Sam Phillips wrote:
> On 4/5/07, Paolo Bonzini <address@hidden> wrote:
>>
>> > My guess is that the gst syntax doesn't like '*' as the initial
>> > character of a variable symbol.
>>
>> Yes, the parameter name should have been "str" (munged from "*str"),
>> it is one byte off.
>>
>> > Is this a known bug?
>>
>> No, it is new. Can you send me your gdate.h header file?
>
> It doesn't just seem to be messing with the gdate functions I've also
> gotten this gem in my gtk/Funcs.st generated:
>
> !GOptionContext methodsFor: 'C call-outs'!
> parse: *arg ***a: ***a **err: **err
> <cCall: 'g_option_context_parse' returning: #boolean
> args: #( #self #cObject #cObject #cObject )>! !
>
> Most of the functions that use any kind of pointer type seem to be
> rewritten strangely.
I cannot reproduce this. Can you:
1) "grep AWK gtk/Makefile" to see which awk implementation you're using
2) do a "gawk --version" (where gawk is the output of the previous
command actually) to see what version it is.