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

Re: [PATCH v2 5/6] clone: fix hostname parsing when guessing dir

From
Patrick Steinhardt <ps@pks.im>
Date
Aug 3, 2015, 08:34 UTC
Message-ID
<20150803083414.GA23997@pks-xps.Speedport_W_723V_1_37_000>
In-Reply-To
<xmqqfv457fkc.fsf@gitster.dls.corp.google.com>
On Thu, Jul 30, 2015 at 09:53:07AM -0700, Junio C Hamano wrote:
Show 35 quoted lines
> Junio C Hamano <gitster@pobox.com> writes:
> 
> > Well, but there is the above "or we may not" ;-)
> >
> >> But actually you are right, currently I still have the old logic
> >> in place that splits on colons in the path component. 
> >
> > Yes.  The reason why I suggested the simple route was exactly
> > because I noticed that you didn't seem to care about the above
> > "$site/foo:bar.git/" => "$site/foo:bar" => "bar" transform.
> >
> > And I think people might depend on that behaviour.  "Fixing" that
> > may even be seen as a regression.
> >
> > When was the last time you created a foo@bar.git repository?
> 
> Actually, this was an unrelated question and a wrong one to ask at
> that.
> 
> Even though I personally haven't created foo:bar.git repository,
> because it is no longer 2005, it is highly likely that somewhere
> there is such a person who depends on the current behaviour of
> turning that to "bar" on the cloned side.  Similarly, even if we the
> people who read the Git mailing list collectively do not know
> anybody who has foo@bar.git repository, it is highly likely that
> somewhere there is such a person who depends on the current
> behaviour of turning that to "foo@bar" on the cloned side.
> 
> So the ideal would be to keep turning $site/foo:bar.git to bar,
> $site/foo@bar.git to foo@bar, and $scheme//u@p:$host/ to $host.
> 
> And it would be ideal if we do so without much code churn.  "The
> whole site is dedicated to host a single repository at the root" is
> a highly unlikely set-up and it feels wasteful to spend too much
> effort on.

One more question for backwards compatibility remains then. Currently when we clone something like 'http://example.com:2222/' we'd create a git repository '2222' as we'd split on the first occurrence of ':'. Should we remain backwards compatible here, as well, or change the behavior to use 'example.com' as repository name?

Previous: Junio C HamanoNext: Jeff King
Message 17 of 53 in “clone: fix repo name when cloning a server's root”
  1. clone: fix repo name when cloning a server's rootPatrick Steinhardt, Jul 27, 2015
  2. Duy NguyenJul 27, 2015
  3. Patrick SteinhardtJul 27, 2015
  4. Junio C HamanoJul 27, 2015
  5. 0/6 clone: fix repo name when cloning a server's rootPatrick Steinhardt, Jul 29, 2015
  6. 1/6 tests: fix broken && chains in t1509-root-worktreePatrick Steinhardt, Jul 29, 2015
  7. 2/6 tests: fix cleanup after tests in t1509-root-worktreePatrick Steinhardt, Jul 29, 2015
  8. 3/6 connect: expose parse_connect_url()Patrick Steinhardt, Jul 29, 2015
  9. 4/6 connect: move error check to caller of parse_connect_urlPatrick Steinhardt, Jul 29, 2015
  10. Eric SunshineJul 29, 2015
  11. Patrick SteinhardtJul 30, 2015
  12. 5/6 clone: fix hostname parsing when guessing dirPatrick Steinhardt, Jul 29, 2015
  13. Junio C HamanoJul 29, 2015
  14. Patrick SteinhardtJul 30, 2015
  15. Junio C HamanoJul 30, 2015
  16. Junio C HamanoJul 30, 2015
  17. Patrick SteinhardtAug 3, 2015
  18. Jeff KingAug 3, 2015
  19. Junio C HamanoAug 3, 2015
  20. 6/6 clone: add tests for cloning with empty pathPatrick Steinhardt, Jul 29, 2015
  21. Eric SunshineJul 30, 2015
  22. Junio C HamanoJul 31, 2015
  23. Patrick SteinhardtJul 31, 2015
  24. 0/6 fix repo name when cloning a server's rootPatrick Steinhardt, Aug 4, 2015
  25. 1/6 tests: fix broken && chains in t1509-root-worktreePatrick Steinhardt, Aug 4, 2015
  26. 2/6 tests: fix cleanup after tests in t1509-root-worktreePatrick Steinhardt, Aug 4, 2015
  27. 3/6 clone: do not include authentication data in guessed dirPatrick Steinhardt, Aug 4, 2015
  28. 4/6 clone: do not use port number as dir namePatrick Steinhardt, Aug 4, 2015
  29. 5/6 clone: abort if no dir name could be guessedPatrick Steinhardt, Aug 4, 2015
  30. 6/6 clone: add tests for cloning with empty pathPatrick Steinhardt, Aug 4, 2015
  31. Eric SunshineAug 4, 2015
  32. Junio C HamanoAug 5, 2015
  33. Jeff KingAug 5, 2015
  34. Torsten BögershausenAug 6, 2015
  35. Junio C HamanoAug 6, 2015
  36. 0/3 fix repo name when cloning a server's rootPatrick Steinhardt, Aug 5, 2015
  37. 1/3 clone: do not include authentication data in guessed dirPatrick Steinhardt, Aug 5, 2015
  38. Junio C HamanoAug 5, 2015
  39. Junio C HamanoAug 5, 2015
  40. Junio C HamanoAug 5, 2015
  41. Patrick SteinhardtAug 6, 2015
  42. Junio C HamanoAug 7, 2015
  43. Patrick SteinhardtAug 8, 2015
  44. 2/3 clone: do not use port number as dir namePatrick Steinhardt, Aug 5, 2015
  45. 3/3 clone: abort if no dir name could be guessedPatrick Steinhardt, Aug 5, 2015
  46. Junio C HamanoAug 5, 2015
  47. 0/5 Improve guessing of repository namesPatrick Steinhardt, Aug 10, 2015
  48. 1/5 clone: add tests for output directoryPatrick Steinhardt, Aug 10, 2015
  49. 2/5 clone: use computed length in guess_dir_namePatrick Steinhardt, Aug 10, 2015
  50. 3/5 clone: do not include authentication data in guessed dirPatrick Steinhardt, Aug 10, 2015
  51. 4/5 clone: do not use port number as dir namePatrick Steinhardt, Aug 10, 2015
  52. 5/5 clone: abort if no dir name could be guessedPatrick Steinhardt, Aug 10, 2015
  53. Junio C HamanoAug 10, 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.