Created attachment 210776 [details] The diff to the Makefile and the distinfo. Hello. dMagnetic saw the release of version 0.20, so the FreeBSD port needs to be updated as well. It is my understanding that having the sha256 sum encoded as base64 is preferable, so this is being written in the distinfo. Please find the patch to the Makefile and the distinfo attached to this email. Enjoy!! Thomas Dettbarn
Created attachment 210779 [details] dMagnetic 0.20 I change the distinfo. FYI, you can create it with "make makesum" https://www.freebsd.org/doc/en/books/porters-handbook/book.html#porting-checksum While I'm here, I move EXTRACT_SUFX to pet portlint. Do you have a Changelog?
games/dMagnetic: Update to 0.20 * Update to 0.20 * Move EXTRACT_SUFX (pet portlint) Changelog: http://www.dettus.net/dMagnetic/ ? QA: * portlint: OK (looks fine.) * testport: OK (poudriere: 11/12/13, i386/amd64)
Use USES=tar:<suffix> instead of EXTRACT_SUFX: https://www.freebsd.org/doc/en/books/porters-handbook/book.html#uses-tar
(In reply to Loïc Bartoletti from comment #1) Hello Loic. Thank you so very much! I must confess that keeping track of all the ports and the packages is a bit of a hassle, so I mostly generate them in one sitting. Which, unfortunately, means that some operating systems are getting more pampered than others.... Sorry! In a perfect world, I would just develop my code, and would not have to worry about the ports and packages.... To answer your question: No, I do not have a changelog. It would have been good practice to keep one, but I did not. What I do have, however, are those: http://freshcode.club/projects/dmagnetic http://www.dettus.net/dMagnetic/old_releases.html
Created attachment 210800 [details] dMagnetic 0.20 Use USES=tar:<suffix> instead of EXTRACT_SUFX
(In reply to Thomas Dettbarn from comment #4) You're welcome! I only asked you a changelog to match with our template: https://wiki.freebsd.org/KubilayKocak/ThePerfectPortsIssue#General I updated the patch. I need your approval to ask to commit it. Thanks.
(In reply to Loïc Bartoletti from comment #6) Well... actually... could I trouble you for a small favor? The OpenBSD guys were so nice as to add a do-test: target to their ports. Is it possible to include this in the FreeBSD Makefile as well? do-test: cd ${WRKSRC}/testcode; if [ `echo Hello | ../dMagnetic \ -mag minitest.mag -gfx minitest.gfx \ -vmode low_ansi -vecho -vcols 300 -vrows 300 | \ sha256 -b` == 6DhbUg1shZBuSXIm3PNK1/fMfRQ5RIHCuLPik+IkeQM= ]; \ then echo ok; else echo expected output not seen; exit 1; fi Explanation: Given the input, the output should always be the same, meaning that the sha256 sum should always be the same. Essentially, my project is a virtual machine, that is running ancient bytecode. I took the liberty of hand-writing such bytecode myself, and running it is testing several aspects of the machine. When it finishes, it has produces a deterministic output, hence the sha256 sum. A match means, that the machine can run the bytecode on the current hardware. This gives users the confidence that it is worth the effort to purchase the original bytecode for the games (which is, of course, copyrighted material).
(In reply to Loïc Bartoletti from comment #6) Otherwise: Yes, approved! Thank you!!!!
(In reply to Thomas Dettbarn from comment #7) There is no "-b" option on FreeBSD for sha256. I modified the test with the hex value: do-test: cd ${WRKSRC}/testcode; if [ `echo Hello | ../dMagnetic \ -mag minitest.mag -gfx minitest.gfx \ -vmode low_ansi -vecho -vcols 300 -vrows 300 | \ sha256 ` == 86ca2731a4333ed24203c5bda33e71a222e864cb22debe656de52157e89ab1ea ]; \ then echo ok; else echo expected output not seen; exit 1; fi
(In reply to Loïc Bartoletti from comment #9) Wonderful! Thank you!! Err.... Btw... I found a bug, and I was forced to do a new release.... 0.21 is the most recent one.... I. AM. TRULY. SORRY.
Created attachment 210857 [details] dMagnetic 0.21 Update to 0.21
(In reply to Thomas Dettbarn from comment #10) No worries!
(In reply to Loïc Bartoletti from comment #12) Well then! One more thing: -vmode low_ansi MIGHT create a different output on machines, where dMagnetic has been configured. A more robust parameter for the test would be -vmode high_ansi (which, of course, will result in a different sha256 output). Gosh, I HOPE that is the last thing... Can you please apply the patch to the svn? ;)
Created attachment 210901 [details] dMagnetic 0.21 test update
APPROVAL! thank you so so much!
A commit references this bug: Author: lbartoletti Date: Wed Jan 22 07:12:04 UTC 2020 New revision: 523796 URL: https://svnweb.freebsd.org/changeset/ports/523796 Log: games/dMagnetic: Update to 0.21 - Update to 0.21 - Remove EXTRACT_SUFX and use USE=tar:bz2 instead (pet portlint) Changelog: http://freshcode.club/projects/dmagnetic http://www.dettus.net/dMagnetic/old_releases.html PR: 243379 Reported by: Thomas Dettbarn (maintainer) Reviewed by: koobs Approved by: tcberner (mentor) Differential Revision: https://reviews.freebsd.org/D23214 Changes: head/games/dMagnetic/Makefile head/games/dMagnetic/distinfo