Re: [perl #31859] [TODO] Plain ole Hash

2005-01-02 Thread Leopold Toetsch
Bernhard Schmalhofer via RT wrote: I assume that the Hash PMC should also be used by PIR code in 'runtime/parrot/library'. In that case returning PMCNULL is not nice, as the returned PMC can't even be queried for it's type. So for now a None PMC is returned for non-existent keys. However this is no

Re: RT#31859, Plain ole Hash

2004-12-18 Thread Leopold Toetsch
Sam Ruby wrote: Any objections to the NCI methods being removed from Coroutine(next), Hash(fromkeys), Iterator(next), and PerlHash(fromkeys) now? The fromkeys in Hash and PerlHash can be removed. The .next methods in iterable PMCs will be needed generally. But iterators need some rework still, e

Re: [perl #31859] [TODO] Plain ole Hash

2004-12-18 Thread Leopold Toetsch
Bernhard Schmalhofer via RT wrote: The attached patch makes the Hash PMC use basic PMCs by default. Native integers, numbers and strings are now stored as Integer, Float and String PMCs. For keys without values, the Undef PMC is returned, as it is laid out in pdd17. As already mentioned: the Hash

[perl #31859] Re: Plain ole Hash

2004-12-18 Thread Bernhard Schmalhofer via RT
> > The attached patch makes the Hash PMC use basic PMCs by default. Sorry again, one shouldn't send any patches after midnight. First I forgot to clean up cruft from hash_pmc_20041218.patch. Then, trying to send the corrected patch, I misspelled the name of the mailing list. So, please note th

Re: RT#31859, Plain ole Hash

2004-12-17 Thread Sam Ruby
Leopold Toetsch wrote: - The NCI method has an implementation in Hash and in PerlHash, this is not nice The NCI method is automatically inherited from Hash. But as it's a python method it'll be removed from Hash anyway. Any objections to the NCI methods being removed from Coroutine(next), Hash(fro

Re: RT#31859, Plain ole Hash

2004-12-13 Thread Leopold Toetsch
Bernhard Schmalhofer <[EMAIL PROTECTED]> wrote: > Hi, > The PerlHash is now an extension of Hash. Great, thanks. > The semantics should not have changed. For example, integer values are > still stored as PerlInt PMCs in the Hash PMC. I'd propose the following: * instead of PerlInt an Integer PM

RT#31859, Plain ole Hash

2004-12-13 Thread Bernhard Schmalhofer
Hi, I have responded to the ticket, http://rt.perl.org/rt3/index.html?q=31859, but the response hasn't made if to [EMAIL PROTECTED] Is there a policy, which responses to RT tickets are posted to [EMAIL PROTECTED] I know that it works fine for new tickets. A patch for moving the implementation o

[perl #31859] [TODO] Plain ole Hash

2004-10-05 Thread via RT
# New Ticket Created by Will Coleda # Please include the string: [perl #31859] # in the subject line of all future correspondence about this issue. # http://rt.perl.org:80/rt3/Ticket/Display.html?id=31859 > Original Message Subject: Re: Plain ole Hash Date: Mon, 04

Re: Plain ole Hash

2004-10-04 Thread Dan Sugalski
At 9:56 AM -0400 10/2/04, William Coleda wrote: Are there any plans to make a Hash (as oppposed to a PerlHash, or an OrderedHash (which is really a PerlHash) ? D'oh! Knew there was something that was getting forgotten. Yes, there should be plans. If someone wants to throw a basic Hash.pmc list-wa

Plain ole Hash

2004-10-02 Thread William Coleda
Are there any plans to make a Hash (as oppposed to a PerlHash, or an OrderedHash (which is really a PerlHash) ?