New release: Glulxe 0.4.5

[Andrew Plotkin]

I've posted a new source package for Glulxe.

Like the recent Git release, this fixes various obscure bugs that have
been revealed by unit-testing. I've also made the definition of the
random-number functions more consistent.

The biggest change in this release (which is still not very big) is a
warning which will appear if you call STRING.print_to_array(a) in I6
code. As described in recent posts, you should always use the
two-argument form of this method.

(The I7 compiler, library, and extensions do not make this mistake, so
this change will not affect the vast majority of I7 authors or games.
If you find a game displaying this warning:

Memory access was much too long -- perhaps a print_to_array call with only one argument

...then you should contact the author and point it out.)

If you want to test this warning in your Glulxe build (or other
interpreter), you can download this game file:
<http://eblong.com/zarf/glulx/arraylimittest.ulx>
It conveniently makes the mistake for you.

--Z

--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*

[David Kinder]

There's now a Windows build too, available from the IF Archive:
http://ifarchive.org/indexes/if-archiveXprogrammingXglulxXinterpretersXglulxe.html

David

[Eliuk Blau]

On 24 nov, 15:23, David Kinder <da...@david.david> wrote:
> There's now a Windows build too, available from the IF Archive:http://ifarchive.org/indexes/if-archiveXprogrammingXglulxXinterpreter...
>
> David

Thanks Andrew and David! =D

[Peter Pears]

I've been thinking about bringing this up, and since there's a handy
thread here I thought I might as well.

I use IFMapper (the one with automatic mapping) for my IF needs - or
at least my IF needs when playing through an interpreter that outputs
to a transcript file as soon as text gets displayed on screen, like
WindowsFrotz does. WinGit doesn't do it, though, and neither does
WinGlulxe.

Sure, it's only a matter of small convenience, but I wonder if I could
persuade the maintainers of WinGit and WinGlulxe (which I believe is,
in both cases, David Kinder) to make it so that, like in WindowsFrotz,
the text in a transcript gets updated in real-time, instead of waiting
for a set number of moves, or characters, or what-have-you. Using
IFMapper has eased my IF experience to a degree I didn't foresee, and
it's quite indispensable (though for standard non-automapping I still
prefer GUEMap).

[David Kinder]

Peter Pears wrote:
> I use IFMapper (the one with automatic mapping) for my IF needs - or
> at least my IF needs when playing through an interpreter that outputs
> to a transcript file as soon as text gets displayed on screen, like
> WindowsFrotz does. WinGit doesn't do it, though, and neither does
> WinGlulxe.

I've had a quick glance at that code, and it's not obvious why Frotz behaves
any differently from Glulxe or Git. I'll make a note to look into this more
thoroughly when I have the chance.

David