intfiction.org

I've made new releases of the Windows Glulxe and Windows Git interpreters, available from here
http://ifarchive.org/indexes/if-archiveXprogrammingXglulxXinterpretersXglulxe.html
http://ifarchive.org/indexes/if-archiveXprogrammingXglulxXinterpretersXgit.html
http://www.davidkinder.co.uk/glulxe.html

The underlying interpreters have not changed: what is new is an update to the Glk implementation, Windows Glk. This brings support for the shiny new Glk 0.7.1 and 0.7.2 features: control over line input echoing and terminator characters; Unicode string decomposition and normalization; and access to the system's date and time.

One part of Glk 0.7.1 that has not entirely been handled is the new "no window border" flag that can be passed to the Glk function that opens windows. Windows Glk defaults to showing no borders anyway, but if window borders are turned on in the Options dialog, then the above "no window border" flag has no effect. This may be looked at at some point in the future.

I just went to the 3rd link and downloaded the Glulxe installation program, installed it, ran it, and then released, opened & ran the example programs from the Inform 7 extensions that expose the new Glk features... but they don't work. The gestalt says they aren't supported, and removing the gestalt check to assume they are anyway still doesn't work. Has the new version of Glulxe not landed on its server or something?

EDIT to add: I'm on Windows, and did not try Git.

EDIT2: oh, but downloading Glulxe's Installer from the first link works just fine.

_________________
Blog at Gamasutra :: Programmer's Guide to Inform 7 :: Seattle I-F

The link from my page is to mirror.ifarchive.org, which will re-direct you to a random IF-Archive mirror - unfortunately some of the mirrors take a bit longer to update than others.

Ah.


And thank you!

_________________
Blog at Gamasutra :: Programmer's Guide to Inform 7 :: Seattle I-F

By the way, do let me know if you find any oddities with the date/time functionality in Windows Glk. I've implemented this in terms of the Windows API, rather than using the Unix/Posix-style functions that Zarf used in CheapGlk and GlkTerm. While this makes the code nice and neat (good) it also means that there's a chance I've made a mistake somewhere (not so good).

It seems to work fine at first blush. For the code I get this (Glulxe)
Which looks correct (at the time I ran it). (This machine is 2.9 GHz, so cycles are measured in nanoseconds, not microseconds.)

Anything specific I should look out for?

_________________
Blog at Gamasutra :: Programmer's Guide to Inform 7 :: Seattle I-F

No, nothing specific. I'm just aware that Windows Glk is probably the only Glk implementation that won't copy Zarf's code verbatim.

Oh, I'll use the native Cocoa toolkit for the iOS version. Just for the sake of variety.

The unit test, again, is at http://eblong.com/zarf/glulx/datetimetest.ulx .

Windows Glk seems okay with your unit test, so hopefully it's right

Thank you, David.

_________________
Eliuk Blau
eliukblau (AT) gmail.com
http://www.caad.es/eliukblau/