Re: popt
- From: Derek Simkowiak <dereks kd-dev com>
- To: Havoc Pennington <hp redhat com>
- Cc: gtk-devel-list gnome org
- Subject: Re: popt
- Date: Mon, 8 May 2000 12:43:23 -0700 (PDT)
-> - has error reporting; we still lack a standard glib error
-> reporting setup
Can anyone recommend a good reference for structured
error-handling techniques?
In my own code, I've used the following (simple out-of-band)
rules:
If the function returns a pointer, NULL means an error occured. This has
the problem that sometimes a pointer-returning function can return NULL as
a valid value, thus breaking the convention.
If the function returns a value that could be a guint (such as
length, time, or any other value where a negative does not make sense) I
have the function return a gint, where <0 indicates an error occurred.
This has the advantage that you can use enumerated error codes, i.e. -1
means one thing, -2 means something else, etc. In my own code, I don't
usually go as far as enumerated error values, but it's possible...
In functions that would normally return void, I have it return a gint
instead, where <0 indicates an error occurred (or 0 for success).
These are obviously very simple rules, but short of orchestrating
some elaborate message-passing, exception-throwing architecture, what else
can you do with C?
-> Thoughts? Maybe we should fork off the codebase? That sort of sucks on
-> one hand, but on the other hand it's only 600 lines of code.
(My $.02: I would have no issue with forking from popt...)
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]