[Swift-devel] iterate behaviour round II

[Michael Wilde]

Cool.  Related to current semantics, Im seeing a case where iterate seems to not terminate correctly with an == test but does with a > test. Is there some float funkiness going on in there too?

">" termination condition works OK:

com$ cat iterategt.swift
iterate i
{
   trace(i);
}  until(i > 3);
com$ swift iterategt.swift | head -15
no sites file specified, setting to default: /scratch/local/wilde/swift/src/devtrunk/cog/modules/swift/dist/swift-svn/etc/sites.xml
Swift svn swift-r4934 (swift modified locally) cog-r3184 (cog modified locally)

RunID: 20110803-1315-qn3cr6s8
Progress:  time: Wed, 03 Aug 2011 13:15:38 -0500
SwiftScript trace: 0
SwiftScript trace: 1
SwiftScript trace: 2
SwiftScript trace: 3
Final status:  time: Wed, 03 Aug 2011 13:15:38 -0500
com$ 

"==" termination condition never terminates:

com$ 
com$ cat iterateeq.swift
iterate i
{
   trace(i);
}  until(i == 3);
com$ 

com$ swift iterateeq.swift | head -15
no sites file specified, setting to default: /scratch/local/wilde/swift/src/devtrunk/cog/modules/swift/dist/swift-svn/etc/sites.xml
Swift svn swift-r4934 (swift modified locally) cog-r3184 (cog modified locally)

RunID: 20110803-1316-f9qhsxig
SwiftScript trace: 0
Progress:  time: Wed, 03 Aug 2011 13:16:06 -0500
SwiftScript trace: 1
SwiftScript trace: 2
SwiftScript trace: 3
SwiftScript trace: 4
SwiftScript trace: 5
SwiftScript trace: 6
SwiftScript trace: 7
SwiftScript trace: 8
SwiftScript trace: 9
SwiftScript trace: 10

^C
com$ swift -version
no sites file specified, setting to default: /scratch/local/wilde/swift/src/devtrunk/cog/modules/swift/dist/swift-svn/etc/sites.xml
Swift svn swift-r4934 (swift modified locally) cog-r3184 (cog modified locally)

com$ which swift
/scratch/local/wilde/swift/src/devtrunk/cog/modules/swift/dist/swift-svn/bin/swift
com$ 



----- Original Message -----
> From: "Mihael Hategan" <hategan at mcs.anl.gov>
> To: "Michael Wilde" <wilde at mcs.anl.gov>
> Cc: "Swift Devel" <swift-devel at ci.uchicago.edu>
> Sent: Wednesday, August 3, 2011 12:56:25 PM
> Subject: Re: [Swift-devel] iterate behaviour round II
> On Wed, 2011-08-03 at 12:53 -0500, Michael Wilde wrote:
> > Right. So lets keep iterate semantically unchanged for now, then
> > deprecate it when we have an approach thats clearly better. For 0.93
> > lets focus on making it work as currently described.
> 
> I agree. I will revert the change I did before.

-- 
Michael Wilde
Computation Institute, University of Chicago
Mathematics and Computer Science Division
Argonne National Laboratory