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

Re: [PATCH] http-push: push <remote> :<branch> deletes remote branch

From
Johannes Schindelin <johannes.schindelin@gmx.de>
Date
Feb 19, 2008, 13:31 UTC
Message-ID
<alpine.LSU.1.00.0802191328040.30505@racer.site>
In-Reply-To
<20080219132410.GA6553@glandium.org>
Hi,
On Tue, 19 Feb 2008, Mike Hommey wrote:
Show 26 quoted lines
> On Tue, Feb 19, 2008 at 01:17:24PM +0000, Johannes Schindelin <Johannes.Schindelin@gmx.de> wrote:
> 
> > On Tue, 19 Feb 2008, Clemens Buchacher wrote:
> > 
> > > On Mon, Feb 18, 2008 at 05:34:45PM +0000, Johannes Schindelin wrote:
> > >
> > > > Thanks for the patch, I think it good.  Now, if we only had 
> > > > automated tests... ;-)
> > > 
> > > Thank you. Implementing a test should be fairly difficult, since we 
> > > need at least a WebDAV capable webserver, right? Do you having 
> > > something in mind?
> > 
> > Actually, thinking about it again, it should not be _that_ difficult.  
> > We would need to ship a httpd.conf.in, pick a port (maybe 8111, 
> > because that is already used in t5703-daemon.sh), generate httpd.conf 
> > and then start Apache with that.
> > 
> > If Apache does not want to start, skip this test.  (This might happen 
> > due to Apache not being installed, or the DAV modules not being 
> > installed, or Apache not finding them, or...)
> 
> Except that means coping with the fact that modules are probably not 
> installed in the same place on all distros, or can have been built in 
> apache instead of being DSOs, etc. The server itself doesn't even have 
> the same name across distros (httpd vs. apache vs. apache2)

I think this would not be too much of a hassle. These days, an Apache2 is installed on many machines, and it is better to have tests, than to have no tests, even if they are not exercised everywhere, right?

Besides, for those machines that do not have Apache installed (such as MinGW based setups), we can always add yet another environment variable, say "HTTP_PUSH_TEST_URL" to tell the test that there is already a web server set up at that URL, ready to be used.

Ciao, Dscho

Previous: Mike HommeyNext: Junio C Hamano
Message 10 of 27 in “http-push: Fix error message for push <remote> :<branch>”
  1. http-push: Fix error message for push <remote> :<branch>Clemens Buchacher, Feb 18, 2008
  2. Johannes SchindelinFeb 18, 2008
  3. http-push: push <remote> :<branch> deletes remote branchClemens Buchacher, Feb 18, 2008
  4. http-push: push <remote> :<branch> deletes remote branchClemens Buchacher, Feb 18, 2008
  5. Johannes SchindelinFeb 18, 2008
  6. Johannes SchindelinFeb 18, 2008
  7. Clemens BuchacherFeb 19, 2008
  8. Johannes SchindelinFeb 19, 2008
  9. Mike HommeyFeb 19, 2008
  10. Johannes SchindelinFeb 19, 2008
  11. Junio C HamanoFeb 19, 2008
  12. Johannes SchindelinFeb 19, 2008
  13. http-push: add regression testsClemens Buchacher, Feb 23, 2008
  14. Mike HommeyFeb 24, 2008
  15. http-push: add regression testsClemens Buchacher, Feb 24, 2008
  16. Mike HommeyFeb 24, 2008
  17. Clemens BuchacherFeb 24, 2008
  18. http-push: add regression testsClemens Buchacher, Feb 25, 2008
  19. Junio C HamanoFeb 26, 2008
  20. http-push: add regression testsClemens Buchacher, Feb 27, 2008
  21. Mike HommeyFeb 27, 2008
  22. Johannes SchindelinFeb 27, 2008
  23. Clemens BuchacherFeb 27, 2008
  24. 1/2 http-push: push <remote> :<branch> deletes remote branchClemens Buchacher, Feb 27, 2008
  25. 2/2 http-push: add regression testsClemens Buchacher, Feb 27, 2008
  26. Johannes SchindelinFeb 27, 2008
  27. Mike HommeyFeb 26, 2008

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.