Re: g_error_free() warning on null pointer

[Michael McConville <mmcconv1 sccs swarthmore edu>]

Jasper St. Pierre wrote:
> Michael McConville wrote:
> > Nicolas George wrote:
> > > But maybe I am forgetting another case: can you imagine a code snippet
> > > where g_error_free(error) would make sense with error == NULL?
> >
> > I may have already mentioned this, but the simplest example is just
> > adding a g_error_free() at the end of a function when adding a GError*
> > declaration at the beginning. This prevents an irrelevant line from
> > cluttering the error-handling logic and makes it more trivially obvious
> > to auditors that there isn't a memory leak.
> But if you have an error, shouldn't you either want to a) handle it,
> or b) propagate it up?

I suspect that this is almost always the case. However:

 1) Sometimes you have to audit other people's code for memory leaks but
    don't have the time/understanding for logic additions.

 2) Even if you are handling the error, the free call is clutter.
    Keeping it in the function's footer contributes readability.