Daniel Marjamäki
Daniel Marjamäki
> It would rely on the input always being in a fixed order though (which might not be case depending on the generator). It's not really a problem as long...
I basically approve this.. however well I assume that theoretically it could cause problems if a ui file would contain the same string in english in two places and the...
> We could also specify -locations none and it would also get rid of the file is originates from reducing the chance of it being modified even more. I have...
> But I wonder why we calculate it based on on the file size anyways. That makes no sense because we are using the size of the actual input file....
> A small file would generally go faster to analyse than a large file. That is my "best guess".. I can't prove it. :-)
> Would it be possible to base it on the length of the token list? nah.. I believe it would be more accurate but that would increase the analysis time....
> I still think we should drop the file size percentage I don't want to drop it.
what does the test-my-pr.py say? are there diffs..?
> This should probably be implemented by introducing Python tests similar to what Cppcheck is doing. @danmar would that be acceptable for tests which actually require (multiple) files? sure, it...
> if I guess correctly, the failure is due to that the header files are not really in the working directory or so, right? yes that sounds probable to me....