Re: [Chicken-users] Difficulty installing eggs on Chicken 5 on Windows 7 MSYS2
Hi guys, This is caused by a bug in CHICKEN 5. I reported a duplacate of it here: https://bugs.call-cc.org/ticket/1584 It is caused by MAX_PATH being too small and chicken-do not able to store all of its arguments. You can apply commit *1e006b65* to your 5.0.0 sources or simply change line 79 in chicken-do.c to read "static TCHAR cmdline[ 1 ];" instead of MAX_PATH. That solved things for me. K. On Fri, Mar 1, 2019, 08:00 John Croisant On 2/28/19 6:50 AM, dignified face wrote: > > Hello, > > I installed Chicken 5.0.0 on Windows 7 64bit with MSYS2, following the > instructions found here: http://wiki.call-cc.org/msys2 > All works fine, but when I try to install an egg - for instance > "chicken-install matchable" - I get the following error: > > building matchable >C:/msys64/usr/local/bin/csc -host -D compiling-extension -J -s -setup-mode > -I C:\msys64\home\username\.chicken-install\cache\matchable -C > -IC:\msys64\home\username\.chicken-install\cache\matchable -O2 -d1 > matchable.scm -o > C:\msys64\home\username\.chicken-install\cache\matchable\matchable.so > creating subprocess failed > > Error: shell command terminated with nonzero exit code > 1 > "C:\\msys64\\home\\username\\.chicken-install\\cache\\matchable\\matchable.build.bat" > > chicken-do seems to be the problem. I managed to build the egg by copying the > csc invocations from matchable.build.bat then running matchable.install.bat > manually, but this seems like a bad approach. Any help would be greatly > appreciated! > > Thanks in advance, > JJ > > Unfortunately I don't have a solution, but I have the same problem on > Windows 10 MSYS2. Here is a paste with a bunch of information in case it > helps someone debug the problem: > > > http://paste.call-cc.org/paste?id=d8878c03d3dbdc1e49f642f3255c893efa18db95 > > One thing I didn't mention in the paste is that I did a "make check" after > compiling and installing CHICKEN, and it seemed to pass. I didn't paste the > output of make check because it is extremely long. > > - John Croisant > ___ > Chicken-users mailing list > Chicken-users@nongnu.org > https://lists.nongnu.org/mailman/listinfo/chicken-users > ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
[Chicken-users] r6rs-bytevectors for chicken 5
With his permission, I've turned John Cowan's r6rs-bytevectors implementation: https://github.com/scheme-requests-for-implementation/srfi-4/tree/master/contrib/cowan into a Chicken 5 egg here: https://github.com/dieggsy/r6rs-bytevectors I'm aware of the separate C4 implementation of r6rs-bytevectors. I originally tried to port that implementation to C5, but I had some trouble with that and I think John's implementation is a little more favorable for a couple reasons: - It's a little more recent and pretty much already written in C5 syntax - my changes were fairly minimal and trivial. - It implements a couple flonum operations in C - It has fairly extensive tests, which I've converted to use the test egg. It'd be nice to have this in the egg repo for Chicken 5, assuming I haven't made any glaring mistakes. - Diego A. Mundo ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
Re: [Chicken-users] [ANN] ck-macros 0.2.0
On Thu, 28 Feb 2019 22:45:42 -0600 John Croisant wrote: > The ck-macros egg version 0.2.0 has been released. This version adds > CHICKEN 5 support, many new features, some R5RS compatibility fixes, and > a few breaking changes to the API. > > https://wiki.call-cc.org/eggref/5/ck-macros > > ck-macros is based on work by Oleg Kiselyov, with many original > contributions by me. It provides a novel kind of macro (or, a novel way > of using macros) with two important properties: > > 1. Aside from wrappers, they are implemented using only standard R5RS > features, mostly syntax-rules. So they are trivially portable to pretty > much any Scheme, not just CHICKEN. And wrappers are trivially portable > to any Scheme that implements er-macro-transformer in the usual way. > > 2. They can be composed (combined) much like Scheme procedures can. It > is even possible to create higher-order macros, analogous to > higher-order functions. For example, you can use "c-compose" to compose > multiple CK-macros together, then use "c-map1" to map the composed macro > over a list. There are many, many examples on the wiki page. > > The biggest new feature in version 0.2.0 is the "ck-wrapper" procedure, > which creates a CK-macro from any Scheme procedure, including lambda > forms and "let over lambda" closures. The egg provides wrappers for many > standard R5RS procedures, and it's trivial to create your own. > > If this sounds useful to you, wonderful. Personally, I just enjoy > exploring the boundaries of what is possible with this idea -- already > much more than I expected! > > - John Croisant > > P.S. Please add this to the CHICKEN 5 coop: > > > https://gitlab.com/jcroisant/ck-macros/raw/master/ck-macros.chicken-5.release-info Thanks a lot, John. Your egg has been added to the coop. All the best. Mario -- http://parenteses.org/mario ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
[Chicken-users] [ANN] ck-macros 0.2.0
The ck-macros egg version 0.2.0 has been released. This version adds CHICKEN 5 support, many new features, some R5RS compatibility fixes, and a few breaking changes to the API. https://wiki.call-cc.org/eggref/5/ck-macros ck-macros is based on work by Oleg Kiselyov, with many original contributions by me. It provides a novel kind of macro (or, a novel way of using macros) with two important properties: 1. Aside from wrappers, they are implemented using only standard R5RS features, mostly syntax-rules. So they are trivially portable to pretty much any Scheme, not just CHICKEN. And wrappers are trivially portable to any Scheme that implements er-macro-transformer in the usual way. 2. They can be composed (combined) much like Scheme procedures can. It is even possible to create higher-order macros, analogous to higher-order functions. For example, you can use "c-compose" to compose multiple CK-macros together, then use "c-map1" to map the composed macro over a list. There are many, many examples on the wiki page. The biggest new feature in version 0.2.0 is the "ck-wrapper" procedure, which creates a CK-macro from any Scheme procedure, including lambda forms and "let over lambda" closures. The egg provides wrappers for many standard R5RS procedures, and it's trivial to create your own. If this sounds useful to you, wonderful. Personally, I just enjoy exploring the boundaries of what is possible with this idea -- already much more than I expected! - John Croisant P.S. Please add this to the CHICKEN 5 coop: https://gitlab.com/jcroisant/ck-macros/raw/master/ck-macros.chicken-5.release-info ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
Re: [Chicken-users] Difficulty installing eggs on Chicken 5 on Windows 7 MSYS2
On 2/28/19 6:50 AM, dignified face wrote: > Hello, > > I installed Chicken 5.0.0 on Windows 7 64bit with MSYS2, following the > instructions found here: http://wiki.call-cc.org/msys2 > All works fine, but when I try to install an egg - for instance > "chicken-install matchable" - I get the following error: > > building matchable >C:/msys64/usr/local/bin/csc -host -D compiling-extension -J -s -setup-mode > -I C:\msys64\home\username\.chicken-install\cache\matchable -C > -IC:\msys64\home\username\.chicken-install\cache\matchable -O2 -d1 > matchable.scm -o > C:\msys64\home\username\.chicken-install\cache\matchable\matchable.so > creating subprocess failed > > Error: shell command terminated with nonzero exit code > 1 > "C:\\msys64\\home\\username\\.chicken-install\\cache\\matchable\\matchable.build.bat" > > chicken-do seems to be the problem. I managed to build the egg by copying the > csc invocations from matchable.build.bat then running matchable.install.bat > manually, but this seems like a bad approach. Any help would be greatly > appreciated! > > Thanks in advance, > JJ Unfortunately I don't have a solution, but I have the same problem on Windows 10 MSYS2. Here is a paste with a bunch of information in case it helps someone debug the problem: http://paste.call-cc.org/paste?id=d8878c03d3dbdc1e49f642f3255c893efa18db95 One thing I didn't mention in the paste is that I did a "make check" after compiling and installing CHICKEN, and it seemed to pass. I didn't paste the output of make check because it is extremely long. - John Croisant ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
Re: [Chicken-users] SRFI-123 ported to Chicken 5
Hi Diego, On Tue, 26 Feb 2019 19:46:11 + Diego wrote: > I've ported SRFI-123 to C5: https://github.com/dieggsy/srfi-123 > > I also made some minor fixes to lingering issues I hadn't yet cleaned up. Thanks. I've added it to the coop. A minor issue: srfi-4 is part of the CHICKEN core, not an egg. Thus, it should not be in `dependencies'. chicken-install doesn't complain about that, but it is a good practice to keep only eggs in the list of dependencies. > A couple questions: > > - Is there a way to make a wiki entry for the egg in the egg repo > itself? I'd rather version control something like that rather than > edit with the web interface. That's not possible (at least not automatically). You can always do it manually though, but then you have to be careful not to clobber changes made to the wiki. Also note that wiki pages are stored in svn, so there is some sort of version control behind the scenes. You don't even have to use the web interface. You can edit files in your favorite text editor and use a svn client to commit your changes. > - How do I go about aliasing srfi.123 to srfi-123? I don't know how to do that. All the best. Mario -- http://parenteses.org/mario ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
Re: [Chicken-users] Chicken 5 support for generalized-arrays egg
Hi Jeremy, On Wed, 27 Feb 2019 22:45:03 -0700 Jeremy Steward wrote: > Spent some time to update and add support for my generalized-arrays egg > today. Find the release info file at: > > https://bitbucket.org/ThatGeoGuy/chicken-generalized-arrays/raw/master/generalized-arrays.release-info Thanks a lot. I've added it to the CHICKEN 5 coop. > Since there isn't yet a salmonella for C5 I haven't been able to test > with that, but this repo should be working if my tests today are any > indication. Apologies for that. I really should find some contiguous hours to release salmonella for C5. In the meantime, you can install the C5-specific version of salmonella from git with: $ git clone https://github.com/mario-goulart/salmonella.git $ cd salmonella $ git checkout chicken-5 $ chicken-install All the best. Mario -- http://parenteses.org/mario ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
Re: [Chicken-users] Back from the dead: pstk
Hi Heinz, On Tue, 26 Feb 2019 14:53:45 +0100 ipc...@arcor.de wrote: > As promised, I've ported the pstk egg to Chicken 5, and would like to > take over as maintainer. > > https://github.com/utz82/pstk > https://raw.githubusercontent.com/utz82/pstk/master/pstk.release-info > > License has been reverted to BSD 2-clause, as the main source actually > never was in the public domain. > > I threw out the SVN backlog when importing to git, so maybe the C4 egg > should continue to use the old SVN repository instead. Otherwise pstk > should replace Chicken/Tk as the "maintained" Tk egg at this point. I > do plan on maintaining both 4 and 5 versions, in any case. > > Haven't done any changes to the code for now, other than C5 > compatibility tweaks, and setting default tclsh to tclsh8.6. I'm > amazed that this decade-old code works on C5 pretty much out of the > box. Thanks a lot. I've added pstk to the CHICKEN 5 coop and update the location for the CHICKEN 4 one. All the best. Mario -- http://parenteses.org/mario ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
[Chicken-users] Difficulty installing eggs on Chicken 5 on Windows 7 MSYS2
Hello, I installed Chicken 5.0.0 on Windows 7 64bit with MSYS2, following the instructions found here: http://wiki.call-cc.org/msys2 All works fine, but when I try to install an egg - for instance "chicken-install matchable" - I get the following error: building matchable C:/msys64/usr/local/bin/csc -host -D compiling-extension -J -s -setup-mode -I C:\msys64\home\username\.chicken-install\cache\matchable -C -IC:\msys64\home\username\.chicken-install\cache\matchable -O2 -d1 matchable.scm -o C:\msys64\home\username\.chicken-install\cache\matchable\matchable.so creating subprocess failed Error: shell command terminated with nonzero exit code 1 "C:\\msys64\\home\\username\\.chicken-install\\cache\\matchable\\matchable.build.bat" chicken-do seems to be the problem. I managed to build the egg by copying the csc invocations from matchable.build.bat then running matchable.install.bat manually, but this seems like a bad approach. Any help would be greatly appreciated! Thanks in advance, JJ ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
Re: [Chicken-users] [Q] Overhead of FFI in Chicken scheme
On Mon, Feb 11, 2019 at 03:16:11PM +0900, Sungjin Chun wrote: > Hi, > > Are there any FFI overhead in case of compiling to a binary using "bind"? Hi Sungjin, As far as I know, the bind egg should not introduce any overhead beyond what the core FFI will have. For the core, there's some overhead regarding type and range checking, and for some types of argument like strings, there'll be copying involved. The best way to figure whether out the overhead is acceptable to your program is to simply measure it. There are ways to make things faster if safety is not important. Sorry for the late reply! Cheers, Peter signature.asc Description: PGP signature ___ Chicken-users mailing list Chicken-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/chicken-users
