On Tue, 2009-12-01 at 17:07 +0100, Sven Neumann wrote: > On Tue, 2009-12-01 at 15:58 +0000, Andre DRASZIK wrote: > > On Tue, 2009-12-01 at 16:36 +0100, Sven Neumann wrote: > > > You probably have a point here and this should be fixed. But I don't > > > quite understand why your test code takes the hard way of calculating > > > the number of bytes from the number of characters. You can simply > > > subtract the pointers into the string buffer, can't you ? > > > > Probably - I didn't write the test case, only the patch :-) > > > > Hm, if you were to subtract the pointers, you'd get a different > > (incorrect) result again, as with the patch only 59 chars would have to > > be drawn (when using ret_str_length to determine the amount of chars) > > whereas subtracting pointers would still give you 60. > > Can the test case be changed then so that it also checks that the > returned pointer to the next line is correct?
The next line pointer is correct though, GetStringBreak() tries to skip ' ' and '\n' at the end of the current line, thus ret_str_length (converted to bytes) can be different from the pointerdiff (61 vs. 62 bytes in this case) a. _______________________________________________ directfb-dev mailing list directfb-dev@directfb.org http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev
