The page you link, where you've likely posted "a month ago" says:
If you do not hear back from the maintainers after posting a message below, please follow up by using the link on the left side of this page or follow this link to contact maintainers. If you still hear nothing back, please follow the package triage process.
So... what about doing so?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I don't really know anything about Chocolatey, so I was not aware they had linked to my GC 3.1.2 build from 2 years ago. That particular build has been removed, and the replacement is available at this link: https://www.arnoldtrembley.com/GC312-BDB-SP2-rename-7z-to-exe.7z
That's assuming they want the current most stable release, which is GC 3.1.2. The "SP2" version has several upgrades to included components.
I do have a GC 3.2 rc2, but that will be replaced (hopefully) in the next few weeks with a stable GC 3.2 final release.
I don't know how Chocolatey is updated, but they would probably need to calculate a new MD5 checksum for the replacement build, since I have never calculated an MD5 checksum.
I've never heard from the Chocolatey maintainer, but if I would be happy to work with them to make sure they always have a good link.
Thanks for "unbreaking" the old external package!
Just to mention - if you want to have nice checksum support with nearly as much archive support as possible on Win32 then you likely would want to checkout a fork of 7z: https://github.com/mcmilk/7-Zip-zstd (7z provides SHA checksums only and does not support zstd or lzip archives, 7-Zip-std adds those while integrating upstream changes).
And just a note: because of recent test results and changes in several "core" COBOL parts I'm going to create 3.2RC3 in the upcoming week...
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Goes into the next update for THANKS file, before or 3.2RC3, which is (over)due for this week.
Current draft:
Past and present maintainers:
Keisuke Nishida <knishida@netlab.jp>
Roger While (1950-2015)
Simon Sobisch <simonsobisch@gnu.org>
Other core developers with major code and design contributions
Ron Norman <rjn@inglenet.com> including but not limited to
- Implementation and ongoing work on the C-/D-/VBISAM handler
- Report Writer development
- Standard EXTFH together with ADDRESS OF FH--FCD and FH--KEYDEF
- Improved compatibility with Micro Focus COBOL
- Runtime configuration
- C-API to inspect/update COBOL data fields
- Dump on abort
- Performance improvements
Edward Hart <edward.dan.hart@gmail.com> including but not limited to
- fine-grained configuration of run-time checks
- Compiler configuration, in particular reserved word handling
- datetime intrinsic functions
- XML GENERATE and JSON GENERATE
- Initial support for internal repository (function and program prototypes)
- Syntax support for standard COBOL and existing implementor extensions
- Bug fixes and test suite enhancements
We are grateful for the work of the Translation Project, its translation teams
and their individual contributors, who provide translations of user messages
for non-English speakers.
Similar thanks go to all the packagers who allow users to install GnuCOBOL from
their distributions' package management systems.
Also to the many people that have helped with testing this software.
We hope that everyone will continue to provide feedback. This is invaluable
to the ongoing development process.
A special mention here goes to people who have given exceptional support in
terms of time and resources on hardware that was not available to the
developers:
Oleg Philon - For his work on the PowerPC
David Wilson - For his work on the MAC (Darwin)
Sergey Kashyrin - For his work on:
- SUN Solaris
- IBM AIX
- HP-UX
And continuing this amazing support:
Arnold Trembley <arnold.trembley@att.net> - Win32 packaging and user support
Chuck Haatvedt <chuck.haatvedt@gmail.com> - Work on optimized BCD handling
Ludwin Janvier <lud.janvier@gmail.com> - patches for build/packaging issues
Hans-Martin Rasch - Work on compiler syntax
Michel Gouget - Work on syntax - IS
Bill Klein - The mainstay for COBOL questions
Frank Swarbrick - Work on compiler syntax
Warren Gay - For testing systems we didn't even know we supported!
(e.g. old DEC Alpha systems)
Fabrizio Calabretta - Work on the internal EXTFH interface
OCamlPro SAS for assigning David Declerck, Nicolas Berthier and
Fabrice Le Fessant to work for adding GCOS (Bull) support and
contributing the result.
Denis Hugonnard-Roche for continuous additions to the internal test suite to
improve code coverage.
Further explicit thanks to
Alain Lucari <eurlix@libertysurf.fr>
Brian Tiffin <btiffin@gnu.org>
David Korn <dgk@research.att.com>
Dave Pitts <dpitts@cozx.com>
Joe Robbins (-2017)
Keiichi Takahashi <bitwalk@jcom.home.ne.jp>
Peg <peg@coboler.com>
Richard Smith <rich@theforest.plus.com>
Thomas Biehler <tobiglob@users.sourceforge.net>
William M. Klein <wmklein@ix.netcom.com>
Yoshiki Kusumoto <webmaster@coboler.com>
-------------------------------------------------------------------------------
For OpenCOBOL 0.10:
Bug reports and patches by:
Ingo D. Rullhusen
Richard A. Painter
Tim Josling
OpenCOBOL was inspired by the TinyCOBOL project.
Thanks to the TinyCOBOL authors and maintainers.
Additional from TinCOBOL:
Rildo Pragana (1958-2020)
Andrew Cameron
Bernard Giroud
David Essex
Jeff Smith
Jim Noeth
Stephen Connolly
Laura Tweedy
Feedback welcomed.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think a citation should be added to Gary and Vincent's documentation contribution to their remarkable Programmer's Guide which is an integral and valuable part of the project.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Those are people I'm thankful for, but are already "quite away" from the GnuCOBOL code we do distribute with the tarball.
There are also others who are active in the discussion board but that's not what this file is about, I think.
... concerning GCSORT - maybe it would be good to check for possible inclusion within the GnuCOBOL source tree directly as gcsort/cobsort utility? @smenna What is your take on this (we'd need a copyright assignment to the FSF)?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Simon,
it is an honor for me to receive this opportunity and I am very pleased and proud.
I do not have a lot of expertise for this licensing step so I am asking for help on how I can proceed.
I have a doubt about licensing since GCSort consists of a baseline originally created by Cedric ISSALY (OCSort module then renamed) in GCSort,
in addition I used, for memory mapped temporary file management, the mmfioc.c file adapting it to the needs of GCSORT which was originally created by Geoff McLane.
In this case, is the transition still possible ?
How can this be done ?
Are there any costs to be incurred ?
Thank you.
Sauro
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Just as a heads up - as per https://community.chocolatey.org/packages/gnucobol#comment-6099134453 the packaged Windows build at https://community.chocolatey.org/packages/gnucobol is broken. What seems to have happened is that new 3.2 builds have been posted at https://arnoldtrembley.com/GnuCOBOL.htm and the old 3.1.2 ones removed.
It looks like the author of https://community.chocolatey.org/packages/gnucobol and https://arnoldtrembley.com/GnuCOBOL.htm might be different people (and of course I've no idea how "authorised" the build there is) - I'm just mentioning it here so that people are aware.
The page you link, where you've likely posted "a month ago" says:
So... what about doing so?
Done - we'll see what happens...
Well, that was a surprise!
I don't really know anything about Chocolatey, so I was not aware they had linked to my GC 3.1.2 build from 2 years ago. That particular build has been removed, and the replacement is available at this link:
https://www.arnoldtrembley.com/GC312-BDB-SP2-rename-7z-to-exe.7z
That's assuming they want the current most stable release, which is GC 3.1.2. The "SP2" version has several upgrades to included components.
I do have a GC 3.2 rc2, but that will be replaced (hopefully) in the next few weeks with a stable GC 3.2 final release.
I don't know how Chocolatey is updated, but they would probably need to calculate a new MD5 checksum for the replacement build, since I have never calculated an MD5 checksum.
I've never heard from the Chocolatey maintainer, but if I would be happy to work with them to make sure they always have a good link.
EDIT - I have just not gone back and restored that older version to my website, so the old link should work once again:
https://arnoldtrembley.com/GC312-BDB-rename-7z-to-exe.7z
I know it's the same file, so the MD5 checksum should still be the same. But you never know.
Kind regards,
Last edit: Arnold Trembley 2023-03-04
EDIT - I have just not gone back and restored that older version to my website, so the old link should work once again:
https://arnoldtrembley.com/GC312-BDB-rename-7z-to-exe.7z
I know it's the same file, so the MD5 checksum should still be the same. But you never know.
It turns out calculating an MD5 checksum is quite easy. Windows versions (32-bit or 64-bit) can be downloaded from:
https://sourceforge.net/projects/filechecksumutility/
And the checksum matches, so Chocolatey should work once again.
Thanks for "unbreaking" the old external package!
Just to mention - if you want to have nice checksum support with nearly as much archive support as possible on Win32 then you likely would want to checkout a fork of 7z: https://github.com/mcmilk/7-Zip-zstd (7z provides SHA checksums only and does not support zstd or lzip archives, 7-Zip-std adds those while integrating upstream changes).
And just a note: because of recent test results and changes in several "core" COBOL parts I'm going to create 3.2RC3 in the upcoming week...
Thanks, Arnold.
Simon, Ron, me, et al, why is Arnold's name not anywhere in SVN? :-)
Joking, not joking.
Arnold, would you mind if we put your name in some of the front matter source file commentary?
Seriously. A credit deserved, I think.
Have good,
Blue
Goes into the next update for THANKS file, before or 3.2RC3, which is (over)due for this week.
Current draft:
Feedback welcomed.
Typo in TinCobol near the bottom for Rildo. Though that wouldn't be a bad small tin COBOL brandname. ;-)
Cheers,
Blue
I think a citation should be added to Gary and Vincent's documentation contribution to their remarkable Programmer's Guide which is an integral and valuable part of the project.
Totally agreed, added as:
Last edit: Simon Sobisch 2023-03-14
Blue for the FAQ !!!
He's already in, if he wants to he can add details (would be FAQ and possibly cobcrun)
Yeah, I was pondering that. Being in the further explicit thanks list seems underwhelming. :-)
Might do something, probably won't.
Cheers,
Blue
László Erdős, contributions
Sauro Menna, GCSORT
Marco Ridoni, Gix
Colin Duquesnoy. OpenCOBOLIDE
?
Not sure how many layers deep this would want to go.
Have good,
Blue
Those are people I'm thankful for, but are already "quite away" from the GnuCOBOL code we do distribute with the tarball.
There are also others who are active in the discussion board but that's not what this file is about, I think.
... concerning GCSORT - maybe it would be good to check for possible inclusion within the GnuCOBOL source tree directly as gcsort/cobsort utility?
@smenna What is your take on this (we'd need a copyright assignment to the FSF)?
Hi Simon,
it is an honor for me to receive this opportunity and I am very pleased and proud.
I do not have a lot of expertise for this licensing step so I am asking for help on how I can proceed.
I have a doubt about licensing since GCSort consists of a baseline originally created by Cedric ISSALY (OCSort module then renamed) in GCSort,
in addition I used, for memory mapped temporary file management, the mmfioc.c file adapting it to the needs of GCSORT which was originally created by Geoff McLane.
In this case, is the transition still possible ?
How can this be done ?
Are there any costs to be incurred ?
Thank you.
Sauro
No costs (other than the copyright clerk but that is paid by the FSF).
But you're right that we'd need both yours and Cedric's assignment.
I guess we may could get mmfioc.c out - otherwise we'd need a third assignment.
I registered with Chocolatey, and emailed the maintainer of the Chocolatey GnuCOBOL install, who sent a very nice reply.
The Chocolatey GnuCOBOL install script has now been updated to link to the updated GC 3.1.2 build, which has several improvements:
Kind regards,
Last edit: Arnold Trembley 2023-03-05
Excellent, thanks!