You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Karl Fogel <kf...@red-bean.com> on 2008/01/16 22:07:30 UTC
failing tests on reintegrate branch
Posting this while I work on it:
Over all RA layers, these four tests fail in the head of the
reintegrate branch right now:
log_tests.py 17: test 'svn log -g' on a branching revision
merge_tests.py 61: merge fails if subtree is deleted on src
merge_tests.py 77: subtrees added after start of merge range are ok
blame_tests.py 10: test 'svn blame -g'
I first looked at log_tests 17. That test makes a bunch of mergey
changes, and then runs 'log -g', expecting that only r10 will show up
in the result (in other words, no "Merged via: ..." lines). But in
IRC just now, hwright said not to worry about log -g tests:
<hwright> kfogel: I wouldn't worry too much about log test 17.
<hwright> *all* the 'log -g' tests are having problems, which I hope
to fix with some of the new goodies on the reintegrate
branch.
<kfogel> hwright: hmmmm
<kfogel> the fact that reintegrate affects it *at all* is what
bothers me
<kfogel> I don't care whether the test is wrong or right right now
<kfogel> what bugs me is that it's behaving differently on trunk vs
reintegrate branch
<kfogel> Is there some reason why this is a reasonable thing for it
to do?
<kfogel> Should I just XFail it and move on to the next of the four
failing tests? (About to post to dev@.)
<hwright> I'm not familiar enough with the guts of the reintegrate
branch to comment on that. (<- reasonableness of test
failure)
<hwright> Yeah, I XFailing is probably appropriate.
So, I'm going to move on to the merge_tests. But in case anyone's
curious, below are some interesting things about log_tests 17.
In the output below, the command 'ts' is a pristine trunk Subversion
build (appx r28940), and 'rs' is a pristine reintegrate branch build.
In order, I invoke 'log -g -r10' like this:
1. using trunk svn, on the test data produced by trunk svn
2. using trunk svn, on the test data produced by reintegrate svn
3. using reintegrate svn, on the test data produced by trunk svn
4. using reintegrate svn, on the test data produced by reintegrate svn
Note how only (4) produces the bad output (the output that causes the
test to fail):
1. $ ts log -g -r10 /home/kfogel/src/subversion/subversion/tests/cmdline/svn-test-work/working_copies/log_tests-17/branches/b
------------------------------------------------------------------------
r10 | jrandom | 2008-01-16 16:30:08 -0500 (Wed, 16 Jan 2008) | 1 line
Create branches/b from trunk
------------------------------------------------------------------------
2. $ ts log -g -r10 /home/kfogel/src/reintegrate/subversion/tests/cmdline/svn-test-work/working_copies/log_tests-17/branches/b
------------------------------------------------------------------------
r10 | jrandom | 2008-01-16 16:46:21 -0500 (Wed, 16 Jan 2008) | 1 line
Create branches/b from trunk
------------------------------------------------------------------------
3. $ rs log -g -r10 /home/kfogel/src/subversion/subversion/tests/cmdline/svn-test-work/working_copies/log_tests-17/branches/b
------------------------------------------------------------------------
r10 | jrandom | 2008-01-16 16:30:08 -0500 (Wed, 16 Jan 2008) | 1 line
Create branches/b from trunk
------------------------------------------------------------------------
4. $ rs log -g -r10 /home/kfogel/src/reintegrate/subversion/tests/cmdline/svn-test-work/working_copies/log_tests-17/branches/b
------------------------------------------------------------------------
r10 | jrandom | 2008-01-16 16:46:21 -0500 (Wed, 16 Jan 2008) | 1 line
Create branches/b from trunk
------------------------------------------------------------------------
r7 | jrandom | 2008-01-16 16:46:21 -0500 (Wed, 16 Jan 2008) | 1 line
Merged via: r10
Add omicron to branches/a. It will be blocked from merging in r8.
------------------------------------------------------------------------
r4 | jrandom | 2008-01-16 16:46:20 -0500 (Wed, 16 Jan 2008) | 1 line
Merged via: r10
Add the file 'upsilon', and change some other files.
------------------------------------------------------------------------
r3 | jrandom | 2008-01-16 16:46:20 -0500 (Wed, 16 Jan 2008) | 1 line
Merged via: r10
Create branches/a from trunk.
------------------------------------------------------------------------
r2 | jrandom | 2008-01-16 16:46:20 -0500 (Wed, 16 Jan 2008) | 1 line
Merged via: r10
Import greek tree into trunk.
------------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: failing tests on reintegrate branch
Posted by Paul Burba <pb...@collab.net>.
> -----Original Message-----
> From: Karl Fogel [mailto:kfogel@red-bean.com]
> Sent: Wednesday, January 16, 2008 5:21 PM
> To: dev@subversion.tigris.org
> Subject: Re: failing tests on reintegrate branch
>
> Karl Fogel <kf...@red-bean.com> writes:
> > Over all RA layers, these four tests fail in the head of the
> > reintegrate branch right now:
> >
> > log_tests.py 17: test 'svn log -g' on a branching revision
> > merge_tests.py 61: merge fails if subtree is deleted on src
> > merge_tests.py 77: subtrees added after start of merge
> range are ok
> > blame_tests.py 10: test 'svn blame -g'
>
> In IRC, we've come to the conclusion that these should be
> ignored on the reintegrate branch for now. The log and blame
> failures are "expected", in some sense (that code is in flux
> anyway; in the log case, the reintegrate branch exposed some
> latent bugs, in the blame case a similar thing has probably
> happened even though hwright didn't say that exactly in our
> conversation).
>
> Meanwhile, the two merge tests are related to:
>
> http://svn.haxx.se/dev/archive-2008-01/0008.shtml
>
> Thanks to hwright, glasser, and pburba for the help.
See the latest in the above thread for the resolution of the failures in
merge_tests.py 61 and 77.
> -Karl
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
> For additional commands, e-mail: dev-help@subversion.tigris.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: failing tests on reintegrate branch
Posted by Karl Fogel <kf...@red-bean.com>.
Karl Fogel <kf...@red-bean.com> writes:
> Over all RA layers, these four tests fail in the head of the
> reintegrate branch right now:
>
> log_tests.py 17: test 'svn log -g' on a branching revision
> merge_tests.py 61: merge fails if subtree is deleted on src
> merge_tests.py 77: subtrees added after start of merge range are ok
> blame_tests.py 10: test 'svn blame -g'
In IRC, we've come to the conclusion that these should be ignored on
the reintegrate branch for now. The log and blame failures are
"expected", in some sense (that code is in flux anyway; in the log
case, the reintegrate branch exposed some latent bugs, in the blame
case a similar thing has probably happened even though hwright didn't
say that exactly in our conversation).
Meanwhile, the two merge tests are related to:
http://svn.haxx.se/dev/archive-2008-01/0008.shtml
Thanks to hwright, glasser, and pburba for the help.
-Karl
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
RE: failing tests on reintegrate branch
Posted by Paul Burba <pb...@collab.net>.
> -----Original Message-----
> From: Karl Fogel [mailto:kfogel@red-bean.com]
> Sent: Wednesday, January 16, 2008 5:08 PM
> To: dev@subversion.tigris.org
> Subject: failing tests on reintegrate branch
>
> Posting this while I work on it:
>
> Over all RA layers, these four tests fail in the head of the
> reintegrate branch right now:
>
> log_tests.py 17: test 'svn log -g' on a branching revision
> merge_tests.py 61: merge fails if subtree is deleted on src
> merge_tests.py 77: subtrees added after start of merge range are ok
Already mentioned this to Karl in IRC, but for the benefit of any other
reintegrate folks, the reasons for the failure of merge_tests.py 77 can
be found in this thread.
See http://svn.haxx.se/dev/archive-2008-01/0008.shtml.
There is a patch there to fix it too...
> blame_tests.py 10: test 'svn blame -g'
>
> I first looked at log_tests 17. That test makes a bunch of
> mergey changes, and then runs 'log -g', expecting that only
> r10 will show up in the result (in other words, no "Merged
> via: ..." lines). But in IRC just now, hwright said not to
> worry about log -g tests:
>
> <hwright> kfogel: I wouldn't worry too much about log test 17.
>
> <hwright> *all* the 'log -g' tests are having problems,
> which I hope
> to fix with some of the new goodies on the reintegrate
> branch.
>
> <kfogel> hwright: hmmmm
>
> <kfogel> the fact that reintegrate affects it *at all* is what
> bothers me
>
> <kfogel> I don't care whether the test is wrong or right right now
>
> <kfogel> what bugs me is that it's behaving differently
> on trunk vs
> reintegrate branch
>
> <kfogel> Is there some reason why this is a reasonable
> thing for it
> to do?
>
> <kfogel> Should I just XFail it and move on to the next
> of the four
> failing tests? (About to post to dev@.)
>
> <hwright> I'm not familiar enough with the guts of the reintegrate
> branch to comment on that. (<- reasonableness of test
> failure)
>
> <hwright> Yeah, I XFailing is probably appropriate.
>
> So, I'm going to move on to the merge_tests. But in case
> anyone's curious, below are some interesting things about
> log_tests 17.
>
> In the output below, the command 'ts' is a pristine trunk
> Subversion build (appx r28940), and 'rs' is a pristine
> reintegrate branch build.
> In order, I invoke 'log -g -r10' like this:
>
> 1. using trunk svn, on the test data produced by trunk svn
> 2. using trunk svn, on the test data produced by reintegrate svn
> 3. using reintegrate svn, on the test data produced by trunk svn
> 4. using reintegrate svn, on the test data produced by
> reintegrate svn
>
> Note how only (4) produces the bad output (the output that
> causes the test to fail):
>
> 1. $ ts log -g -r10
> /home/kfogel/src/subversion/subversion/tests/cmdline/svn-test-
> work/working_copies/log_tests-17/branches/b
> --------------------------------------------------------------
> ----------
> r10 | jrandom | 2008-01-16 16:30:08 -0500 (Wed, 16 Jan 2008) | 1 line
>
> Create branches/b from trunk
> --------------------------------------------------------------
> ----------
>
> 2. $ ts log -g -r10
> /home/kfogel/src/reintegrate/subversion/tests/cmdline/svn-test
> -work/working_copies/log_tests-17/branches/b
> --------------------------------------------------------------
> ----------
> r10 | jrandom | 2008-01-16 16:46:21 -0500 (Wed, 16 Jan 2008) | 1 line
>
> Create branches/b from trunk
> --------------------------------------------------------------
> ----------
>
> 3. $ rs log -g -r10
> /home/kfogel/src/subversion/subversion/tests/cmdline/svn-test-
> work/working_copies/log_tests-17/branches/b
> --------------------------------------------------------------
> ----------
> r10 | jrandom | 2008-01-16 16:30:08 -0500 (Wed, 16 Jan 2008) | 1 line
>
> Create branches/b from trunk
> --------------------------------------------------------------
> ----------
>
> 4. $ rs log -g -r10
> /home/kfogel/src/reintegrate/subversion/tests/cmdline/svn-test
> -work/working_copies/log_tests-17/branches/b
> --------------------------------------------------------------
> ----------
> r10 | jrandom | 2008-01-16 16:46:21 -0500 (Wed, 16 Jan 2008) | 1 line
>
> Create branches/b from trunk
> --------------------------------------------------------------
> ----------
> r7 | jrandom | 2008-01-16 16:46:21 -0500 (Wed, 16 Jan 2008) |
> 1 line Merged via: r10
>
> Add omicron to branches/a. It will be blocked from merging in r8.
> --------------------------------------------------------------
> ----------
> r4 | jrandom | 2008-01-16 16:46:20 -0500 (Wed, 16 Jan 2008) |
> 1 line Merged via: r10
>
> Add the file 'upsilon', and change some other files.
> --------------------------------------------------------------
> ----------
> r3 | jrandom | 2008-01-16 16:46:20 -0500 (Wed, 16 Jan 2008) |
> 1 line Merged via: r10
>
> Create branches/a from trunk.
> --------------------------------------------------------------
> ----------
> r2 | jrandom | 2008-01-16 16:46:20 -0500 (Wed, 16 Jan 2008) |
> 1 line Merged via: r10
>
> Import greek tree into trunk.
> --------------------------------------------------------------
> ----------
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
> For additional commands, e-mail: dev-help@subversion.tigris.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org