CEDET

Hi all,

  There have been a bunch of failures reported to the lists lately. I
really appreciate the help of running these tests on all these
platforms to help make CEDET more robust.

  For failures related to using semantic-symref with the grep
back-end, there has been more than one solution to the problem
proposed.

  I have finally figured out how to use the grep-template stuff to
fabricate new grep commands in a way that works for semantic symref.
This should get the unit test running for any platform supported by
these templates IFF you use an Emacs version with the template
support, such as Emacs 23.

  If you don't have the template support, I still removed the
offending command argument to xargs, so that should work too.

  If you encountered one of the many failures with this, please give
it another try from CVS.  If you reported something on this, I'm going
to consider the issue closed without a direct reply unless I hear back
from you.  Thanks!

  There was also a bug related to BSD Make and the integration tests.
I checked in a change to EDE to try to identify that a version of GNU
Make is available.  If you encountered this problem of a Makefile
syntax error, please try it out again.  Thanks!

Eric

-- 
          Eric Ludlam:                       eric@...
   Siege: http://www.siege-engine.com          Emacs: http://cedet.sourceforge.net

Re: [CEDET-devel] Fixes for making CEDET tests work. From: Alex Ott <alexott@gm...> - 2009-03-08 17:37 Hello symref-grep test now passed on Mac OS X/Carbon Emacs -- With best wishes, Alex Ott, MBA http://alexott.blogspot.com/ http://xtalk.msk.su/~ott/ http://alexott-ru.blogspot.com/

Re: [CEDET-devel] Fixes for making CEDET tests work. From: Eric M. Ludlam <eric@si...> - 2009-03-08 17:54 >>> Alex Ott <alexott@...> seems to think that: >Hello > >symref-grep test now passed on Mac OS X/Carbon Emacs Huzzah! Thanks for checking. Eric -- Eric Ludlam: eric@... Siege: http://www.siege-engine.com Emacs: http://cedet.sourceforge.net

Re: [CEDET-devel] Fixes for making CEDET tests work. From: David Engster <deng@ra...> - 2009-03-10 14:07 "Eric M. Ludlam" <eric@...> writes: > I have finally figured out how to use the grep-template stuff to > fabricate new grep commands in a way that works for semantic symref. > This should get the unit test running for any platform supported by > these templates IFF you use an Emacs version with the template > support, such as Emacs 23. > > If you don't have the template support, I still removed the > offending command argument to xargs, so that should work too. cedet-utest now works on FreeBSD 7.1, including symref-grep. > There was also a bug related to BSD Make and the integration tests. > I checked in a change to EDE to try to identify that a version of GNU > Make is available. If you encountered this problem of a Makefile > syntax error, please try it out again. Thanks! There are still problems with Make on FreeBSD, simply because 'compile' is still called with "make" instead of ede-make-command. Additionally, ede-make-check-version fails because GNU Make has a version output like "GNU Make 3.81" on my systems. Attached are the changes I made to make it work under FreeBSD. -David Index: ede/ede-make.el =================================================================== RCS file: /cvsroot/cedet/cedet/ede/ede-make.el,v retrieving revision 1.1 diff -u -r1.1 ede-make.el --- ede/ede-make.el 8 Mar 2009 12:51:27 -0000 1.1 +++ ede/ede-make.el 10 Mar 2009 14:03:47 -0000 @@ -84,7 +84,7 @@ "--version") ;; Check the buffer for the string (goto-char (point-min)) - (when (looking-at "GNU Make version \\([0-9][^,]+\\),") + (when (looking-at "GNU Make\\(?: version\\)? \\([0-9][^,]+\\),") (setq rev (match-string 1)) (setq ans (not (inversion-check-version rev nil ede-make-min-version)))) Index: tests/cedet-integ-test.el =================================================================== RCS file: /cvsroot/cedet/cedet/tests/cedet-integ-test.el,v retrieving revision 1.8 diff -u -r1.8 cedet-integ-test.el --- tests/cedet-integ-test.el 24 Jan 2009 04:56:26 -0000 1.8 +++ tests/cedet-integ-test.el 10 Mar 2009 14:03:47 -0000 @@ -242,7 +242,7 @@ ;; 1 f) Create a build file. (ede-proj-regenerate) ;; 1 g) build the sources. - (compile "make") + (compile ede-make-command) (while compilation-in-progress (accept-process-output) Index: tests/cit-cpp.el =================================================================== RCS file: /cvsroot/cedet/cedet/tests/cit-cpp.el,v retrieving revision 1.3 diff -u -r1.3 cit-cpp.el --- tests/cit-cpp.el 6 Mar 2009 14:24:33 -0000 1.3 +++ tests/cit-cpp.el 10 Mar 2009 14:03:47 -0000 @@ -152,7 +152,7 @@ ;; 1 g) build the sources. ;; Direct compile to test that make fails properly. - (compile "make") + (compile ede-make-command) ;; @todo - verify make error status (while compilation-in-progress (accept-process-output)

Re: [CEDET-devel] Fixes for making CEDET tests work. From: Eric M. Ludlam <eric@si...> - 2009-03-12 02:30 >>> David Engster <deng@...> seems to think that: >"Eric M. Ludlam" <eric@...> writes: [ ... ] >> If you don't have the template support, I still removed the >> offending command argument to xargs, so that should work too. > >cedet-utest now works on FreeBSD 7.1, including symref-grep. Huzzah! >> There was also a bug related to BSD Make and the integration tests. >> I checked in a change to EDE to try to identify that a version of GNU >> Make is available. If you encountered this problem of a Makefile >> syntax error, please try it out again. Thanks! > >There are still problems with Make on FreeBSD, simply because 'compile' >is still called with "make" instead of ede-make-command. Additionally, >ede-make-check-version fails because GNU Make has a version output like >"GNU Make 3.81" on my systems. > >Attached are the changes I made to make it work under FreeBSD. [ ... ] Thanks for the patches. I applied them. I also figured out how to figure out why the gnustep project was asking questions, and fixed that too. It should work much better now. Eric -- Eric Ludlam: eric@... Siege: http://www.siege-engine.com Emacs: http://cedet.sourceforge.net

Re: [CEDET-devel] Fixes for making CEDET tests work. From: Matzi Kratzi <matzikratzi@gm...> - 2009-03-15 11:42 Hi Eric, I just tried to build cedet for Vista. The build has some problems. I build cedet with cedet-build: load cedet-build.el M-x eval-buffer M-x cedet-build-in-default-emacs After a while I get the following questions. I answer yes to them all. "Save buffer wisent-java-tags-wy.el first? (y or n)" "Save buffer wisent-awk-wy.el first? (y or n)" "Save buffer wisent-calc-wy.el first? (y or n)" "Save buffer wisent-cim-wy.el first? (y or n)" "Save buffer wisent-java-wy.el first? (y or n)" "Save buffer wisent-python-wy.el first? (y or n)" "Save buffer wisent-jv-wy.el first? (y or n)" "Save buffer semantic-c-by.el first? (y or n)" "Save buffer semantic-make-by.el first? (y or n)" "Save buffer semantic-scm-by.el first? (y or n)" "Save buffer semantic-erlang-by.el first? (y or n)" "Save buffer srecode-tempate-wy.el first? (y or n)" "Save buffer wisent-dot- wy.el first? (y or n)" On Sun, Mar 8, 2009 at 4:49 PM, Eric M. Ludlam <eric@...> wrote: > Hi all, > >  There have been a bunch of failures reported to the lists lately. I > really appreciate the help of running these tests on all these > platforms to help make CEDET more robust. > >  For failures related to using semantic-symref with the grep > back-end, there has been more than one solution to the problem > proposed. > >  I have finally figured out how to use the grep-template stuff to > fabricate new grep commands in a way that works for semantic symref. > This should get the unit test running for any platform supported by > these templates IFF you use an Emacs version with the template > support, such as Emacs 23. > >  If you don't have the template support, I still removed the > offending command argument to xargs, so that should work too. > >  If you encountered one of the many failures with this, please give > it another try from CVS.  If you reported something on this, I'm going > to consider the issue closed without a direct reply unless I hear back > from you.  Thanks! > >  There was also a bug related to BSD Make and the integration tests. > I checked in a change to EDE to try to identify that a version of GNU > Make is available.  If you encountered this problem of a Makefile > syntax error, please try it out again.  Thanks! > > Eric > > -- >          Eric Ludlam:                       eric@... >   Siege: http://www.siege-engine.com          Emacs: http://cedet.sourceforge.net > > ------------------------------------------------------------------------------ > Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA > -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise > -Strategies to boost innovation and cut costs with open source participation > -Receive a $600 discount off the registration fee with the source code: SFAD > http://p.sf.net/sfu/XcvMzF8H > _______________________________________________ > Cedet-devel mailing list > Cedet-devel@... > https://lists.sourceforge.net/lists/listinfo/cedet-devel >

Re: [CEDET-devel] Fixes for making CEDET tests work. From: Matzi Kratzi <matzikratzi@gm...> - 2009-03-15 11:52 (I try to write the whole mail this time before sending it) Hi Eric, I just tried to build cedet for Vista. The build has some problems. I build cedet with cedet-build: load cedet-build.el M-x eval-buffer M-x cedet-build-in-default-emacs After a while I get the following questions. I answer yes to them all. "Save buffer wisent-java-tags-wy.el first? (y or n)" "Save buffer wisent-awk-wy.el first? (y or n)" "Save buffer wisent-calc-wy.el first? (y or n)" "Save buffer wisent-cim-wy.el first? (y or n)" "Save buffer wisent-java-wy.el first? (y or n)" "Save buffer wisent-python-wy.el first? (y or n)" "Save buffer wisent-jv-wy.el first? (y or n)" "Save buffer semantic-c-by.el first? (y or n)" "Save buffer semantic-make-by.el first? (y or n)" "Save buffer semantic-scm-by.el first? (y or n)" "Save buffer semantic-erlang-by.el first? (y or n)" "Save buffer srecode-tempate-wy.el first? (y or n)" "Save buffer wisent-dot- wy.el first? (y or n)" "Save buffer wisent-csharp- wy.el first? (y or n)" "Save buffer wisent-ruby-wy.el first? (y or n)" "Save buffer wisent-php-wy.el first? (y or n)" *CEDET BYTECOMPILE* ends with Building project tests Project: cit Target Lisp... 7 compiled, 0 up to date. Thank you for your work. /Mats On Sun, Mar 15, 2009 at 12:42 PM, Matzi Kratzi <matzikratzi@...> wrote: > Hi Eric, > I just tried to build cedet for Vista. The build has some problems. > > I build cedet with cedet-build: > load cedet-build.el > M-x eval-buffer > M-x cedet-build-in-default-emacs > > After a while I get the following questions. I answer yes to them all. > > "Save buffer wisent-java-tags-wy.el first? (y or n)" > "Save buffer wisent-awk-wy.el first? (y or n)" > "Save buffer wisent-calc-wy.el first? (y or n)" > "Save buffer wisent-cim-wy.el first? (y or n)" > "Save buffer wisent-java-wy.el first? (y or n)" > "Save buffer wisent-python-wy.el first? (y or n)" > "Save buffer wisent-jv-wy.el first? (y or n)" > "Save buffer semantic-c-by.el first? (y or n)" > "Save buffer semantic-make-by.el first? (y or n)" > "Save buffer semantic-scm-by.el first? (y or n)" > "Save buffer semantic-erlang-by.el first? (y or n)" > "Save buffer srecode-tempate-wy.el first? (y or n)" > "Save buffer wisent-dot- wy.el first? (y or n)" > > > On Sun, Mar 8, 2009 at 4:49 PM, Eric M. Ludlam <eric@...> wrote: >> Hi all, >> >>  There have been a bunch of failures reported to the lists lately. I >> really appreciate the help of running these tests on all these >> platforms to help make CEDET more robust. >> >>  For failures related to using semantic-symref with the grep >> back-end, there has been more than one solution to the problem >> proposed. >> >>  I have finally figured out how to use the grep-template stuff to >> fabricate new grep commands in a way that works for semantic symref. >> This should get the unit test running for any platform supported by >> these templates IFF you use an Emacs version with the template >> support, such as Emacs 23. >> >>  If you don't have the template support, I still removed the >> offending command argument to xargs, so that should work too. >> >>  If you encountered one of the many failures with this, please give >> it another try from CVS.  If you reported something on this, I'm going >> to consider the issue closed without a direct reply unless I hear back >> from you.  Thanks! >> >>  There was also a bug related to BSD Make and the integration tests. >> I checked in a change to EDE to try to identify that a version of GNU >> Make is available.  If you encountered this problem of a Makefile >> syntax error, please try it out again.  Thanks! >> >> Eric >> >> -- >>          Eric Ludlam:                       eric@... >>   Siege: http://www.siege-engine.com          Emacs: http://cedet.sourceforge.net >> >> ------------------------------------------------------------------------------ >> Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA >> -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise >> -Strategies to boost innovation and cut costs with open source participation >> -Receive a $600 discount off the registration fee with the source code: SFAD >> http://p.sf.net/sfu/XcvMzF8H >> _______________________________________________ >> Cedet-devel mailing list >> Cedet-devel@... >> https://lists.sourceforge.net/lists/listinfo/cedet-devel >> >

Re: [CEDET-devel] Fixes for making CEDET tests work. From: Eric M. Ludlam <eric@si...> - 2009-03-15 12:24 Hi, It looks like the full build succeeded, but for some reason the compiled grammars did not save, or were modified after they were compiled. Things to check are: When it is asking the questions, do the files exist? Are their time-stamps up to date? Most are in cedet/semantic/bovine and cedet/semantic/wisent. Others are in a few of the other directories. Is there a matching .elc file for them? If these things exist, then I'd guess they are being changed after being saved, perhaps by a coding system shift? Perhaps by some hook you have? Try starting emacs with -Q or -q to not load your configuration. Perhaps that will fix it. Thanks Eric >>> Matzi Kratzi <matzikratzi@...> seems to think that: >(I try to write the whole mail this time before sending it) >Hi Eric, >I just tried to build cedet for Vista. The build has some problems. > >I build cedet with cedet-build: >load cedet-build.el >M-x eval-buffer >M-x cedet-build-in-default-emacs > >After a while I get the following questions. I answer yes to them all. > >"Save buffer wisent-java-tags-wy.el first? (y or n)" >"Save buffer wisent-awk-wy.el first? (y or n)" >"Save buffer wisent-calc-wy.el first? (y or n)" >"Save buffer wisent-cim-wy.el first? (y or n)" >"Save buffer wisent-java-wy.el first? (y or n)" >"Save buffer wisent-python-wy.el first? (y or n)" >"Save buffer wisent-jv-wy.el first? (y or n)" >"Save buffer semantic-c-by.el first? (y or n)" >"Save buffer semantic-make-by.el first? (y or n)" >"Save buffer semantic-scm-by.el first? (y or n)" >"Save buffer semantic-erlang-by.el first? (y or n)" >"Save buffer srecode-tempate-wy.el first? (y or n)" >"Save buffer wisent-dot- wy.el first? (y or n)" >"Save buffer wisent-csharp- wy.el first? (y or n)" >"Save buffer wisent-ruby-wy.el first? (y or n)" >"Save buffer wisent-php-wy.el first? (y or n)" > >*CEDET BYTECOMPILE* ends with > >Building project tests > Project: cit > Target Lisp... 7 compiled, 0 up to date. > >Thank you for your work. > [ ... ]

1 message has been excluded from this view by a project administrator.