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

Re: [RFC] dropping support for ancient versions of curl

From
Tom G. Christensen <tgc@jupiterrise.com>
Date
Apr 5, 2017, 13:04 UTC
Message-ID
<4e30ae33-d508-9bb8-74e6-8204967c4538@jupiterrise.com>
In-Reply-To
<CACBZZX450tRRsy-Sj8igZthYov7UxFMRJ51M-b1cgYBLo782jQ@mail.gmail.com>
On 05/04/17 12:51, Ævar Arnfjörð Bjarmason wrote:
Show 8 quoted lines
> On Wed, Apr 5, 2017 at 11:33 AM, Tom G. Christensen <tgc@jupiterrise.com> wrote:
> Whoah. So my assumption in
> <CACBZZX78oKU5HuBEqb9qLy7--wcwhC_mW6x7Q+tB4suxohSCsQ@mail.gmail.com>
> that nobody was compiling this & thus not reporting failures was
> false. Rather there's an entire community & distribution mechanism
> around patching git for older EL versions, but the patches aren't
> making it upstream.
>

The community as I know it consists of me and EPEL5 (now dead and archived). The packages that I build are probably only used by me and the company I work for as they are not exactly easy to find via search engines. EPEL5 supported git 1.8.3.2 but the Fedora git specfile still contains all the infrastructure though I cannot know if it actually got used with anything later than 1.8.3.2.

I don't know of anyone that actually *needs* to use the latest git on RHEL < 5, myself included. I kept the support for RHEL < 5 because I could and it was good fun to tinker with.

Also I should say that testresults are good, no problems there except a few small nits as revealed in the specfile: %if %{?el4:1}0 # These tests fail with subversion 1.1.4 export GIT_SKIP_TESTS="t9140.4" %ifarch x86_64 # These tests fail with subversion 1.1.4 but only on x86_64 export GIT_SKIP_TESTS="$GIT_SKIP_TESTS t9106.7 t9106.8 t9106.9 t9106.10 t9137.4 t9164.5 t9164.6 t9164.7 t9164.8" %endif %endif %if %{?el3:1}%{?el4:1}0 # not ok 6 - url high-bit escapes export GIT_SKIP_TESTS="$GIT_SKIP_TESTS t0110.6" # not ok 32 - ref name 'heads/foo' is invalid export GIT_SKIP_TESTS="$GIT_SKIP_TESTS t1402.32" %endif %if %{?el3:1}0 # t7800 failed 17 among 56 test(s) export GIT_SKIP_TESTS="$GIT_SKIP_TESTS t7800" %endif

It's been a little while since I did a build without those exclusions but I doubt much has changed.

Show 15 quoted lines
> $ grep -h -e ^Subject -e ^Date *patch
> Date: Tue, 13 Oct 2009 11:34:11 +0200
> Subject: [PATCH 1/7] Make NO_PERL_MAKEMAKER behave more like
> Date: Fri, 13 Jun 2014 11:02:02 +0200
> Subject: [PATCH 2/7] Install man pages when NO_PERL_MAKEMAKER is used
> Date: Mon, 22 Sep 2014 13:42:50 +0200
> Subject: [PATCH 3/7] Allow svnrdump_sim.py to be used with Python 2.2
> Date: Tue, 8 Mar 2016 09:31:56 +0100
> Subject: [PATCH 4/7] Handle missing HTTP_CONNECTCODE in curl < 7.10.7
> Date: Tue, 23 Aug 2016 10:32:51 +0200
> Subject: [PATCH 5/7] Add support for gnupg < 1.4
> Date: Tue, 23 Aug 2016 18:15:13 +0200
> Subject: [PATCH 6/7] Handle missing CURLINFO_SSL_DATA_{IN,OUT}
> Date: Tue, 23 Aug 2016 18:26:54 +0200
> Subject: [PATCH 7/7] Do not use curl_easy_strerror with curl < 7.12.0
All original work done by me.
Show 5 quoted lines
> Date: Wed, 2 Feb 2011 21:24:44 -0500
> Subject: [PATCH] Restore vc-git.el for basic compatibility on EL-5
> Date: Mon, 23 Mar 2009 00:03:36 -0400
> Subject: [PATCH] git-cvsimport: Ignore cvsps-2.2b1 Branches: output
>

These two I can't claim credit for. They are lifted verbatim from Fedora/EPEL and as the headers reveal they were created by Todd Zullinger. I won't submit them for inclusion since I am not familiar with nor a user of the parts they touch.

Show 9 quoted lines
>> Patches can be found in the src.rpm, though I can also post them here as
>> patch series, they cover more than just curl.
>>
>> I don't use the el3 and el4 versions much any more and el5 use will also
>> drop of now as I'm busy converting machines from el5 to el7.
>
> It would be great to have them on-list, as far as I can tell they were
> never submitted? Is there some time/administrative reason for why
> you're not submitting them?

Well I recently took the time to clean them up with the intention of maybe finally submitting them but I never got that far.

The first one in the series I actually submitted many years ago but it was ultimately rejected.

I've submitted a few patches over the years to support older RHEL releases and some of them ended up being included.

> Some of these are many years old, it would
> be great to have them on-list for wider review & included so vanilla
> git works on these platforms.
>
I just posted them now. The series was made against the v2.12.2 tag.
-tgc
Previous: Jacob KellerNext: brian m. carlson
Message 39 of 48 in “[RFC] dropping support for ancient versions of curl”
  1. Jeff KingApr 4, 2017
  2. Jeff KingApr 4, 2017
  3. Jessie HernandezApr 4, 2017
  4. Ævar Arnfjörð BjarmasonApr 4, 2017
  5. Jeff KingApr 4, 2017
  6. Ævar Arnfjörð BjarmasonApr 4, 2017
  7. Johannes SchindelinApr 4, 2017
  8. Ævar Arnfjörð BjarmasonApr 4, 2017
  9. Brandon WilliamsApr 4, 2017
  10. Johannes SchindelinApr 4, 2017
  11. Brandon WilliamsApr 4, 2017
  12. Stefan BellerApr 4, 2017
  13. Johannes SchindelinApr 5, 2017
  14. Jeff KingApr 5, 2017
  15. Jeff KingApr 4, 2017
  16. Frank GevaertsApr 4, 2017
  17. Tom G. ChristensenApr 5, 2017
  18. Ævar Arnfjörð BjarmasonApr 5, 2017
  19. 0/7 Patches to support older RHEL releasesTom G. Christensen, Apr 5, 2017
  20. 3/7 Allow svnrdump_sim.py to be used with Python 2.2Tom G. Christensen, Apr 5, 2017
  21. Ævar Arnfjörð BjarmasonApr 5, 2017
  22. Tom G. ChristensenApr 5, 2017
  23. 1/7 Make NO_PERL_MAKEMAKER behave more like ExtUtils::MakeMakerTom G. Christensen, Apr 5, 2017
  24. 4/7 Handle missing HTTP_CONNECTCODE in curl < 7.10.7Tom G. Christensen, Apr 5, 2017
  25. Ævar Arnfjörð BjarmasonApr 5, 2017
  26. Franke, KnutApr 5, 2017
  27. 2/7 Install man pages when NO_PERL_MAKEMAKER is usedTom G. Christensen, Apr 5, 2017
  28. 5/7 Add support for gnupg < 1.4Tom G. Christensen, Apr 5, 2017
  29. Ævar Arnfjörð BjarmasonApr 5, 2017
  30. Junio C HamanoApr 13, 2017
  31. Ævar Arnfjörð BjarmasonApr 13, 2017
  32. 6/7 Handle missing CURLINFO_SSL_DATA_{IN,OUT}Tom G. Christensen, Apr 5, 2017
  33. Ævar Arnfjörð BjarmasonApr 5, 2017
  34. 7/7 Do not use curl_easy_strerror with curl < 7.12.0Tom G. Christensen, Apr 5, 2017
  35. Ævar Arnfjörð BjarmasonApr 5, 2017
  36. Jeff KingApr 6, 2017
  37. Junio C HamanoApr 13, 2017
  38. Jacob KellerApr 13, 2017
  39. Tom G. ChristensenApr 5, 2017
  40. brian m. carlsonApr 6, 2017
  41. Todd ZullingerApr 6, 2017
  42. Jeff KingApr 6, 2017
  43. Johannes SchindelinApr 7, 2017
  44. Jeff KingApr 10, 2017
  45. Jeff KingApr 6, 2017
  46. Tom G. ChristensenApr 6, 2017
  47. Jeff KingApr 7, 2017
  48. Junio C HamanoApr 14, 2017

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.