Sorry for lated response about this issue. Yes, about this case, I agree with the removal of redundant default case after the complete list of possible enum values.
The variable, gxvalid->statetable.entry_glyphoffset_fmt, it is set by gxv module internally not imported from external font file (thus, malicious input data cannot set the value to the undefined value). So it is OK to remove the default case. Regards, mpsuzuki Werner LEMBERG wrote: >>> The `default' case is redundant, since all possible enum values are >>> covered in the `switch' statement. >> I personally prefer having code in front of me that I can prove >> correct, than having to go fish for all the places that value in the >> switch statement is set to convince myself that it can't possibly >> have any other value... > > Well, I've commented this code out to fix a clang warning... Today, > compilers tell you reliably whether a `default' label is redundant or > missing, so I consider your constraints as a non-issue. > >> At any rate, having "#if 0" code like this is not a good idea as 5 >> years from now no one will remember why that code is there and why >> it's disabled. > > True. > > > Werner _______________________________________________ Freetype-devel mailing list Freetype-devel@nongnu.org https://lists.nongnu.org/mailman/listinfo/freetype-devel
