On maandag 31 oktober 2011, Derek Atkins wrote: > Geert Janssens <[email protected]> writes: > > - if(strcmp(order, DEFAULT_SORT_ORDER) != 0) > > + if (order && (strcmp (order, DEFAULT_SORT_ORDER) != 0)) > > Why not use safe_strcmp()? > > -derek
I could have, but in this case only using safe_strcmp would still yield the wrong result. Using safe_strcmp, if order were NULL, then the test would evaluate to TRUE (a NULL value is always not equal to a non-null value) and the "then" part would be executed. However, in this case if order is NULL, the "then" part has to be skipped. And by the way, I don't see much use for safe_strcmp anymore in gnucash. I'd rather user g_strcmp0, which behaves exactly the same, but is part of glibc. Our code is heavily dependent on glibc anyway, so I prefer to use its convenience functions over maintaining our own. Geert _______________________________________________ gnucash-devel mailing list [email protected] https://lists.gnucash.org/mailman/listinfo/gnucash-devel
