----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://git.reviewboard.kde.org/r/129248/#review100220 -----------------------------------------------------------
Ship it! Seems reasonable to me, if indeed both need to be supported then I don't see a better way. - David Faure On Oct. 23, 2016, 6:23 p.m., Krzysztof Nowicki wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://git.reviewboard.kde.org/r/129248/ > ----------------------------------------------------------- > > (Updated Oct. 23, 2016, 6:23 p.m.) > > > Review request for KDE Frameworks, Dawit Alemayehu and David Faure. > > > Repository: kio > > > Description > ------- > > The default KNTLM behaviour when the client supplies no domain name is to use > the target name provided by the server. This however prevents from explicitly > using an empty domain name. > > The change makes use of the fact that QString can be either NULL or empty and > retains the old behaviour if the domain name is NULL (for ex. "username"), > but when a valid empty string is provided (for ex. "\username") the empty > domain name is used instead of the target name. > > This fixes some rare authentication failures for oddly configured servers. > > Autotests have been updated to reflect new behaviour. > > > Diffs > ----- > > autotests/http/httpauthenticationtest.cpp > a4367404003bff6b60cbfe633385125df8820a4e > src/kntlm/kntlm.h 5e0c2be6ad4dd62705da038f70fbbc4daacd60fb > src/kntlm/kntlm.cpp 269204e3c1d0a4615afefe385047ba54231ecaa6 > > Diff: https://git.reviewboard.kde.org/r/129248/diff/ > > > Testing > ------- > > Executed autotests including updated test cases. > > > Thanks, > > Krzysztof Nowicki > >
