git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH 0/2] Getopt::Long workaround in send-email

From
Brandon Casey <drafnel@gmail.com>
Date
Feb 15, 2015, 06:13 UTC
Message-ID
<CA+sFfMetZv_sr9EUOQDJpdq0E9bYYKWrSh4OeFfRDBY3SjRxeg@mail.gmail.com>
In-Reply-To
<1423858769-1565-1-git-send-email-gitster@pobox.com>

[apparently it is impossible to send a plain text email using Google Inbox, maybe people on this list know someone to talk to about that? Sorry for the dup for those on cc]

On Fri, Feb 13, 2015 at 12:19 PM, Junio C Hamano <gitster@pobox.com> wrote:
Show 6 quoted lines
> The first one is a replay of Kyle's workaround for older versions of
> Getopt::Long that did not take "--no-option" to negate a boolean
> option "--option".  The second one revert the workarounds made to
> the test script over time, and should break if the first one does
> not work well for older Getopt::Long (I have no reason to suspect it
> would break, though).

The only downside I can see is that we're going to end up carrying around these extra options for the forseeable future and possibly adding more over time with this precedent. Maybe that's not so bad. The extra options are not ugly at all. My original thinking in just fixing up the tests was that the platforms with ancient versions of perl/Getopt::Long would just disappear over time and we'd eventually stop fixing up the tests to be backwards compatible when people stopped showing up saying that the tests failed on their ancient system.

What platforms are actually affected? RHEL3? Other ancient UNIX? I know the systems I was fixing up were ancient SunOS and IRIX.

Unfortunately (or fortunately, depending on how you look at it), I don't have access to any ancient systems to test on anymore. So I can't run the updated tests to make sure they still pass. The patches "look" fine to me though. :-)

-Brandon
Previous: brian m. carlsonNext: Tom G. Christensen
Message 29 of 31 in “[ANNOUNCE] Git v2.3.0-rc2”
  1. Junio C HamanoJan 27, 2015
  2. Broken makefile check for curl version on el4 [Re: [ANNOUNCE] Git v2.3.0-rc2]Tom G. Christensen, Jan 29, 2015
  3. Makefile: Handle broken curl version number in version checkTom G. Christensen, Jan 30, 2015
  4. Andreas SchwabJan 30, 2015
  5. Tom G. ChristensenJan 30, 2015
  6. Kyle J. McKayJan 30, 2015
  7. Junio C HamanoJan 30, 2015
  8. All gnupg tests broken on el4 [Re: [ANNOUNCE] Git v2.3.0-rc2]Tom G. Christensen, Jan 29, 2015
  9. Jeff KingJan 29, 2015
  10. Jeff KingJan 29, 2015
  11. Tom G. ChristensenJan 29, 2015
  12. Junio C HamanoJan 29, 2015
  13. Testsuite regression with perl 5.8.0 [Re: [ANNOUNCE] Git v2.3.0-rc2]Tom G. Christensen, Jan 29, 2015
  14. Jeff KingJan 29, 2015
  15. Tom G. ChristensenJan 30, 2015
  16. t9001: use older Getopt::Long boolean prefix '--no' rather than '--no-'Tom G. Christensen, Jan 30, 2015
  17. brian m. carlsonJan 30, 2015
  18. Kyle J. McKayJan 31, 2015
  19. Junio C HamanoFeb 2, 2015
  20. Kyle J. McKayFeb 2, 2015
  21. Junio C HamanoFeb 2, 2015
  22. Junio C HamanoFeb 12, 2015
  23. 0/2 Getopt::Long workaround in send-emailJunio C Hamano, Feb 13, 2015
  24. 1/2 git-send-email.perl: support no- prefix with older GetOptionsJunio C Hamano, Feb 13, 2015
  25. Brandon CaseyFeb 15, 2015
  26. 2/2 SQUASH??? t9001: turn --no$option workarounds to --no-$optionJunio C Hamano, Feb 13, 2015
  27. Kyle J. McKayFeb 13, 2015
  28. brian m. carlsonFeb 13, 2015
  29. Brandon CaseyFeb 15, 2015
  30. Tom G. ChristensenFeb 16, 2015
  31. Brandon CaseyFeb 16, 2015

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.