longer be created after analysis. Links and metrics belonging to these objects will now belong to their respective container artifacts. Therefore, after upgrade, QR results will change.
Resolved Issues
The following table lists the bugs fixed in Fortran 1.1.1-funcrel.
Internal ID | Description | Impact? |
---|
FORTRAN-34 | Getting many more FORTRAN warnings and not getting any violation in dashboard | The warnings for types FortranIfConstruct, FortranSelectCase, FortranDoConstruct have been removed as these objects are no longer created. |
1.1.0-funcrel
What to expect after an upgrade
1.1.0-funcrel is a major release and includes many bug fixes to improve the results produced by the extension. Therefore, after upgrading to 1.1.0 and generating a post upgrade consistency snapshot on unchanged source code, analysis results (objects, links, rules) will be impacted as follows:
- Less files are skipped during the analysis, providing greater accuracy.
- More objects are discovered and included in the analysis.
- Object boundaries are discovered more accurately.
- Strings and comments are now detected correctly.
Resolved Issues
The following table lists all bugs fixed in Fortran 1.1.0-funcrel.
Internal ID | Description | Impact? |
---|
FORTRAN-35 | Snapshot is stuck from 2 days | Snapshot should now complete more quickly. |
FORTRAN-36 | Functions are named "<unnamed>" when return value type is precedes keyword function | Function names will be correctly resolved in this situation. |
FORTRAN-40 | Exception while calculating "Number of Lines of Code (Fortran)" | Exception will no longer occur. |
FORTRAN-43 | Warning "Unable to find the object end for type 'FortranExternalSubroutine'" during analysis | The warning message is no longer displayed. Object end types are now correctly identified. |
FORTRAN-44 | Warning "Duplicate object of type 'FortranExternalSubroutine' has been detected" is given during analysis | The warning message is no longer displayed. |
FORTRAN-45 | Warning "Unable to find the object end for type 'FortranInterface'" | The warning message is no longer displayed. Object end types are now correctly identified. |
FORTRAN-46 | Preprocessor log file becomes huge | The log file will now be much smaller. |
FORTRAN-47 | Fortran pre-processor processes files of any type | Only Fortran related files will now be processed. |
FORTRAN-48 | Fortran pre-processor may fail to create files (intermittent, not easy to replicate) | Files without fortran specific extensions are now correctly handled. |
FORTRAN-52 | Fortran pre-processor process errors out if file cannot be created | A message will now be displayed in the log stating which file cannot be created. Any subsequent files will also be correctly processed instead of being skipped. |
FORTRAN-53 | Warning "Unable to find the object end for type 'FortranInternalSubroutine'" is given during analysis | The warning message is no longer displayed. Object end types are now correctly identified. |
FORTRAN-54 | Warning "Duplicate object of type 'FortranInternalSubroutine' has been detected" is given during analysis | The warning message is no longer displayed. |
FORTRAN-55 | Warning "Unable to find the object end for type 'FortranModule'" is given during analysis | The warning message is no longer displayed. Object end types are now correctly identified. |
FORTRAN-56 | Fortran extension does not consider endfunction and endsubroutine as end of artifacts | endfunction and endsubroutine clauses are now correctly identified as the end of artifacts |