On 6/1/06, James Antill <[EMAIL PROTECTED]> wrote: > "Axel Liljencrantz" <[EMAIL PROTECTED]> writes: > > > I've released fish 1.21.7. > > This is actually against 1.21.8, even though I've not seen that email > yet (freshmeat sent me one :).
Thanks for the reminder! I'd forgotten. > > There appear to be a few 64bit unclean lines in fish[1]: Careless of me. > > % uname -i > x86_64 > % ( ./configure && make ) > /dev/null > expand.c: In function 'expand_variables': > expand.c:804: warning: cast to pointer from integer of different size > expand.c:817: warning: cast from pointer to integer of different size > highlight.c: In function 'highlight_universal_internal': > highlight.c:759: warning: cast from pointer to integer of different size > history.c: In function 'history_test': > history.c:485: warning: cast from pointer to integer of different size > sanity.c: In function 'validate_pointer': > sanity.c:63: warning: cast from pointer to integer of different size > Notice: Output directory `builtin_doc' does not exist. I have created it for > you. > fish_pager.c: In function 'join_completions': > fish_pager.c:739: warning: cast to pointer from integer of different size > > ...all but history.c are storing ints in void *s, which is ugly but > probably works and isn't trivial to fix, but history has: Yep. Almost makes you miss C++ templates. > > return (int)wcsstr( haystack, needle ); > > ...which is almost certainly more idiomatic as: > > return !!wcsstr( haystack, needle ); > > ...and doesn't have a problem when the lower 32bits of the return > value are zero. Didn't your mother teach you not to use double negations? ;-) I guess that is cleaner, but it _does_ look a bit weird to me. > > [1] Please forgive the use of zsh to compile fish ;). ... Ok. I officially forgive you. > > -- > James Antill -- [EMAIL PROTECTED] > http://www.and.org/and-httpd > -- Axel _______________________________________________ Fish-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/fish-users
