On Thu, Oct 07, 2004 at 04:54:13PM -0400, Allan Edwards wrote: > Joe Orton wrote: > >Why is this a macro? It's not like apr_uint32_t is a name which is going > >to change any time soon? > > It's a macro because we don't want to lose sight of the > fact that we did something there that we utimately want to > back out (i.e. in APR 2.0 when we can change the API). If > we used apr_uint32_t it would be easy to lose track of > these places - make sense?
The fact that these casts are unnecessary can be tracked using comments, a list of issues in STATUS, or, not wishing to rock the boat, bugzilla. Using a macro just obfuscates the code. joe
