From naesten at gmail.com Thu Apr 3 01:16:08 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 3 01:12:04 2008 Subject: [jhc] darcs patch: Mark output from -ddatatable Message-ID: Thu Apr 3 01:15:09 EDT 2008 Samuel Bronson * Mark output from -ddatatable -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 7808 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080403/ad1ee76a/attachment.bin From naesten at gmail.com Thu Apr 3 19:58:30 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 3 19:54:22 2008 Subject: [jhc] patches to fix *haskell* typechecking of (->) Message-ID: Okay, now before you get all excited, I must warn you that you still won't be able to compile Control.Arrow. There is a good bit of work that must be done in E before this will, in fact, compile. Much of this work has to do with turning fully-applied Jhc.Basics.->s into EPis, but there also seems to be some kind of problem with multi-parameter newtypes... From naesten at gmail.com Thu Apr 3 20:02:59 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 3 19:58:53 2008 Subject: [jhc] Re: patches to fix *haskell* typechecking of (->) In-Reply-To: References: Message-ID: Hmm, do you suppose I should have attached the patches? -------------- next part -------------- A non-text attachment was scrubbed... Name: fix-typechecking.dpatch Type: application/octet-stream Size: 21502 bytes Desc: not available Url : http://www.haskell.org/pipermail/jhc/attachments/20080403/cf38ddc8/fix-typechecking-0001.obj From john at repetae.net Thu Apr 3 20:14:25 2008 From: john at repetae.net (John Meacham) Date: Thu Apr 3 20:10:17 2008 Subject: [jhc] patches to fix *haskell* typechecking of (->) In-Reply-To: References: Message-ID: <20080404001425.GA29316@sliver.repetae.net> On Thu, Apr 03, 2008 at 07:58:30PM -0400, Samuel Bronson wrote: > Okay, now before you get all excited, I must warn you that you still > won't be able to compile Control.Arrow. There is a good bit of work > that must be done in E before this will, in fact, compile. Much of > this work has to do with turning fully-applied Jhc.Basics.->s into > EPis, but there also seems to be some kind of problem with > multi-parameter newtypes... Could you give a summary of what these patches specifically fix about type checking? It is not immediatly obvious to me from looking at the patches. Also, could you add 'send --unified' to your .darcs/defaults, that will make patches you send me be in unified format which is easier to read. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Thu Apr 3 20:18:07 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 3 20:14:11 2008 Subject: [jhc] darcs patch: Fix kind inference + test (and 6 more) Message-ID: Mon Mar 31 07:28:01 EDT 2008 Samuel Bronson * Fix kind inference + test Also, comment out the case in kindOf that makes things look fine when they aren't -- unless of course * isn't the right kind Mon Mar 31 11:18:39 EDT 2008 Samuel Bronson * Work around type inference bug relating to main (I didn't do it!) Fri Mar 21 00:29:22 EDT 2008 Samuel Bronson * Split TArrows where appropriate Thu Mar 20 16:42:02 EDT 2008 Samuel Bronson * Actually define (->)... Tue Mar 18 18:43:45 EDT 2008 Samuel Bronson * Add Control.Monad.Instances to base package Tue Mar 18 18:56:40 EDT 2008 Samuel Bronson * Add Control.Category Thu Mar 20 17:33:26 EDT 2008 Samuel Bronson * Make Control.Category actually not import Prelude... -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 19293 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080403/53031036/attachment.bin From naesten at gmail.com Thu Apr 3 20:47:28 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 3 20:43:19 2008 Subject: [jhc] darcs patch: Fix kind inference + test (and 6 more) In-Reply-To: References: Message-ID: On Thu, Apr 3, 2008 at 7:18 PM, Samuel Bronson wrote: > Mon Mar 31 07:28:01 EDT 2008 Samuel Bronson > * Fix kind inference + test > Also, comment out the case in kindOf that makes things look fine when > they aren't -- unless of course * isn't the right kind This one makes the kind inference actually run when FrontEnd.KindInfer.hsTypeToType, and makes kindOf stop reporting * when the name in question is not in the environment -- signifying that kind inference has definitely not seen it. The former behaviour was useless, serving only to make things appear to work when they in fact did not, merely because * happened to be the correct kind in most cases. > Mon Mar 31 11:18:39 EDT 2008 Samuel Bronson > * Work around type inference bug relating to main (I didn't do it!) This one just makes the test case for the previous patch actually run. > Fri Mar 21 00:29:22 EDT 2008 Samuel Bronson > * Split TArrows where appropriate This one breaks apart TArrows when they must be matched with other things in type inference. > Thu Mar 20 16:42:02 EDT 2008 Samuel Bronson > * Actually define (->)... This establishes the kind of (->) as ?? -> ? -> *, as copied from GHCi output for :kind (->). > Tue Mar 18 18:43:45 EDT 2008 Samuel Bronson > * Add Control.Monad.Instances to base package > > Tue Mar 18 18:56:40 EDT 2008 Samuel Bronson > * Add Control.Category > > Thu Mar 20 17:33:26 EDT 2008 Samuel Bronson > * Make Control.Category actually not import Prelude... The first two are needed by the still-failing testcase; the third one just makes it build faster by side-stepping Prelude. I seem to have left some patches out this time, though... From naesten at gmail.com Thu Apr 3 21:01:53 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 3 20:57:47 2008 Subject: [jhc] darcs patch: Fix kind inference + test (and 6 more) In-Reply-To: References: Message-ID: > I seem to have left some patches out this time, though... Okay, this one has the final patch -- this one mostly replaces hard-coded TAp constructors with tAp invocations -- tAp returns a TArrow instead of a TAp in the appropriate case. -------------- next part -------------- A non-text attachment was scrubbed... Name: fix-typechecking.dpatch Type: application/octet-stream Size: 28587 bytes Desc: not available Url : http://www.haskell.org/pipermail/jhc/attachments/20080403/04dc6f7c/fix-typechecking-0001.obj From john at repetae.net Thu Apr 3 21:06:50 2008 From: john at repetae.net (John Meacham) Date: Thu Apr 3 21:02:41 2008 Subject: [jhc] darcs patch: Fix kind inference + test (and 6 more) In-Reply-To: References: Message-ID: <20080404010650.GA31945@sliver.repetae.net> On Thu, Apr 03, 2008 at 07:47:28PM -0500, Samuel Bronson wrote: > On Thu, Apr 3, 2008 at 7:18 PM, Samuel Bronson wrote: > > Mon Mar 31 07:28:01 EDT 2008 Samuel Bronson > > * Fix kind inference + test > > Also, comment out the case in kindOf that makes things look fine when > > they aren't -- unless of course * isn't the right kind > > This one makes the kind inference actually run when > FrontEnd.KindInfer.hsTypeToType, and makes kindOf stop reporting * > when the name in question is not in the environment -- signifying that > kind inference has definitely not seen it. The former behaviour was > useless, serving only to make things appear to work when they in fact > did not, merely because * happened to be the correct kind in most > cases. This was due to the kind inference algorithm specified in the report, any unconstrained kinds are supposed to be defaulted to *. This may not be necessary anymore though as I think I added an explicit defaulting step. In any case, It wasn't put there by accident or a hack even if it is no longer needed. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Thu Apr 3 21:31:33 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 3 21:27:26 2008 Subject: [jhc] darcs patch: Fix kind inference + test (and 6 more) In-Reply-To: <20080404010650.GA31945@sliver.repetae.net> References: <20080404010650.GA31945@sliver.repetae.net> Message-ID: On Thu, Apr 3, 2008 at 9:06 PM, John Meacham wrote: > > This one makes the kind inference actually run when > > FrontEnd.KindInfer.hsTypeToType, and makes kindOf stop reporting * > > when the name in question is not in the environment -- signifying that > > kind inference has definitely not seen it. The former behaviour was > > useless, serving only to make things appear to work when they in fact > > did not, merely because * happened to be the correct kind in most > > cases. > > This was due to the kind inference algorithm specified in the report, > any unconstrained kinds are supposed to be defaulted to *. This may not > be necessary anymore though as I think I added an explicit defaulting > step. In any case, It wasn't put there by accident or a hack even if it > is no longer needed. Well, that's what I thought it must be an attempt at. But in fact there is other code that does a much better job -- or at least, it does it in a way that provides less false comfort. The other code only defaults kinds for things that have made it into the kind environment. Anyway, the rule of thumb for compilers and RTSs is "crash ASAP", so... From john at repetae.net Thu Apr 3 21:50:20 2008 From: john at repetae.net (John Meacham) Date: Thu Apr 3 21:46:12 2008 Subject: [jhc] darcs patch: Fix kind inference + test (and 6 more) In-Reply-To: References: <20080404010650.GA31945@sliver.repetae.net> Message-ID: <20080404015020.GB31945@sliver.repetae.net> On Thu, Apr 03, 2008 at 09:31:33PM -0400, Samuel Bronson wrote: > Well, that's what I thought it must be an attempt at. But in fact > there is other code that does a much better job -- or at least, it > does it in a way that provides less false comfort. The other code only > defaults kinds for things that have made it into the kind environment. > Anyway, the rule of thumb for compilers and RTSs is "crash ASAP", > so... Yeah, the original kind checker was out of the THIH in haskell paper, where only simple kinds were allowed so just defaulting to * was a straightforward way to do it. When I added the kind constraint stuff, it became necessary to be a bit smarter in the defaulting stage so that code was no longer needed. You will still find remenants of the old THIH kind checker and type checker around, though I have rewritten both a couple times since then, some code constructs and terminology has a way of sticking around. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Fri Apr 4 14:42:46 2008 From: naesten at gmail.com (Samuel Bronson) Date: Fri Apr 4 14:38:36 2008 Subject: [jhc] So, what's going on with my patches right now? Message-ID: Tried them? Any comments? Advice? Questions? From john at repetae.net Mon Apr 7 08:10:14 2008 From: john at repetae.net (John Meacham) Date: Mon Apr 7 08:05:53 2008 Subject: [jhc] So, what's going on with my patches right now? In-Reply-To: References: Message-ID: <20080407121014.GG31945@sliver.repetae.net> On Fri, Apr 04, 2008 at 02:42:46PM -0400, Samuel Bronson wrote: > Tried them? Any comments? Advice? Questions? Oh, sorry. I have been prettty busy the last bit so have not gotten a chance to look too critically at them yet. I am going to add them in, but I am holding off on the patches adding stuff taken from ghc base to the standard libraries because I am working on a method to build those directly out of the ghc repo now so we don't need to replicate a lot of code. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Mon Apr 7 10:00:29 2008 From: naesten at gmail.com (Samuel Bronson) Date: Mon Apr 7 09:56:21 2008 Subject: [jhc] So, what's going on with my patches right now? In-Reply-To: <20080407121014.GG31945@sliver.repetae.net> References: <20080407121014.GG31945@sliver.repetae.net> Message-ID: On 4/7/08, John Meacham wrote: > On Fri, Apr 04, 2008 at 02:42:46PM -0400, Samuel Bronson wrote: > > Tried them? Any comments? Advice? Questions? > > Oh, sorry. I have been prettty busy the last bit so have not gotten a > chance to look too critically at them yet. I am going to add them in, > but I am holding off on the patches adding stuff taken from ghc base to > the standard libraries because I am working on a method to build those > directly out of the ghc repo now so we don't need to replicate a lot of > code. That's a mighty fine excuse ;-). Btw, I hope your method involves GHC's base package keeping the proper name, because without the name "base" on the "base" package, Cabal has little hope of working. (Did you know that Cabal is supposed to have JHC support? Neither did I, until dcoutts mentioned that I was welcome to fix the JHC support since he hadn't been able to test it...) From naesten at gmail.com Wed Apr 9 21:02:38 2008 From: naesten at gmail.com (Samuel Bronson) Date: Wed Apr 9 20:58:10 2008 Subject: [jhc] Doc repository Message-ID: Having the Doc repository seperate from the main jhc repository is beginning to cramp my style... I'm trying to get line breaks in sensible places in E code, and some of it seems to need coordinated changes in the Doc repository. Or might. Would it be possible to move that code into the main repository? From john at repetae.net Thu Apr 10 00:06:43 2008 From: john at repetae.net (John Meacham) Date: Thu Apr 10 00:02:14 2008 Subject: [jhc] Doc repository In-Reply-To: References: Message-ID: <20080410040643.GD25982@sliver.repetae.net> On Wed, Apr 09, 2008 at 09:02:38PM -0400, Samuel Bronson wrote: > Having the Doc repository seperate from the main jhc repository is > beginning to cramp my style... I'm trying to get line breaks in > sensible places in E code, and some of it seems to need coordinated > changes in the Doc repository. Or might. Would it be possible to move > that code into the main repository? I'd prefer not to, I use 'Doc' in a number of different projects and it is meant to be independent of jhc. I don't like copying source into the jhc repo that isn't jhc itself. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Thu Apr 10 13:31:37 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 10 13:27:10 2008 Subject: [jhc] Doc repository In-Reply-To: <20080410040643.GD25982@sliver.repetae.net> References: <20080410040643.GD25982@sliver.repetae.net> Message-ID: On 4/10/08, John Meacham wrote: > On Wed, Apr 09, 2008 at 09:02:38PM -0400, Samuel Bronson wrote: > > Having the Doc repository seperate from the main jhc repository is > > beginning to cramp my style... I'm trying to get line breaks in > > sensible places in E code, and some of it seems to need coordinated > > changes in the Doc repository. Or might. Would it be possible to move > > that code into the main repository? > > I'd prefer not to, I use 'Doc' in a number of different projects and it > is meant to be independent of jhc. I don't like copying source into the > jhc repo that isn't jhc itself. Fine. I think I figured out how to do what I want without changes in Doc anyway... From john at repetae.net Thu Apr 10 16:51:48 2008 From: john at repetae.net (John Meacham) Date: Thu Apr 10 16:47:17 2008 Subject: [jhc] Doc repository In-Reply-To: References: <20080410040643.GD25982@sliver.repetae.net> Message-ID: <20080410205147.GF25982@sliver.repetae.net> On Thu, Apr 10, 2008 at 01:31:37PM -0400, Samuel Bronson wrote: > On 4/10/08, John Meacham wrote: > > On Wed, Apr 09, 2008 at 09:02:38PM -0400, Samuel Bronson wrote: > > > Having the Doc repository seperate from the main jhc repository is > > > beginning to cramp my style... I'm trying to get line breaks in > > > sensible places in E code, and some of it seems to need coordinated > > > changes in the Doc repository. Or might. Would it be possible to move > > > that code into the main repository? > > > > I'd prefer not to, I use 'Doc' in a number of different projects and it > > is meant to be independent of jhc. I don't like copying source into the > > jhc repo that isn't jhc itself. > > Fine. I think I figured out how to do what I want without changes in > Doc anyway... Yeah, the layout of core has always been a little tricky. On the other hand, it is good motivation for code simplifying transformations. :) John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Fri Apr 11 06:50:50 2008 From: naesten at gmail.com (Samuel Bronson) Date: Fri Apr 11 06:46:21 2008 Subject: [jhc] Doc repository In-Reply-To: <20080410205147.GF25982@sliver.repetae.net> References: <20080410040643.GD25982@sliver.repetae.net> <20080410205147.GF25982@sliver.repetae.net> Message-ID: On Thu, Apr 10, 2008 at 4:51 PM, John Meacham wrote: > Yeah, the layout of core has always been a little tricky. On the other > hand, it is good motivation for code simplifying transformations. Well, with such long names, I don't know if that will help much ;-) From naesten at gmail.com Fri Apr 11 17:18:03 2008 From: naesten at gmail.com (Samuel Bronson) Date: Fri Apr 11 17:13:59 2008 Subject: [jhc] darcs patch: Use $(HC) --interactive instead of ghci, in case a non... Message-ID: Thu Apr 10 15:34:13 EDT 2008 Samuel Bronson * Use $(HC) --interactive instead of ghci, in case a non-default version of GHC is in use -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 7877 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080411/dd9aaf7c/attachment.bin From naesten at gmail.com Fri Apr 11 19:20:33 2008 From: naesten at gmail.com (Samuel Bronson) Date: Fri Apr 11 19:16:05 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code Message-ID: Fri Apr 11 19:01:59 EDT 2008 Samuel Bronson * Improve lambda/pi printing code Only only put as many lambda/pi abstractions on a line as will fit. Use sortKindLike to get the uppercase lambdas for type variables of kind other than *. (For instance, * -> *.) I had to make E.TypeCheck, E.Eval and E.Subst import E.Show with a {-# SOURCE #-} pragma before I could use sortKindLike. (I also had to write the .hs-boot file, of course). -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 12131 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080411/4c586ec8/attachment.bin From naesten at gmail.com Sat Apr 12 14:14:47 2008 From: naesten at gmail.com (Samuel Bronson) Date: Sat Apr 12 14:10:11 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code In-Reply-To: References: Message-ID: On 4/11/08, Samuel Bronson wrote: > Fri Apr 11 19:01:59 EDT 2008 Samuel Bronson > * Improve lambda/pi printing code > Only only put as many lambda/pi abstractions on a line as will fit. > Use sortKindLike to get the uppercase lambdas for type variables of > kind other than *. (For instance, * -> *.) Hmm. I accidentally disabled kind-printing in doing this... so don't apply this. From john at repetae.net Sat Apr 12 20:53:33 2008 From: john at repetae.net (John Meacham) Date: Sat Apr 12 20:48:55 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code In-Reply-To: References: Message-ID: <20080413005333.GG25982@sliver.repetae.net> On Fri, Apr 11, 2008 at 07:20:33PM -0400, Samuel Bronson wrote: > Fri Apr 11 19:01:59 EDT 2008 Samuel Bronson > * Improve lambda/pi printing code > Only only put as many lambda/pi abstractions on a line as will fit. > Use sortKindLike to get the uppercase lambdas for type variables of > kind other than *. (For instance, * -> *.) the uppercase lambda is not valid for kinds other than *, actually, the uppercase lambda is just there to help people used to system F, there is no such thing in jhc core since PTS's 'pi' terms subsume explicit type application terms as well as functional types. papers on PTS's (pure type systems) such as the henk paper have more details on the difference between PTS's and system F. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Sat Apr 12 22:19:47 2008 From: naesten at gmail.com (Samuel Bronson) Date: Sat Apr 12 22:15:09 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code In-Reply-To: <20080413005333.GG25982@sliver.repetae.net> References: <20080413005333.GG25982@sliver.repetae.net> Message-ID: On Sat, Apr 12, 2008 at 8:53 PM, John Meacham wrote: > On Fri, Apr 11, 2008 at 07:20:33PM -0400, Samuel Bronson wrote: > > Fri Apr 11 19:01:59 EDT 2008 Samuel Bronson > > * Improve lambda/pi printing code > > Only only put as many lambda/pi abstractions on a line as will fit. > > Use sortKindLike to get the uppercase lambdas for type variables of > > kind other than *. (For instance, * -> *.) > > the uppercase lambda is not valid for kinds other than *, actually, the > uppercase lambda is just there to help people used to system F, there is > no such thing in jhc core since PTS's 'pi' terms subsume explicit type > application terms as well as functional types. papers on PTS's (pure > type systems) such as the henk paper have more details on the difference > between PTS's and system F. Eh? How can it be invalid when uppercase lambda is just sugar for lowercase lambda? Henk paper didn't say anything about that... From john at repetae.net Sat Apr 12 23:00:34 2008 From: john at repetae.net (John Meacham) Date: Sat Apr 12 22:55:55 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code In-Reply-To: References: <20080413005333.GG25982@sliver.repetae.net> Message-ID: <20080413030034.GH25982@sliver.repetae.net> On Sat, Apr 12, 2008 at 10:19:47PM -0400, Samuel Bronson wrote: > > the uppercase lambda is not valid for kinds other than *, actually, the > > uppercase lambda is just there to help people used to system F, there is > > no such thing in jhc core since PTS's 'pi' terms subsume explicit type > > application terms as well as functional types. papers on PTS's (pure > > type systems) such as the henk paper have more details on the difference > > between PTS's and system F. > > Eh? How can it be invalid when uppercase lambda is just sugar for > lowercase lambda? Henk paper didn't say anything about that... Well, when I originally made the shorthand, I just wanted terms in System F2 to look the same in jhc core. Since F2 doesn't have functions at the type level, I made only * turn into a capital lambda. But I suppose it does make sense to expand the shorthand to higher ordered types since jhc core is more powerful than system F2... so yeah, expanding capital lambda to all kind terms does make sense. cool. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Sun Apr 13 12:38:27 2008 From: naesten at gmail.com (Samuel Bronson) Date: Sun Apr 13 12:33:48 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code In-Reply-To: <20080413030034.GH25982@sliver.repetae.net> References: <20080413005333.GG25982@sliver.repetae.net> <20080413030034.GH25982@sliver.repetae.net> Message-ID: On Sat, Apr 12, 2008 at 11:00 PM, John Meacham wrote: > On Sat, Apr 12, 2008 at 10:19:47PM -0400, Samuel Bronson wrote: > > > the uppercase lambda is not valid for kinds other than *, actually, the > > > uppercase lambda is just there to help people used to system F, there is > > > no such thing in jhc core since PTS's 'pi' terms subsume explicit type > > > application terms as well as functional types. papers on PTS's (pure > > > type systems) such as the henk paper have more details on the difference > > > between PTS's and system F. > > > > Eh? How can it be invalid when uppercase lambda is just sugar for > > lowercase lambda? Henk paper didn't say anything about that... > > Well, when I originally made the shorthand, I just wanted terms in > System F2 to look the same in jhc core. Since F2 doesn't have functions > at the type level, I made only * turn into a capital lambda. > > But I suppose it does make sense to expand the shorthand to higher > ordered types since jhc core is more powerful than system F2... so yeah, > expanding capital lambda to all kind terms does make sense. cool. But as I said, I accidentally suppressed the kind printing at the same time... From naesten at gmail.com Sun Apr 13 20:47:56 2008 From: naesten at gmail.com (Samuel Bronson) Date: Sun Apr 13 20:43:31 2008 Subject: [jhc] darcs patch: Show both types when printing rules Message-ID: Sun Apr 13 20:45:01 EDT 2008 Samuel Bronson * Show both types when printing rules This turns up trouble with the Kleisli instance RULES. I guess if you copy and paste that stuff from Control.Arrow, or comment out the (->) stuff, you should see what I mean... anyway, the LHS type doesn't seem to match the RHS type. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 8749 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080413/4e97e07d/attachment.bin From naesten at gmail.com Mon Apr 14 12:40:54 2008 From: naesten at gmail.com (Samuel Bronson) Date: Mon Apr 14 12:36:13 2008 Subject: [jhc] darcs patch: Check that lhs and rhs have compatible t... (and 1 more) Message-ID: Mon Apr 14 12:31:02 EDT 2008 Samuel Bronson * Check that lhs and rhs have compatible types in E.Rules.makeRule Also remove the unused DataTable argument to typesCompatable, since makeRule hasn't got a DataTable. Mon Apr 14 12:38:30 EDT 2008 Samuel Bronson * Demonstrate the Kleisli problem in distilled form -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 10519 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080414/4b3b582e/attachment-0001.bin From naesten at gmail.com Mon Apr 14 18:08:49 2008 From: naesten at gmail.com (Samuel Bronson) Date: Mon Apr 14 18:04:05 2008 Subject: [jhc] createInstanceRules Message-ID: Could you please take a look at the type error that happens when you apply both of the patches from "[jhc] darcs patch: Check that lhs and rhs have compatible t... (and 1 more)" and then run "perl regress/regress.prl Kleisli"? (Maybe you'll want to unpull the first one and instead just print out both types in printRule...) I'm suspicious of the line: Just (n,_) -> foldl EAp (EVar n) (vs ++ map EVar args) but not sure what it should be... From naesten at gmail.com Mon Apr 14 20:08:51 2008 From: naesten at gmail.com (Samuel Bronson) Date: Mon Apr 14 20:04:07 2008 Subject: [jhc] Re: createInstanceRules In-Reply-To: References: Message-ID: Pencil-and-paper summary, below copied-and-pasted code: http://picasaweb.google.com/lh/photo/2qJOWq0cFKrl0FsnDRfLhA From naesten at gmail.com Wed Apr 16 19:27:56 2008 From: naesten at gmail.com (Samuel Bronson) Date: Wed Apr 16 19:23:07 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code Message-ID: Fri Apr 11 19:01:59 EDT 2008 Samuel Bronson * Improve lambda/pi printing code Only only put as many lambda/pi abstractions on a line as will fit. Use sortKindLike to get the uppercase lambdas for type variables of kind other than *. (For instance, * -> *.) I had to make E.TypeCheck, E.Eval and E.Subst import E.Show with a {-# SOURCE #-} pragma before I could use sortKindLike. (I also had to write the .hs-boot file, of course). -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 12383 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080416/3c082a4e/attachment.bin From naesten at gmail.com Wed Apr 16 19:28:48 2008 From: naesten at gmail.com (Samuel Bronson) Date: Wed Apr 16 19:23:58 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code (and 1 more) Message-ID: Fri Apr 11 19:01:59 EDT 2008 Samuel Bronson * Improve lambda/pi printing code Only only put as many lambda/pi abstractions on a line as will fit. Use sortKindLike to get the uppercase lambdas for type variables of kind other than *. (For instance, * -> *.) I had to make E.TypeCheck, E.Eval and E.Subst import E.Show with a {-# SOURCE #-} pragma before I could use sortKindLike. (I also had to write the .hs-boot file, of course). Wed Apr 16 19:26:09 EDT 2008 Samuel Bronson * Fix kind printing for Lambdas/foralls -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 13557 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080416/e16e0783/attachment-0001.bin From naesten at gmail.com Wed Apr 16 19:29:45 2008 From: naesten at gmail.com (Samuel Bronson) Date: Wed Apr 16 19:24:54 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code In-Reply-To: References: Message-ID: Oops... I meant to send this and the kind-printing fix... From john at repetae.net Wed Apr 16 20:42:32 2008 From: john at repetae.net (John Meacham) Date: Wed Apr 16 20:37:47 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code (and 1 more) In-Reply-To: References: Message-ID: <20080417004231.GP25982@sliver.repetae.net> Thanks. patches applied. sorry I have not had a chance to look at the type error you were getting with rules yet, I'll try to take a look later tonight. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Thu Apr 17 16:10:08 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 17 16:05:20 2008 Subject: [jhc] darcs patch: Improve lambda/pi printing code (and 1 more) In-Reply-To: <20080417004231.GP25982@sliver.repetae.net> References: <20080417004231.GP25982@sliver.repetae.net> Message-ID: On 4/16/08, John Meacham wrote: > Thanks. patches applied. sorry I have not had a chance to look at the > type error you were getting with rules yet, I'll try to take a look > later tonight. Well, did you do it? From naesten at gmail.com Thu Apr 17 23:22:04 2008 From: naesten at gmail.com (Samuel Bronson) Date: Thu Apr 17 23:17:10 2008 Subject: [jhc] =?windows-1252?q?Announcing_DarcsWatch_-_nomeata=92s_mind_?= =?windows-1252?q?shares?= Message-ID: Announcing DarcsWatch - nomeata's mind shares: http://www.joachim-breitner.de/blog/archives/290-guid.html John, could this be the software you've been looking for? From john at repetae.net Fri Apr 18 00:50:25 2008 From: john at repetae.net (John Meacham) Date: Fri Apr 18 00:45:30 2008 Subject: [jhc] Announcing =?utf-8?Q?DarcsWatch_?= =?utf-8?Q?-_nomeata=E2=80=99s?= mind shares In-Reply-To: References: Message-ID: <20080418045025.GD17560@sliver.repetae.net> On Thu, Apr 17, 2008 at 10:22:04PM -0500, Samuel Bronson wrote: > Announcing DarcsWatch - nomeata's mind shares: > http://www.joachim-breitner.de/blog/archives/290-guid.html > > John, could this be the software you've been looking for? Ah, yes! that looks nice. John -- John Meacham - ?repetae.net?john? From gwern0 at gmail.com Sun Apr 20 21:43:54 2008 From: gwern0 at gmail.com (gwern0@gmail.com) Date: Sun Apr 20 21:39:31 2008 Subject: [jhc] darcs patch: Makefile.am: swap order of args to sed Message-ID: <480bf180.0589300a.5ccb.ffffe37e@mx.google.com> Sun Apr 20 21:42:52 EDT 2008 gwern0@gmail.com * Makefile.am: swap order of args to sed This is necessary for building on my system; otherwise sed always fails with an error. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/x-darcs-patch Size: 4416 bytes Desc: A darcs patch for your repository! Url : http://www.haskell.org/pipermail/jhc/attachments/20080420/cbffe400/attachment.bin From john at repetae.net Sun Apr 20 23:11:14 2008 From: john at repetae.net (John Meacham) Date: Sun Apr 20 23:06:12 2008 Subject: [jhc] darcs patch: Makefile.am: swap order of args to sed In-Reply-To: <480bf180.0589300a.5ccb.ffffe37e@mx.google.com> References: <480bf180.0589300a.5ccb.ffffe37e@mx.google.com> Message-ID: <20080421031114.GQ17560@sliver.repetae.net> Thanks! patch applied. What sort of system do you have by the way? I am guessing something BSD based? I know jhc worked on OpenBSD for a while, but I have not tested it much. John -- John Meacham - ?repetae.net?john? From gwern0 at gmail.com Sun Apr 20 23:28:17 2008 From: gwern0 at gmail.com (Gwern Branwen) Date: Sun Apr 20 23:24:24 2008 Subject: [jhc] darcs patch: Makefile.am: swap order of args to sed In-Reply-To: <20080421031114.GQ17560@sliver.repetae.net> References: <480bf180.0589300a.5ccb.ffffe37e@mx.google.com> <20080421031114.GQ17560@sliver.repetae.net> Message-ID: <20080421032817.GA20977@localhost> On 2008.04.20 20:11:14 -0700, John Meacham scribbled 0.3K characters: > Thanks! patch applied. > > What sort of system do you have by the way? I am guessing something BSD > based? I know jhc worked on OpenBSD for a while, but I have not tested > it much. > > John No, not BSD - ~amd64 Gentoo Linux. Does no one else try JHC on Gentoo? -- gwern NSOF to Patel GOSIP electronic 127 DI Ingram Gamma Keyhole -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://www.haskell.org/pipermail/jhc/attachments/20080420/fe4d998e/attachment.bin From john at repetae.net Mon Apr 21 00:21:44 2008 From: john at repetae.net (John Meacham) Date: Mon Apr 21 00:16:40 2008 Subject: [jhc] darcs patch: Makefile.am: swap order of args to sed In-Reply-To: <20080421032817.GA20977@localhost> References: <480bf180.0589300a.5ccb.ffffe37e@mx.google.com> <20080421031114.GQ17560@sliver.repetae.net> <20080421032817.GA20977@localhost> Message-ID: <20080421042144.GR17560@sliver.repetae.net> On Sun, Apr 20, 2008 at 11:28:17PM -0400, Gwern Branwen wrote: > On 2008.04.20 20:11:14 -0700, John Meacham scribbled 0.3K characters: > > What sort of system do you have by the way? I am guessing something BSD > > based? I know jhc worked on OpenBSD for a while, but I have not tested > > it much. > No, not BSD - ~amd64 Gentoo Linux. Does no one else try JHC on Gentoo? Ah, as far as I knew all linuxess used the same sed implementation. Interesting. I have been meaning to try out Gentoo sometime. John -- John Meacham - ?repetae.net?john? From naesten at gmail.com Tue Apr 22 16:04:57 2008 From: naesten at gmail.com (Samuel Bronson) Date: Tue Apr 22 16:07:04 2008 Subject: [jhc] Still waiting to hear back about that type error Message-ID: John, have you looked at it yet? Are you at least working on getting Jhc to use the normal base package? From john at repetae.net Wed Apr 23 00:10:16 2008 From: john at repetae.net (John Meacham) Date: Wed Apr 23 00:05:06 2008 Subject: [jhc] Still waiting to hear back about that type error In-Reply-To: References: Message-ID: <20080423041016.GW17560@sliver.repetae.net> On Tue, Apr 22, 2008 at 04:04:57PM -0400, Samuel Bronson wrote: > John, have you looked at it yet? Are you at least working on getting > Jhc to use the normal base package? Sorry, I have been swamped with my day job the last week or so. I will hopefully have time this weekend to get back to jhc. I should say I am not trying to get jhc to use the ghc base package directly, meerly have the build automatically pick out the specific files that would have been imported to jhc unchanged. (like Arrow,Applicative, etc...) Basically it just involves setting up the '-i' paths properly and attempting to compile base, but there are a few things jhc still trips over I am fixing up, I think your recent patches solved some of them, but I will have to look. I think I need a '-fno-ghc-rules' option that can be applied to specific files which will make jhc accept RULES_JHC but not RULES, since ghc rules don't always make sense applied to jhc, and the phase numbering stuff is very compiler dependent (and largely unneeded due to CATALYSTs) John -- John Meacham - ?repetae.net?john?