Re: Parrot::Coroutine

2006-10-24 Thread Bob Rogers
From: François PERRAD <[EMAIL PROTECTED]> Date: Mon, 23 Oct 2006 18:51:03 +0200 In languages/lua/lib/thread.pir, I create a Lua thread type by extension of Parrot::Coroutine. So I add a lot of methods for Lua type, but I think that 2 of these methods could be integra

Re: Parrot::Coroutine

2006-10-23 Thread François PERRAD
In languages/lua/lib/thread.pir, I create a Lua thread type by extension of Parrot::Coroutine. So I add a lot of methods for Lua type, but I think that 2 of these methods could be integrated in Parrot::Coroutine : - __clone - __get_pointer (equivalent of get_pointer() in src/pmc/sub.pmc

Re: Parrot::Coroutine

2006-10-17 Thread Leopold Toetsch
Am Dienstag, 17. Oktober 2006 10:47 schrieb François PERRAD: > Now, Lua doesn't depend of the deprecated Coroutine PMC. Coroutine PMC isn't deprecated, it's limited in its functionality. leo

Re: Parrot::Coroutine

2006-10-17 Thread François PERRAD
At 13:48 14/10/2006 -0400, Bob Rogers wrote: So while we could redesign Coroutine.pmc now, I don't think we are ready to reimplement it. In the mean time, having an alternative implementation allows those who want a full implementation (hi, Fran,Ag(Bois!) to get on with it. I've removed th

Re: Parrot::Coroutine

2006-10-14 Thread Bob Rogers
From: Leopold Toetsch <[EMAIL PROTECTED]> Date: Sat, 14 Oct 2006 23:47:13 +0200 Am Samstag, 14. Oktober 2006 23:20 schrieb Bob Rogers: > This sounds more like a coroutine tutorial to me.  I was hoping that the > 'SEE ALSO' links in the source would cover that.  Still, I agree it

Re: Parrot::Coroutine

2006-10-14 Thread Leopold Toetsch
Am Samstag, 14. Oktober 2006 23:20 schrieb Bob Rogers: > This sounds more like a coroutine tutorial to me.  I was hoping that the > 'SEE ALSO' links in the source would cover that.  Still, I agree it > would be good to have a detailed explanation of how coroutines work in > Parrot terms. > >    The

Re: Parrot::Coroutine

2006-10-14 Thread Leopold Toetsch
Am Samstag, 14. Oktober 2006 04:21 schrieb Bob Rogers: A further note, while the patch has hit svn: this namespace > +.namespace ["Parrot::Coroutine"] is actually the absolute namespace: [ 'parrot'; 'Parrot::Coroutine' ] which isn't importable by a lot

Re: Parrot::Coroutine

2006-10-14 Thread Bob Rogers
more ] Thanks for the detailed explanation. I now see the point of a Parrot::Coroutine thingy. Great; committed as r14925. OTOH I'd still like to have some proposals explaining all the planned use cases (incl. HLL examples) what some more flexible coroutines could do. Some

Re: Parrot::Coroutine

2006-10-14 Thread Leopold Toetsch
Am Samstag, 14. Oktober 2006 19:48 schrieb Bob Rogers: > True, but this is something different.  Coroutine.pmc is limited in a > way that can't be cured by extension. [ and a lot more ] Thanks for the detailed explanation. I now see the point of a Parrot::Coroutine thingy. OTO

Re: Parrot::Coroutine

2006-10-14 Thread Bob Rogers
From: Leopold Toetsch <[EMAIL PROTECTED]> Date: Sat, 14 Oct 2006 13:22:07 +0200 Am Samstag, 14. Oktober 2006 04:21 schrieb Bob Rogers: > Diffs between last version checked in and current workfile(s): Index: compilers/pge/PGE/Regex.pir ==

Re: Parrot::Coroutine

2006-10-14 Thread Leopold Toetsch
Am Samstag, 14. Oktober 2006 04:21 schrieb Bob Rogers: >    Here's the Parrot::Coroutine patch; I will commit this if nobody > objects. Please note that you can extend existing PMCs with PIR methods. >    It seems to me that this should be mentioned in docs/compiler_faq.pod >

Parrot::Coroutine

2006-10-13 Thread Bob Rogers
Here's the Parrot::Coroutine patch; I will commit this if nobody objects. It seems to me that this should be mentioned in docs/compiler_faq.pod as an alternative way to do coroutines . . . I will take a look at this. In fact, it might be safer to deprecate Coroutine.pmc until it c