[cairo] CAIRO_STATUS_NO_MEMORY when passing incorrect scaling

[Behdad Esfahbod]

Peter Clifton wrote:
> On Thu, 2009-01-01 at 23:57 -0500, Behdad Esfahbod wrote:
>> Peter Clifton wrote:
>>> Aha.
>>>
>>> Right after the failed operation, I set cairo_matrix_identity() on the
>>> context, and still, further drawing was not functional on that context.
>>> Is that considered a known bug, or something to be fixed in the future?
>> I think I answered this already.  Yes, that's a known bug.  Yes, it will be
>> fixed in *some* future.  No, no idea when it will be fixed.  Whenever someone
>> works on it.  No one currently is.  No, it has nothing to do with backends
>> randomly setting NO_MEMORY on misc errors.
> 
> Sorry, I just wanted to make sure I had it clear. I wasn't sure if
> context errors were supposed to be transient, or whether you are just
> supposed to go get a new context once you've hit an error error. I draw
> the distinction, since originally I had thought it was the surface which
> was hitting an error - not the context.

Well, the errors are supposed to only happen on irrecoverable situations and
as a result of programming errors.  So, getting a new context typically
doesn't help.  In this case though, the error is bogus and creating a new
context "fixes" it, yes.

behdad

> Thanks for your help!
>