On Tue, Nov 29, 2016 at 6:14 AM, Matthew Flatt <[email protected]> wrote:
> > Otherwise, I agree that using the `errno` field is the way to go when > you have an `exn:fail:filesystem:errno` exception. The `lookup-errno` > function provided by `ffi/unsafe` can be helpful for recognizing Posix > errors more portably. > lookup-errno will go from a symbol to a code, but the ...:errno exception provides the code. Is there a way to go the other direction? > > The built-in exception table is currently defined by "makeexn" in > "racket/src/racket/src/", and various primitives the runtime system > explicitly raise instances of those exceptions. > > We could try to translate more OS error cores into new subtrees of the > hierarchy, but that has never seemed like an especially good idea > compared to exposing the error codes in a more generic and automatic > way. If there are useful distinctions that aren't reflected by `errno`, > then those cases are certainly worth a closer look. > Yeah, it's not essential. Between the errno and the message I can figure out what's going on. It's a little irksome, but it's fine. -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
