Camelia, the Perl 6 bug

IRC log for #cdk, 2008-10-20

| Channels | #cdk index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:37 CIA-36 cdk: rajarshi * r12738 /cdk/branches/cdk-1.2.x/src/ (3 files in 3 dirs): Updated to add covalent radii information to the periodic table class. Data taken from BODR
01:28 CIA-36 cdk: rajarshi * r12739 /cdk/branches/cdk-1.2.x/src/main/org/o​penscience/cdk/qsar/descriptors/molecu​lar/LengthOverBreadthDescriptor.java: Updated code to covalent radii from the periodic table - avoids the issue of using covalent radii from atom type information. Fixes bug #1862142
01:32 CIA-36 cdk: rajarshi * r12740 /cdk/branches/cdk-1.2.x/src/test/org/op​enscience/cdk/qsar/descriptors/molecula​r/LengthOverBreadthDescriptorTest.java: Updated unit tests, to use a lower level of accuracy
04:14 CIA-36 cdk: gilleain * r12741 /cdk/branches/gilleain-jchempaint/src/ (21 files in 6 dirs): Merge of goglepox's branch with mine to compare architectures.
05:27 bag_ joined #cdk
05:29 jbrefort joined #cdk
06:40 carsten joined #cdk
07:00 egonw joined #cdk
07:42 rojasm joined #cdk
08:11 jonalv joined #cdk
08:20 CIA-36 cdk: miguelrojasch * r12742 /cdk/branches/miguelrojasch-reacti​ons/src/test/org/openscience/cdk/r​eaction/ReactionProcessTest.java: Added test for ReactionProcess
08:26 maclean joined #cdk
08:42 maclean joined #cdk
08:56 rojasm hi egonw!
08:56 rojasm I have in my local project
08:56 rojasm $git branch
08:56 rojasm * git-miguelrojasch-reaction
08:56 rojasm master
08:56 rojasm which is my local branch of miguelrojasch-reaction. To merge to trunk I check all test if they are ok. And they are?
08:56 rojasm To merge should I do with the command:
08:56 rojasm $ git checkout master
08:56 rojasm $ git merge git-miguelrojasch-reaction
08:56 rojasm Or should I do first test other thing?
08:57 egonw first make sure there are no regressions (no new failing tests)
08:57 egonw then, you update your branch to make it in full synch with trunk:
08:57 egonw git svn fetch; git checkout master; git svn rebase
08:57 egonw then
08:57 egonw git checkout git-miguelrojasch-reaction; git rebase master
08:57 egonw then recompile and run the unit tests again
08:58 egonw this "git rebase master" makes your branch up to date with repsect to commits to trunk
08:58 egonw rojasm: that's the better way of doing: cd miguelrojash-reaction; svn merge trunk
08:59 egonw this "git rebase master" basically converts your patches into patches against the latest trunk version
08:59 egonw then, when you detect no new fails, you can do:
08:59 egonw git checkout master; git merge git-miguelrojasch-reaction
08:59 egonw and then
08:59 egonw git svn dcommit
09:05 rojasm ok thanks I try it
09:10 rojasm other question. Now I am working with eclipse to modify the files in svn and then I commit via svn. Do you know where is localized exactly the git branch miguelrojasch-rection for example?
09:10 rojasm may be in cdk/.git/svn/
09:11 egonw no idea
09:11 egonw if you use git-svn, you cannot commit from inside eclipse
09:12 rojasm but I mean where is the files which you are modifying?
09:12 rojasm At least I would like to import them as a project
09:17 rojasm strange I have a error when I didn't change any thing about this method :(
09:17 rojasm src/main/org/openscience/cdk/modeling/build​er3d/data/ringTemplateFingerprints.txt.gz: needs merge
09:17 rojasm error: you need to resolve your current index first
09:23 rojasm mmm
09:27 egonw I don't understand what you mean?
09:27 egonw I don't understand what you mean
09:27 egonw (no ?)
09:27 rojasm mmm
09:27 rojasm I was changing the files with eclipse by svn like we did always
09:28 rojasm but than I tried to have a git local branch to be able to merge afterwards
09:29 rojasm Because I was not able to find how to change the files in git directory
09:29 rojasm a bit all stupid
09:30 egonw ah, I have my git cdk/ folder as project in Eclipse
09:30 egonw so, if I change something in Eclipse, it changes in my git branch
09:30 egonw this does mean I have to do F5 in Eclipse each time I change branches, but that's fast enough
09:31 rojasm also the complet cdk project
09:31 rojasm ok
09:32 egonw no
09:32 egonw SVN cdk/trunk == git cdk/
09:32 egonw umm... yeah, the complete "cdk" project
09:32 egonw but not a checkout of all cdk projects :)
09:32 egonw hahaha
09:33 egonw each 'cdk project' from [0], I have as separate git project:
09:33 egonw 0. http://cdk.svn.sourceforge.net/viewvc/cdk/
09:33 zarah egonw's link is also http://tinyurl.com/6fqe5j
09:41 rojasm But I think doing that as you explain me, If I modify some thing in eclipse I will modify the cdk/trunk and not the branches. or am I wrong?
09:43 egonw depends on what you have openened in Eclipse
09:43 egonw right click on your CDK project in the Eclipse navigator
09:43 egonw and pick
09:43 egonw Properites
09:44 egonw then click in the left side "Resource"
09:44 egonw that will give you the "Location" of what you have opened
09:46 rojasm a /home/mrc/Development/Sourceforge/cdkDir/git/cdk which means for the trunk project
09:48 rojasm that is the problem I don't find where is the directory in git which contains the brunch projects :(
09:48 egonw depends...
09:48 egonw ah...
09:48 egonw but git only has one folder
09:48 egonw I assume git/cdk is your git checkout
09:48 egonw where you type "git branch"
09:48 egonw so to say
09:49 egonw so, everytime you type "git checkout someBranch"
09:49 egonw it will change to (visible) content of that folder
09:49 egonw (but it always remembers all commits to any branch)
09:53 rojasm yes my checkout is the git/cdk directory, yes where I type "git branch"
09:53 rojasm Ok I think I understand!
09:53 rojasm each time I modify something of the cdk/trunk and than I point to mybranch doing "git checkout mybranch" It will remember this change was doing for this branch.
09:54 rojasm what happens if I want to work with two different branches at the same time without checkout
09:54 rojasm anyway I need to read more about git
09:54 egonw clone
09:54 rojasm I bit confuse still
09:54 egonw cd git/cdk
09:55 egonw cd ..
09:55 egonw mkdir cdk-myBranch
09:55 egonw cd cdk-myBranch
09:55 egonw git init
09:55 egonw nope...
09:55 egonw from the start
09:55 egonw cd git/cdk
09:55 egonw cd ..
09:56 egonw git clone cdk cdk-clone
09:56 egonw (I think)
09:56 egonw add -l
09:56 egonw git clone -l cdk cdk-clone
09:57 egonw but remember, this clone can not use git-svn to commit things
09:57 egonw so, you might want to add --origin someBranch
09:58 egonw but I have not used this approach myself yet
09:58 egonw maybe ask jonalv
09:59 egonw who was interested in such a set up too
09:59 * jonalv is backlogging now...
10:00 jonalv I hve not done any further studying into that.
10:00 jonalv I talked a bit with one of the developers of the git plugin for Eclipse though
10:00 jonalv He suggested having a git repository for each pluign....:/
10:01 jonalv I am thinking use SVN for branching and daily work and git just for merging...
10:04 egonw jonalv: yes, not too bad an idea
10:04 egonw jonalv: but you can also consider the above set up
10:05 egonw use git for branching, using different git repository copies (git clone)
10:05 egonw and use git pull for merging
10:06 jonalv And if you want your branch in the svn repo?
10:09 egonw svn copy
10:09 egonw and keep a git-svn clone around
10:09 egonw so that you can do:
10:09 egonw git svn fetch; git checkout -b localNewBranch newUpstreamSVNBranch
10:10 egonw git pull ../../what/ever/git/clone/with/cool/new/feature master
10:10 egonw git svn dcommit
10:10 egonw well, sort of like that
10:10 egonw full version:
10:11 egonw svn copy https://bla/cdk/trunk https://bla/cdk/branches/newUpstreamSVNBranch
10:11 egonw cd git-svn/cdk
10:11 egonw git svn fetch
10:11 egonw git checkout trunk
10:11 egonw git svn rebase
10:11 egonw cd ..
10:12 egonw git clone -l cdk cdkMyLocalCopyGitDirForInEclipse
10:12 egonw cd cdkMyLocalCopyGitDirForInEclipse
10:12 egonw nano bla
10:13 egonw (or open this git copy folder in Eclipse, but mind tuning .project)
10:13 egonw gti commit
10:13 egonw # wait until you are happy for copying into the newUpstreamSVNBranch
10:13 egonw cd ../../git-svn/cdk
10:13 egonw git svn fetch; git checkout newUpstreamSVNBranch; git svn rebase
10:14 egonw git pull ../cdkMyLocalCopyGitDirForInEclipse
10:14 egonw git svn dcommit
10:14 egonw OK, that should do it :)
10:17 egonw http://postgenomic.com/story.php?cluster_id=26333
10:17 zarah egonw's link is also http://tinyurl.com/5wu864
10:22 rojasm egonw: Could I merge anyway using the commands
10:22 rojasm git checkout master; git merge git-miguelrojasch-reaction
10:22 rojasm (10:59:56 AM) Egon Willighagen: and then
10:22 rojasm (10:59:59 AM) Egon Willighagen: git svn dcommit
10:22 rojasm I have not conflicts
10:24 egonw rojasm: formally, you need to announce and describe your plugin
10:25 egonw so that it can be reviewer before it goes into trunk
10:25 egonw so that it can be reviewed before it goes into trunk
10:26 thomasku joined #cdk
10:32 maclean egonw : on that blog post about chemical editing, the link that says "wiki page" points to my blog..
10:32 egonw oops
10:32 maclean :)
10:35 egonw fixed
10:35 thomasku moin
10:35 egonw thanx for letting me know
10:35 zarah hi thomasku
10:35 egonw hi thomasku
10:35 maclean np.
10:36 egonw nasty spam: http://chem-bla-ics.blogspot.com/2008/10/g​ittodo-support-for-freemind-graphical.html
10:36 zarah egonw's link is also http://tinyurl.com/6hhn6g
11:34 shk3 joined #cdk
11:45 timvdm joined #cdk
12:01 maclean joined #cdk
12:51 CIA-36 cdk: egonw * r12743 /paper-jchempaint/ (6 files in 2 dirs): Added template for the JChemPaint 3.0 paper
12:53 malick left #cdk
13:36 CIA-36 cdk: shk3 * r12744 /cdk/branches/shk3-cleanuptests/ (5 files in 4 dirs): moved test for PartialTChargeMMFF94Descriptor to test-builder3d package (the class itself was in builder3d, the test in test-)qsartomic
14:05 CIA-36 cdk: rajarshi * r12745 /cdk/branches/cdk-1.2.x/src/main/or​g/openscience/cdk/tools/manipulator​/MolecularFormulaManipulator.java: Updated code so that it does not modify the input formula by configuring it's isotopes, but rather pulls isotope info from the factory on the fly
14:11 rojasm joined #cdk
14:21 CIA-36 cdk: goglepox * r12746 /cdk/branches/goglepox-jchempaint/sr​c/main/org/openscience/cdk/renderer/ (19 files in 3 dirs): Trying a new approach in generators
14:52 CIA-36 cdk: rajarshi * r12747 /cdk/trunk/src/test/org/openscience/cdk/tools/​manipulator/AtomContainerManipulatorTest.java: Added a test for bug #1969156
15:34 maclean joined #cdk
16:21 maclean left #cdk
16:31 thomasku joined #cdk
16:47 jbrefort joined #cdk
16:58 edrin joined #cdk
22:37 malic1 joined #cdk

| Channels | #cdk index | Today | | Search | Google Search | Plain-Text | summary