xazax.hun added a comment. I was thinking a bit more, what would be the best way to determine what should we include in the hash.
In order to determine that first we need to define the scope of the hash itself. There are several sensible choices such as: - identify bugs that were generated by a specific checker in a translation unit. (i.e. check name is not included in the hash) - identify bugs that has the same bugtype in the same translation unit. - identify bugs in a translation unit. - identify bug within a project. - identify bugs globally. - etc... Note that the scope of the hash does not necesserily affects the scope of diffing results, since the information omitted from the hash can be processed separately. If we exclude something from the hash that is available in the plist we give flexibility to the tools that consume plists, but also give more possibility to diverge in the way bugs are identified. http://reviews.llvm.org/D12906 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
