Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 38 Next »

[00:18:20 EDT(-0400)] * jayshao (n=jayshao@ool-44c59034.dyn.optonline.net) has joined ##uportal
[07:54:43 EDT(-0400)] * jayshao (n=jayshao@ool-44c59034.dyn.optonline.net) has joined ##uportal
[08:53:50 EDT(-0400)] * esm (n=esm@clue.mse.jhu.edu) has joined ##uportal
[09:41:51 EDT(-0400)] * jayshao (n=jayshao@wg-206-hill012.rutgers.edu) has joined ##uportal
[09:55:15 EDT(-0400)] * jayshao_ (n=jayshao@jshao.rutgers.edu) has joined ##uportal
[09:57:04 EDT(-0400)] * jayshao (n=jayshao@wg-206-hill012.rutgers.edu) has joined ##uportal
[11:11:20 EDT(-0400)] <EricDalquist> morning all
[11:13:20 EDT(-0400)] <esm> morning
[11:13:37 EDT(-0400)] <EricDalquist> so I think I am going to do a uP3 release before the dev meeting
[11:13:43 EDT(-0400)] <EricDalquist> but call it a milestone not RC
[11:13:56 EDT(-0400)] <EricDalquist> you have anything you want to get in before I do that today?
[11:15:57 EDT(-0400)] <esm> no i don't think so
[11:16:01 EDT(-0400)] <esm> are you goign to use the release plguin?
[11:16:19 EDT(-0400)] <EricDalquist> um, if you can show me how (smile)
[11:16:29 EDT(-0400)] <EricDalquist> one thing I need to do is fix up version numbers
[11:16:33 EDT(-0400)] <esm> right.
[11:16:38 EDT(-0400)] <esm> i saw that jira issue
[11:16:42 EDT(-0400)] <esm> sooooo.
[11:16:50 EDT(-0400)] <EricDalquist> I think that is all I need to do before releasing
[11:16:55 EDT(-0400)] <esm> i assume the milestone will get its own version?
[11:17:09 EDT(-0400)] <esm> 3.0.0-M1 or some such?
[11:17:16 EDT(-0400)] <EricDalquist> yeah
[11:17:35 EDT(-0400)] <EricDalquist> so actually I guess I need to do two things
[11:17:46 EDT(-0400)] <EricDalquist> 1. merge the refactor branch back to the trunk
[11:17:51 EDT(-0400)] <esm> right
[11:17:53 EDT(-0400)] <EricDalquist> 2. fix the version in the trunk
[11:19:07 EDT(-0400)] <EricDalquist> so ... off to do the merge
[11:19:09 EDT(-0400)] <esm> so the release plugin can be a bit of a headache (only because of the assemblies). It may be best to go over the use of the release plugin at the dev meeting
[11:19:12 EDT(-0400)] <esm> ok
[11:19:30 EDT(-0400)] <EricDalquist> ok
[11:19:40 EDT(-0400)] <EricDalquist> so this will be called 3.0.0-M4
[11:19:40 EDT(-0400)] <esm> just remember when you fix versions to update each pom's parent element.
[11:19:51 EDT(-0400)] <EricDalquist> ok
[11:20:15 EDT(-0400)] <esm> with the release will you also deploy maven 2 artifacts to the repo?
[11:20:35 EDT(-0400)] <esm> i don't think we have to per se
[11:20:54 EDT(-0400)] <esm> (who would be depending on -M4 artifacts?)
[11:21:48 EDT(-0400)] <esm> you could release -M4 as a binary zip and then we can use the release plugin at the dev meeting to release M5 which would automatically push the artifacts out.
[11:22:15 EDT(-0400)] <EricDalquist> hrm
[11:22:31 EDT(-0400)] <EricDalquist> doesn't mvn deploy push the artifacts to the repo?
[11:22:39 EDT(-0400)] <esm> yes it does
[11:22:49 EDT(-0400)] <EricDalquist> I was planning on doing that
[11:23:00 EDT(-0400)] <esm> k no problem that will work
[11:23:13 EDT(-0400)] <esm> final question (smile)
[11:23:25 EDT(-0400)] <esm> what will the next version number be : 3.0.0-SNAPSHOT?
[11:23:37 EDT(-0400)] <EricDalquist> um I'm not sure
[11:23:51 EDT(-0400)] <EricDalquist> 3.0.0-SNAPSHOT or 3.0.0-M5-SNAPSHOT
[11:23:57 EDT(-0400)] <EricDalquist> what do you think of those?
[11:24:16 EDT(-0400)] <esm> It probably doesn't make too much of a difference, so I'm cool with either.
[11:24:25 EDT(-0400)] <esm> it might make a difference if we were publishing snapshots
[11:25:00 EDT(-0400)] <EricDalquist> ah
[11:51:10 EDT(-0400)] <EricDalquist> grr need to find the revision of the branch again
[11:51:19 EDT(-0400)] <EricDalquist> they annoying bit of merging with svn
[11:52:59 EDT(-0400)] <esm> yeah
[11:53:27 EDT(-0400)] <EricDalquist> ok, merge is going
[12:02:26 EDT(-0400)] * jayshao (n=jayshao@wg-206-hill012.rutgers.edu) has joined ##uportal
[12:19:18 EDT(-0400)] <EricDalquist> crap
[12:19:42 EDT(-0400)] <EricDalquist> so
[12:19:53 EDT(-0400)] <EricDalquist> since I had done some trunk > refactor merges
[12:20:04 EDT(-0400)] <EricDalquist> I need to figure out what the last rev of those were
[12:20:07 EDT(-0400)] <EricDalquist> and use that rev
[12:20:18 EDT(-0400)] <EricDalquist> I tried the branch rev and got a bunch of conflicts
[12:20:33 EDT(-0400)] <EricDalquist> even though the conflicting files are the same in trunk and refactor
[12:25:49 EDT(-0400)] <esm> thats annoying...
[12:25:55 EDT(-0400)] <EricDalquist> very
[12:29:19 EDT(-0400)] <esm> hmm that could really complicate things
[12:29:37 EDT(-0400)] <EricDalquist> I think I just need to use the latest rev from trunk as my base
[12:29:39 EDT(-0400)] <esm> you may end up having to do multiple merges
[12:31:01 EDT(-0400)] <EricDalquist> we'll see
[12:31:08 EDT(-0400)] <EricDalquist> I'm reverting my checked out trunk
[12:31:15 EDT(-0400)] <EricDalquist> once that is done I'll give it another try
[12:32:15 EDT(-0400)] <esm> if the branch contains all of the changes in trunk since the branch was created, you might be able to update trunk to the revision # when the branch was created
[12:32:27 EDT(-0400)] <esm> and merge against that
[12:32:35 EDT(-0400)] <EricDalquist> ah
[12:33:04 EDT(-0400)] <esm> but if you have to cherry-pick stuff...
[12:33:51 EDT(-0400)] <esm> so you update a WC of trunk to the rev # when the branch was created
[12:33:52 EDT(-0400)] <esm> merge
[12:33:54 EDT(-0400)] <esm> update
[12:33:55 EDT(-0400)] <esm> commit
[12:34:17 EDT(-0400)] <EricDalquist> hrm
[12:37:52 EDT(-0400)] <esm> the update step will be needed b/c svn won't let you commit on an old rev #
[12:38:08 EDT(-0400)] <esm> hopefully you won't see the same conflicts
[12:38:08 EDT(-0400)] <EricDalquist> ah
[12:38:12 EDT(-0400)] <EricDalquist> yeah
[12:46:16 EDT(-0400)] * jayshao_ (n=jayshao@jshao.rutgers.edu) has joined ##uportal
[12:48:34 EDT(-0400)] * jayshao_ (n=jayshao@jshao.rutgers.edu) has joined ##uportal
[12:49:57 EDT(-0400)] <EricDalquist> ok ... revert done (that took a long time)
[12:50:04 EDT(-0400)] <EricDalquist> doing the switch to rev 11913 now
[12:58:50 EDT(-0400)] <esm> ok i've updated to 11913 and performed the merge in my WC
[12:59:12 EDT(-0400)] <esm> (working copy not water closet, tho I have been known to do computer work in the water closet)
[12:59:22 EDT(-0400)] <EricDalquist> lol
[12:59:45 EDT(-0400)] <EricDalquist> ok
[13:00:03 EDT(-0400)] <EricDalquist> well if you get it done before me
[13:00:07 EDT(-0400)] <EricDalquist> that would be great
[13:00:17 EDT(-0400)] <EricDalquist> I'm having issues of things being very slow on my laptop right now:/
[13:01:20 EDT(-0400)] <esm> hm
[13:01:53 EDT(-0400)] <esm> here comes a paste
[13:01:55 EDT(-0400)] <esm> one sec
[13:02:05 EDT(-0400)] <EricDalquist> k
[13:02:45 EDT(-0400)] <esm> http://rafb.net/p/zD0FI953.html
[13:03:16 EDT(-0400)] <EricDalquist> hrm
[13:03:25 EDT(-0400)] <esm> thats a new one for me
[13:03:30 EDT(-0400)] <esm> R == replaced
[13:03:35 EDT(-0400)] <esm> + means with history
[13:03:47 EDT(-0400)] <esm> so i'm not sure what the dilly is
[13:04:13 EDT(-0400)] <EricDalquist> well I did work in trunk (added/removed/modified) and merged it into refactor
[13:05:17 EDT(-0400)] <esm> well, i don't know why i get the error on line 15
[13:05:37 EDT(-0400)] <esm> and the message on line 16 seems to be a lie. the file is in the repository, at least at HEAD
[13:05:49 EDT(-0400)] <esm> it may not exist in r11913
[13:06:06 EDT(-0400)] <esm> but then I don't know why it gets a status code of 'R'
[13:06:49 EDT(-0400)] <EricDalquist> :/
[13:07:37 EDT(-0400)] <EricDalquist> I don't think it did exist in 11913
[13:07:47 EDT(-0400)] <esm> ok
[13:08:08 EDT(-0400)] <esm> http://rafb.net/p/0uX3Cx57.html
[13:08:38 EDT(-0400)] <esm> it seems to be kind of like an add. Add was what I was expecting, not a replace.
[13:09:00 EDT(-0400)] <esm> so i coudl just try to commit
[13:09:34 EDT(-0400)] <EricDalquist> everything else look good?
[13:09:46 EDT(-0400)] <esm> do you want to see my svn st output?
[13:10:04 EDT(-0400)] <esm> it seems to look good
[13:10:15 EDT(-0400)] <esm> it has the uportal3-war module rename
[13:10:25 EDT(-0400)] <esm> it has assemblies
[13:10:26 EDT(-0400)] <EricDalquist> cool, well I say go for it
[13:10:38 EDT(-0400)] <esm> ok here goes....
[13:12:25 EDT(-0400)] <esm> Boom. http://rafb.net/p/AI2aHT96.html
[13:13:46 EDT(-0400)] <esm> hmm somthing is up
[13:13:56 EDT(-0400)] <esm> i keep on getting svn: REPORT request failed on '/svn/!svn/vcc/default'
[13:13:56 EDT(-0400)] <esm> svn: REPORT of '/svn/!svn/vcc/default': 200 OK (https://developer.ja-sig.org)
[13:14:04 EDT(-0400)] <esm> imma revert
[13:14:09 EDT(-0400)] <EricDalquist> hrm ok
[13:14:42 EDT(-0400)] <EricDalquist> I'm do a dry run of the merge in eclipse right now
[13:15:46 EDT(-0400)] <EricDalquist> I'm doing*
[13:18:04 EDT(-0400)] <esm> which commits on the branch were merges from trunk?
[13:19:02 EDT(-0400)] <esm> oh nvm i think i can tell from svn log -v
[13:19:05 EDT(-0400)] <EricDalquist> I think I killed eclipse ...
[13:19:21 EDT(-0400)] * esm observes moment of silence
[13:20:01 EDT(-0400)] <esm> er no i can't
[13:20:07 EDT(-0400)] <EricDalquist> :/
[13:20:16 EDT(-0400)] <EricDalquist> I hope I didn't really hose this
[13:20:47 EDT(-0400)] <esm> nah
[13:21:47 EDT(-0400)] <esm> worse case senario is that we apply a merge per commit
[13:22:05 EDT(-0400)] <esm> per branch commit
[13:22:06 EDT(-0400)] <EricDalquist> uhg
[13:22:51 EDT(-0400)] <esm> well we can take some of them as a whole unit.
[13:23:00 EDT(-0400)] <esm> r11943 - HEAD should apply fine.
[13:23:18 EDT(-0400)] <esm> that is r11943-HEAD of the branch should merge fine with trunk
[13:23:54 EDT(-0400)] <EricDalquist> ok ... I'm going to try one more big merge
[13:24:13 EDT(-0400)] <esm> k
[13:24:25 EDT(-0400)] <esm> another opt
[13:24:30 EDT(-0400)] <esm> is to do a URL to URL merge
[13:24:59 EDT(-0400)] <esm> i don't know if that will help or not. it takes the WC out of the equation.
[13:25:00 EDT(-0400)] <esm> but
[13:26:09 EDT(-0400)] <esm> so workign back from head on the branch
[13:26:24 EDT(-0400)] <esm> r11938 on the branch was a merge from trunk
[13:27:05 EDT(-0400)] <EricDalquist> yup
[13:27:49 EDT(-0400)] <EricDalquist> I'm trying a 'merge -rHEAD:11937 '
[13:27:57 EDT(-0400)] <EricDalquist> so instead of the branch creation rev
[13:28:17 EDT(-0400)] <EricDalquist> that is the rev of the last rev that was merged into the refactor branch
[13:29:20 EDT(-0400)] <esm> are those revisions in the correct order?
[13:29:57 EDT(-0400)] <EricDalquist> I think so
[13:30:32 EDT(-0400)] <esm> i think they may be reversed
[13:30:39 EDT(-0400)] <EricDalquist> hrm oop
[13:30:40 EDT(-0400)] <EricDalquist> oops
[13:30:44 EDT(-0400)] <esm> svn merge -r OLD:NEW
[13:30:47 EDT(-0400)] <esm> so
[13:31:28 EDT(-0400)] <esm> e.g. doing a merge is almost, but not quite the same as creating a diff and applying it
[13:31:31 EDT(-0400)] <esm> so you can do
[13:31:54 EDT(-0400)] <esm> svn diff -r OLD:NEW http://path/to/branch to see the diff that will be applied to trunk
[13:32:43 EDT(-0400)] <esm> and kind of get a sanity check on what the merge will be trying to apply
[13:33:23 EDT(-0400)] <esm> http://rafb.net/p/zmnT1W23.html
[13:34:27 EDT(-0400)] <esm> i'm running the merge on trunk now, using the HEAD of trunk and things are progressing smoothly so far
[13:35:00 EDT(-0400)] <EricDalquist> with that 11937 rev?
[13:36:15 EDT(-0400)] <esm> no with r11913
[13:36:27 EDT(-0400)] <EricDalquist> ah ok
[13:37:12 EDT(-0400)] <esm> there are some conflicts though. which is really really annoying
[13:37:27 EDT(-0400)] <EricDalquist> yeah
[13:37:30 EDT(-0400)] <esm> i only see a couple though so far
[13:37:42 EDT(-0400)] <esm> on .springBeans, README, build.xml
[13:37:50 EDT(-0400)] <EricDalquist> there are a bunch in the source
[13:38:40 EDT(-0400)] <esm> ok. i'm not there yet i guess.
[13:38:42 EDT(-0400)] <esm> woops yeah
[13:38:51 EDT(-0400)] <esm> uportal3-impl/src/main/java/org/jasig/portal/portlet/registry/event/RegistryEvent.java
[13:38:55 EDT(-0400)] <esm> WTF
[13:39:42 EDT(-0400)] <esm> i've counted 13 or so conflicts so far
[13:39:45 EDT(-0400)] <esm> 14
[13:43:29 EDT(-0400)] * jayshao (n=jayshao@wg-206-hill012.rutgers.edu) has joined ##uportal
[13:44:36 EDT(-0400)] <esm> http://rafb.net/p/mmhLDl78.html
[13:44:40 EDT(-0400)] <esm> 21 conflicts
[13:44:44 EDT(-0400)] <esm> pfffft
[13:47:50 EDT(-0400)] <esm> http://svnbook.red-bean.com/nightly/en/svn.branchmerge.copychanges.html#svn.branchmerge.copychanges.bestprac
[13:47:59 EDT(-0400)] <esm> Unfortunately, Subversion is not such a system; it does not yet record any information about merge operations. [23] When you commit local modifications, the repository has no idea whether those changes came from running svn merge, or from just hand-editing the files.
[13:48:15 EDT(-0400)] <EricDalquist> ah
[13:50:20 EDT(-0400)] <esm> Scroll down to the "Merge Conflicts"; it is informative also
[13:50:41 EDT(-0400)] <esm> esp the example regarding "Skipped missing target"
[13:53:12 EDT(-0400)] <EricDalquist> interesting
[13:53:22 EDT(-0400)] <esm> yeah that chapter is worth reading.
[13:55:01 EDT(-0400)] <esm> so anyway i think its just as easy to perform the big merge, resolve the conflicts, run the tests then commit. what do you think
[13:55:10 EDT(-0400)] <EricDalquist> yeah
[13:55:25 EDT(-0400)] <EricDalquist> I'm re-trying my r11937 merge in the right order this time
[13:55:30 EDT(-0400)] <esm> ok
[13:55:33 EDT(-0400)] <EricDalquist> since that may resovle the conflicts
[13:55:33 EDT(-0400)] <esm> cool
[13:57:44 EDT(-0400)] <esm> http://subversion.tigris.org/roadmap.html
[13:57:50 EDT(-0400)] <esm> Medium-term Goals:
[13:57:50 EDT(-0400)] <esm> *
[13:57:50 EDT(-0400)] <esm> Merge tracking (tracking of merge history coming in 1.5.0, see info)
[13:57:56 EDT(-0400)] <esm> first item in the roadmap (smile)
[13:58:06 EDT(-0400)] <EricDalquist> cool
[13:58:13 EDT(-0400)] <esm> Sparsely populated checkouts (full API and basic command-line support coming in 1.5)
[13:58:16 EDT(-0400)] <esm> THANK YOU
[13:58:25 EDT(-0400)] <EricDalquist> ?
[13:58:26 EDT(-0400)] <esm> if that means what I think that means
[13:59:01 EDT(-0400)] <esm> so normally if you checkout http://repo/path/to/the/project it just checks out "project"
[13:59:32 EDT(-0400)] <esm> but sometimes you want to check out the entire hierarchy, "/path/to/the/project" instead of just "/project"
[13:59:58 EDT(-0400)] <esm> right now that means you have to do a full check out of everything under "/path" which of course on a large repo, sucks.
[14:00:17 EDT(-0400)] <EricDalquist> ah
[14:01:25 EDT(-0400)] <esm> apparently there is a contrib called 'svnmerge' which is not part of svn itself but can be used to help with merges
[14:02:55 EDT(-0400)] <EricDalquist> well I'm seeing the same conflicts with this merge
[14:03:02 EDT(-0400)] <EricDalquist> is there a way to just list conflicts?
[14:03:10 EDT(-0400)] <esm> http://www.orcaware.com/svn/wiki/Svnmerge.py
[14:03:14 EDT(-0400)] <esm> hum
[14:03:18 EDT(-0400)] <esm> i use egrep (smile)
[14:03:49 EDT(-0400)] <esm> eg svn st | egrep '^C'
[14:04:04 EDT(-0400)] <esm> if you are on windows, you migth be able to use "find"
[14:04:10 EDT(-0400)] <EricDalquist> ok
[14:04:39 EDT(-0400)] <esm> but i don't see it in any svn commands
[14:05:15 EDT(-0400)] <esm> i'm on version 1.3.1 of the client 1.4.3 may have that option
[14:06:10 EDT(-0400)] <esm> It looks like I need to add svnmerge.py into my arsenal er i mean toolbox
[14:06:26 EDT(-0400)] <esm> ...
[14:06:27 EDT(-0400)] <esm> Bidirectional merges: merges changes forth and back between a branch and its head.
[14:06:29 EDT(-0400)] <esm> ...
[14:08:56 EDT(-0400)] <esm> wow
[14:08:58 EDT(-0400)] <esm> have you seen this
[14:08:59 EDT(-0400)] <esm> http://www.justin.tv/
[14:09:29 EDT(-0400)] <EricDalquist> oh yeah
[14:09:33 EDT(-0400)] <EricDalquist> weird
[14:09:35 EDT(-0400)] <EricDalquist> ok
[14:09:38 EDT(-0400)] <EricDalquist> so svn st
[14:09:45 EDT(-0400)] <EricDalquist> shows me lots of A/D/M
[14:09:47 EDT(-0400)] <EricDalquist> and some ?
[14:09:50 EDT(-0400)] <EricDalquist> no C though
[14:10:07 EDT(-0400)] <EricDalquist> the ? are all .working, .merge-left and .merge-right files
[14:11:05 EDT(-0400)] <esm> ok what version of the client do you have
[14:11:11 EDT(-0400)] <EricDalquist> 1.4.0
[14:11:45 EDT(-0400)] <esm> ok so that should be fine. this is from r11937?
[14:12:37 EDT(-0400)] <esm> you will still need to merge the changes from the branch inception up to r11937 at some point though
[14:13:40 EDT(-0400)] <EricDalquist> ?
[14:13:51 EDT(-0400)] <EricDalquist> they are already in both trunk and the branch
[14:14:09 EDT(-0400)] <EricDalquist> and it looks like the .merge-right files are what I want
[14:19:00 EDT(-0400)] <esm> well
[14:19:30 EDT(-0400)] <esm> so do an 'svn log -r 11913:11937' on the branch
[14:19:36 EDT(-0400)] <EricDalquist> I'm just going to batch rename and rm some files
[14:20:02 EDT(-0400)] <EricDalquist> what will that give me?
[14:20:43 EDT(-0400)] <esm> Ok so the maven-refactor branch goes from 11913 - HEAD, which is now 11954
[14:21:03 EDT(-0400)] <esm> you've merged 11937-HEAD back to trunk right?
[14:21:24 EDT(-0400)] <esm> so that still leaves 11913-11936 that hasn't been merged back to trunk
[14:21:30 EDT(-0400)] <EricDalquist> yup
[14:21:43 EDT(-0400)] <EricDalquist> except all of those changes are already sync'd I believe
[14:21:57 EDT(-0400)] <EricDalquist> oh crap
[14:21:58 EDT(-0400)] <EricDalquist> no they aren't
[14:22:05 EDT(-0400)] <EricDalquist> man my head is just not with this today
[14:22:16 EDT(-0400)] <esm> that is ok (smile)
[14:24:05 EDT(-0400)] <EricDalquist> so what do I need to do about 11913-11936
[14:24:16 EDT(-0400)] <EricDalquist> or do I really just need to do the merge from 11913
[14:25:38 EDT(-0400)] <esm> well there are changes there I think you want
[14:26:06 EDT(-0400)] <esm> all of the commits by me were only done on the branch
[14:26:26 EDT(-0400)] <EricDalquist> yeah
[14:26:44 EDT(-0400)] <esm> a portion of those commits were with objectID's, which you've fixed in later revisions, so those may be able to be left out
[14:27:40 EDT(-0400)] <esm> r11926, r11928 are two that need to be merged
[14:28:32 EDT(-0400)] <EricDalquist> since I know I'm taking everything from the 'right' of the merge I might just re-do it from 11913 and just run my little script again
[14:28:58 EDT(-0400)] <esm> ok
[14:29:42 EDT(-0400)] <esm> one way or the other I think you'll have to manually resolve conflicts
[14:29:52 EDT(-0400)] <EricDalquist> yeah
[14:29:52 EDT(-0400)] <esm> i need to run to a meeting but i'll be back on.
[14:30:01 EDT(-0400)] <EricDalquist> luckily the conflicts are pretty simple
[14:30:05 EDT(-0400)] <esm> yeah
[14:30:12 EDT(-0400)] <EricDalquist> and the resolving is just taking the version from the branch and using it
[14:30:16 EDT(-0400)] <EricDalquist> no actualy file editing
[14:30:24 EDT(-0400)] <esm> right
[14:30:35 EDT(-0400)] <esm> you could try this with svnmerge.py
[14:30:46 EDT(-0400)] <esm> if you had the time or energy
[14:30:55 EDT(-0400)] <EricDalquist> or python
[14:30:56 EDT(-0400)] <EricDalquist> (smile)
[14:30:58 EDT(-0400)] <esm> ahahah
[14:30:59 EDT(-0400)] <esm> yeah
[14:31:08 EDT(-0400)] <esm> brb
[14:31:11 EDT(-0400)] <EricDalquist> k

  • No labels