On 31 August 2015 at 10:44, Dmitry Kazakov <dimul...@gmail.com> wrote: > > >> >> 2) I am personally against of automated whitespace reformatting, >> >> because >> >> it pollutes history of files without any use. Includes, >> >> slots, forward declarations reformattings are ok. Whitespace no. >> > >> > >> > There is a lot of use in whitespace cleanup: it makes the code >> > consistent and readable. >> >> Yes, we discussed that a bit a while ago. Initially I was critical >> too. But whitespace changes can be ignored when generating diff _and_ >> blame (the -w option). Anyone can add this flag in git options for >> diff/blame. I'd do so. >> >> I'd like to note that we don't have a proven formatting tool for now >> for all C++ elements. astyle got broken, and clang formatter can be >> better but we need an expert. Missing things may be minor but we can't >> make the result worse in problematic areas than the original. Not even >> KF5 has a formatting tool, last I checked. > > > Yes, the last time we discussed it, I raised quite a few issues with the > formatter and got a response that "they are limitations of the formatter and > cannot be overcome easily". So why we should spend time on it at all if it > doesn't solve any real issue? At least in Krita I don't see any real problem > with formatting. Yes, there are a few places where it is non-standard, but > we fix it when refactoring these lines.
How about that: A kind of the same here.Personally I can live in kexi.git without formatting and renames for some time, in this dept non-app repos like kreport.git deserve renames first for basic consistency (there are classes with no prefixes, Ko prefixes and KReport prefixes, all mixed now). Because it's more like a 'public' code, headers are actually installed. When kexi.git and krita.git and ( cut-off ) calligra.git gets created, repo maintainers can decide with to do regarding formatting. I hope it can wait. First we're about to have some renames. Every calligra subproject has specific time frames and workforce. Let's distribute the decision.... PS: Re-wrapping looks like the most intrusive change. git diff -w won't ignore it. Example: *Original: one line* void FoobarClass::setFoobar(VeryLongType a1, VeryLongType a2, VeryLongType a3) -> *Formatted: 2 lines* ** void FoobarClass::setFoobar(VeryLongType a1, VeryLongType a2, VeryLongType a3) -- regards, Jaroslaw Staniek KDE: : A world-wide network of software engineers, artists, writers, translators : and facilitators committed to Free Software development - http://kde.org Calligra Suite: : A graphic art and office suite - http://calligra.org Kexi: : A visual database apps builder - http://calligra.org/kexi Qt Certified Specialist: : http://www.linkedin.com/in/jstaniek
_______________________________________________ calligra-devel mailing list calligra-devel@kde.org https://mail.kde.org/mailman/listinfo/calligra-devel
