Comments for lambda.oasis :: poucet -> content http://cpoucet.wordpress.com Thoughts of an expat geek Thu, 10 Sep 2009 13:40:56 +0000 http://wordpress.com/ hourly 1 Comment on Change of blog by sandrar http://cpoucet.wordpress.com/2008/02/13/hello-world/#comment-174 sandrar Thu, 10 Sep 2009 13:40:56 +0000 #comment-174 Hi! I was surfing and found your blog post... nice! I love your blog. :) Cheers! Sandra. R. Hi! I was surfing and found your blog post… nice! I love your blog. :) Cheers! Sandra. R.

]]> Comment on Flattening Data.Map by Edward Kmett http://cpoucet.wordpress.com/2009/04/18/flattening-datamap/#comment-170 Edward Kmett Mon, 20 Apr 2009 14:57:39 +0000 http://cpoucet.wordpress.com/?p=120#comment-170 Actually, the issue with TH is simpler than that. It currently doesn't support type and data families. I ran into that issue when I started doing this same thing to Data.Set. I love the idea of abusing the views for flattening. I've been working towards that same goal with a flattened trie implementation. Normally a ternary search tree based trie that winds up with a bunch of Maybe nodes floating around to keep the implementation simple, but I can flatten the cases into the constructors and only extract them during view. I've also been exploring if I can store the trie internally in compact PATRICIA form, but expose only a simple view that is ignorant of this storage optimization. Actually, the issue with TH is simpler than that. It currently doesn’t support type and data families.

I ran into that issue when I started doing this same thing to Data.Set. I love the idea of abusing the views for flattening. I’ve been working towards that same goal with a flattened trie implementation. Normally a ternary search tree based trie that winds up with a bunch of Maybe nodes floating around to keep the implementation simple, but I can flatten the cases into the constructors and only extract them during view. I’ve also been exploring if I can store the trie internally in compact PATRICIA form, but expose only a simple view that is ignorant of this storage optimization.

]]> Comment on Flattening Data.Map by cpoucet http://cpoucet.wordpress.com/2009/04/18/flattening-datamap/#comment-169 cpoucet Sun, 19 Apr 2009 21:27:55 +0000 http://cpoucet.wordpress.com/?p=120#comment-169 We're currently working on some benchmarks that will give us these numbers. There are no benchmarks out there that we know of for associative containers. We’re currently working on some benchmarks that will give us these numbers. There are no benchmarks out there that we know of for associative containers.

]]> Comment on Flattening Data.Map by Dafydd Harries http://cpoucet.wordpress.com/2009/04/18/flattening-datamap/#comment-168 Dafydd Harries Sun, 19 Apr 2009 19:48:54 +0000 http://cpoucet.wordpress.com/?p=120#comment-168 It would be nice to see some numbers on what differences these sorts of changes can achieve. It would be nice to see some numbers on what differences these sorts of changes can achieve.

]]> Comment on Flattening Data.Map by Felipe Lessa http://cpoucet.wordpress.com/2009/04/18/flattening-datamap/#comment-167 Felipe Lessa Sun, 19 Apr 2009 11:37:57 +0000 http://cpoucet.wordpress.com/?p=120#comment-167 Hmmm... but at the very least -funbox-strict-fields could be applied, maybe with a file pragma (not to force the entire library/program to use it). Hmmm… but at the very least -funbox-strict-fields could be applied, maybe with a file pragma (not to force the entire library/program to use it).

]]> Comment on Flattening Data.Map by cpoucet http://cpoucet.wordpress.com/2009/04/18/flattening-datamap/#comment-166 cpoucet Sun, 19 Apr 2009 09:56:07 +0000 http://cpoucet.wordpress.com/?p=120#comment-166 We were actually wondering whether TH could be used for this. The problem is that we want the {-# UNPACK #-} specifications, and I'm not sure you can force that sort of annotations with TH. If anyone knows whether this is possible, that would be great. We were actually wondering whether TH could be used for this. The problem is that we want the {-# UNPACK #-} specifications, and I’m not sure you can force that sort of annotations with TH.

If anyone knows whether this is possible, that would be great.

]]> Comment on Flattening Data.Map by Felipe Lessa http://cpoucet.wordpress.com/2009/04/18/flattening-datamap/#comment-165 Felipe Lessa Sun, 19 Apr 2009 02:23:36 +0000 http://cpoucet.wordpress.com/?p=120#comment-165 It seems like TH could be used to generete the instances automagically. It seems like TH could be used to generete the instances automagically.

]]> Comment on Bootstrapping cabal by cpoucet http://cpoucet.wordpress.com/2009/03/12/bootstrapping-cabal/#comment-154 cpoucet Thu, 12 Mar 2009 17:10:22 +0000 http://cpoucet.wordpress.com/?p=96#comment-154 I have not looked at the code of bootstrap.sh, but I can say that at the time, bootstrap.sh did not work for me while this did. I have not looked at the code of bootstrap.sh, but I can say that at the time, bootstrap.sh did not work for me while this did.

]]> Comment on Bootstrapping cabal by liesen http://cpoucet.wordpress.com/2009/03/12/bootstrapping-cabal/#comment-153 liesen Thu, 12 Mar 2009 12:28:48 +0000 http://cpoucet.wordpress.com/?p=96#comment-153 doesn't bootstrap.sh (bundled with cabal-installed) do exactly this? doesn’t bootstrap.sh (bundled with cabal-installed) do exactly this?

]]> Comment on Bootstrapping cabal by lilac http://cpoucet.wordpress.com/2009/03/12/bootstrapping-cabal/#comment-152 lilac Thu, 12 Mar 2009 12:07:39 +0000 http://cpoucet.wordpress.com/?p=96#comment-152 FYI, the 'cabal-install' tarball on hackage includes a script 'bootstrap.sh' which does much the same as this (but seems to get by installing fewer packages). FYI, the ‘cabal-install’ tarball on hackage includes a script ‘bootstrap.sh’ which does much the same as this (but seems to get by installing fewer packages).

]]>