zaks.anna added a comment. This definitely needs tests!
I agree that the file name should be used for bug identification. I am not sure if it should be included in the hash by the analyzer or if it should be included by the processing script. Have you looked at CmpRuns.py and SATestBuild.py? > On the other hand, the tools will be much faster if they do not have to > process more information from the report to identify the bugs. The file name is another field in the same report; I do not see how this would result in a much slower scripts. Here is one problem to consider: What is included as "filename"? Is it the full path to the file where the issue is found? Is it just the file name without the path? Does your solution support uniquing the issues when the same project is analyzed from different directories? How will you handle issues coming from 2 different projects that have the same file name? Our testing scrips that is based on CmpRuns has the notion of multiple projects and bug reports coming from different projects are considered to be different. We include project name and the location of the file within the project to identify the issue. Not including a filename in the hash allows for that model. http://reviews.llvm.org/D10305 _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
