Thread

1. .gitignore patch for coverage builds

   Kevin Grittner <kevin.grittner@wicourts.gov> — 2011-01-26T21:04:33Z

   Building for coverage and running the reports littered my tree with
   files which should probably be in .gitignore for just such a
   contingency.  Patch attached.
    
   -Kevin
   
   

2. Re: .gitignore patch for coverage builds

   Kevin Grittner <kevin.grittner@wicourts.gov> — 2011-01-26T21:38:31Z

   "Kevin Grittner" <Kevin.Grittner@wicourts.gov> wrote:
    
   > Patch attached.
    
   The coverage directory belongs under "Local excludes in root
   directory".  Version 2.
    
   -Kevin
   
   

3. Re: .gitignore patch for coverage builds

   Tom Lane <tgl@sss.pgh.pa.us> — 2011-01-26T21:44:11Z

   "Kevin Grittner" <Kevin.Grittner@wicourts.gov> writes:
   > Building for coverage and running the reports littered my tree with
   > files which should probably be in .gitignore for just such a
   > contingency.  Patch attached.
   
   Ick.  That's an awful lot of stuff to have global ignores for.
   
   Perhaps we should recommend people do coverage tests in separate
   build trees, instead.
   
   			regards, tom lane
   
   

4. Re: .gitignore patch for coverage builds

   Robert Haas <robertmhaas@gmail.com> — 2011-01-26T22:01:17Z

   On Wed, Jan 26, 2011 at 4:44 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
   > "Kevin Grittner" <Kevin.Grittner@wicourts.gov> writes:
   >> Building for coverage and running the reports littered my tree with
   >> files which should probably be in .gitignore for just such a
   >> contingency.  Patch attached.
   >
   > Ick.  That's an awful lot of stuff to have global ignores for.
   
   The "coverage" directory ignore seems a little icky, but the rest
   seems unlikely to pick up anything incidental.
   
   -- 
   Robert Haas
   EnterpriseDB: http://www.enterprisedb.com
   The Enterprise PostgreSQL Company
   
   

5. Re: .gitignore patch for coverage builds

   Tom Lane <tgl@sss.pgh.pa.us> — 2011-01-26T22:20:52Z

   Robert Haas <robertmhaas@gmail.com> writes:
   > On Wed, Jan 26, 2011 at 4:44 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
   >> Ick. That's an awful lot of stuff to have global ignores for.
   
   > The "coverage" directory ignore seems a little icky, but the rest
   > seems unlikely to pick up anything incidental.
   
   Tying /coverage to the root as in his V2 makes that better, but I'm
   still unexcited about the thesis that we should auto-ignore the results
   of any random tool somebody wants to run in their source tree.
   
   			regards, tom lane
   
   

6. Re: .gitignore patch for coverage builds

   Kevin Grittner <kevin.grittner@wicourts.gov> — 2011-01-26T22:26:07Z

   Tom Lane <tgl@sss.pgh.pa.us> wrote:
    
   > I'm still unexcited about the thesis that we should auto-ignore
   > the results of any random tool somebody wants to run in their
   > source tree.
    
   Hos about just the tools supported by our documentation, configure
   file and make file?
    
   -Kevin
   
   

7. Re: .gitignore patch for coverage builds

   Alvaro Herrera <alvherre@commandprompt.com> — 2011-01-26T22:41:39Z

   Excerpts from Tom Lane's message of mié ene 26 19:20:52 -0300 2011:
   > Robert Haas <robertmhaas@gmail.com> writes:
   > > On Wed, Jan 26, 2011 at 4:44 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
   > >> Ick. That's an awful lot of stuff to have global ignores for.
   > 
   > > The "coverage" directory ignore seems a little icky, but the rest
   > > seems unlikely to pick up anything incidental.
   > 
   > Tying /coverage to the root as in his V2 makes that better,
   
   Hmm, I don't think that works, because you can run "make coverage" in
   any subdir and it will create a "coverage" subdir there.
   
   > but I'm
   > still unexcited about the thesis that we should auto-ignore the results
   > of any random tool somebody wants to run in their source tree.
   
   Well, in this case it's not any random tool, because it's integrated
   into our makefiles.
   
   -- 
   Álvaro Herrera <alvherre@commandprompt.com>
   The PostgreSQL Company - Command Prompt, Inc.
   PostgreSQL Replication, Consulting, Custom Development, 24x7 support
   
   

8. Re: .gitignore patch for coverage builds

   Jeff Janes <jeff.janes@gmail.com> — 2011-02-15T00:38:27Z

   On Wed, Jan 26, 2011 at 2:41 PM, Alvaro Herrera
   <alvherre@commandprompt.com> wrote:
   > Excerpts from Tom Lane's message of mié ene 26 19:20:52 -0300 2011:
   >> Robert Haas <robertmhaas@gmail.com> writes:
   >> > On Wed, Jan 26, 2011 at 4:44 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
   >> >> Ick. That's an awful lot of stuff to have global ignores for.
   >>
   >> > The "coverage" directory ignore seems a little icky, but the rest
   >> > seems unlikely to pick up anything incidental.
   >>
   >> Tying /coverage to the root as in his V2 makes that better,
   >
   > Hmm, I don't think that works, because you can run "make coverage" in
   > any subdir and it will create a "coverage" subdir there.
   
   I like being told that I have a coverage directory outstanding when I
   run "git status".
   
   The hundreds of other files, not so much.
   
   
   >> but I'm
   >> still unexcited about the thesis that we should auto-ignore the results
   >> of any random tool somebody wants to run in their source tree.
   >
   > Well, in this case it's not any random tool, because it's integrated
   > into our makefiles.
   
   I agree.  Should this be added to commit-fest 2011-Next?
   
   Also, should "make clean-coverage" be changed to remove all of those
   files from the entire tree and not just root?
   
   Cheers,
   
   Jeff
   
   

9. Re: .gitignore patch for coverage builds

   Robert Haas <robertmhaas@gmail.com> — 2011-02-15T03:35:27Z

   On Mon, Feb 14, 2011 at 7:38 PM, Jeff Janes <jeff.janes@gmail.com> wrote:
   > On Wed, Jan 26, 2011 at 2:41 PM, Alvaro Herrera
   > <alvherre@commandprompt.com> wrote:
   >> Excerpts from Tom Lane's message of mié ene 26 19:20:52 -0300 2011:
   >>> Robert Haas <robertmhaas@gmail.com> writes:
   >>> > On Wed, Jan 26, 2011 at 4:44 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
   >>> >> Ick. That's an awful lot of stuff to have global ignores for.
   >>>
   >>> > The "coverage" directory ignore seems a little icky, but the rest
   >>> > seems unlikely to pick up anything incidental.
   >>>
   >>> Tying /coverage to the root as in his V2 makes that better,
   >>
   >> Hmm, I don't think that works, because you can run "make coverage" in
   >> any subdir and it will create a "coverage" subdir there.
   >
   > I like being told that I have a coverage directory outstanding when I
   > run "git status".
   >
   > The hundreds of other files, not so much.
   >
   >
   >>> but I'm
   >>> still unexcited about the thesis that we should auto-ignore the results
   >>> of any random tool somebody wants to run in their source tree.
   >>
   >> Well, in this case it's not any random tool, because it's integrated
   >> into our makefiles.
   >
   > I agree.  Should this be added to commit-fest 2011-Next?
   
   I think there's little reason not to go ahead and commit this now.
   It's a trivial patch, Tom is the only one objecting, and there are at
   least four votes on the other side.  The only question in my mind is
   whether we ought to try to ignore the coverage directories, or just
   the other glob patterns.
   
   -- 
   Robert Haas
   EnterpriseDB: http://www.enterprisedb.com
   The Enterprise PostgreSQL Company
   
   

10. Re: .gitignore patch for coverage builds

    Peter Eisentraut <peter_e@gmx.net> — 2011-03-08T20:11:47Z

    On ons, 2011-01-26 at 15:38 -0600, Kevin Grittner wrote:
    > "Kevin Grittner" <Kevin.Grittner@wicourts.gov> wrote:
    >  
    > > Patch attached.
    >  
    > The coverage directory belongs under "Local excludes in root
    > directory".  Version 2.
    
    I have committed a simplified version of this, except the coverage/
    directory, which some people had issues with.  I think it's OK to
    "manually" ignore that.