Camelia, the Perl 6 bug

IRC log for #cdk, 2012-03-22

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

All times shown according to UTC.

Time Nick Message
01:26 slyrus joined #cdk
03:35 Conrad joined #cdk
06:05 egonw joined #cdk
06:24 sneumann joined #cdk
07:22 konditorn joined #cdk
07:52 Conrad joined #cdk
08:15 Gpox joined #cdk
08:30 egonw joined #cdk
08:32 egonw joined #cdk
08:34 egonw Gpox: ping
08:38 zaetnick joined #cdk
08:41 egonw Gpox: can I ask you to review two patches for renderbasic?
08:42 egonw they are small, and mostly around testing, not functionality...
08:42 egonw so, not complex code changes that need checking
08:48 egonw https://github.com/egonw/cdk​/tree/380-14x-rendercleanup
08:48 egonw two patches, in fact
08:51 jonalv joined #cdk
08:51 egonw hi jonalv
08:51 jonalv hi egonw
09:04 egonw konditorn: please do ping me if you need help with setting up a CDK development environment
09:05 egonw and afterwards, so that I can talk you through the unit tests for the DBST
10:12 jbrefort joined #cdk
10:47 egonw Gpox: the other things is, you did see my renderextra branch on github?
10:53 jbrefort joined #cdk
12:59 jbrefort joined #cdk
12:59 konditorn egonw: I think I've set up the workspace for CDK right, but I've one error: Project 'org.openscience.cdk' is missing required source folder: 'build'
13:30 konditorn egonw: ping
14:35 egonw pong
14:35 egonw back from a meeting
14:35 egonw yes, that's why you need to run 'ant clean runDoclet' from the command line
14:36 egonw (or from within Eclipse, with the Ant view)
14:40 konditorn from what director should I run it?
14:41 egonw cdk/
14:41 egonw the root folder, where the build.xml file os
14:42 egonw is
14:48 konditorn fixed, after refreshing the workspace as well…=)
14:49 egonw good, good
14:59 konditorn so… whats the next step? continue to read your book and familiarize me with the code?
15:15 zaetnick_ joined #cdk
15:25 egonw so, strategic discussion here :)
15:25 egonw konditorn:
15:25 egonw all unit tests are found in src/test/
15:25 konditorn egonw:  yes
15:25 egonw the test class .java name follows the pattern as the class they test
15:26 egonw so: org.openscience.cdk.smiles.DeduceBondSystemTool
15:26 egonw and
15:26 egonw org.openscience.cdk.smiles​.DeduceBondSystemToolTest
15:26 egonw just in a different folder
15:26 egonw (applied 98% of the time :)
15:26 egonw I guess you use Eclipse?
15:27 konditorn yes, I do
15:29 egonw ok, so you can just right click on the test class name, and Run as -> JUnit test
15:29 egonw (not the plugin flavor)
15:31 konditorn mm, it say something about 0 error and 1 failure...
15:31 egonw good!
15:31 egonw you can compare your results with the "current" state on our Nightly service
15:32 egonw there is indeed 1 known failure
15:32 egonw check this website: http://pele.farmbio.uu.se/supernightly/
15:32 egonw did you change the branch to cdk-1.4.x?
15:32 egonw because I like to aim for a tool for that branch, as Bioclipse uses that branch
15:32 egonw (it will be part of master automatically too then)
15:33 egonw but forward porting is easier than backwards
15:33 konditorn no, but soon…
15:33 egonw good :)
15:33 egonw branching you know already anyway :)
15:33 egonw so, SUpernightly aggregates some results from various Nightly runs
15:33 egonw somewhere in the middle are the CDK versions listed
15:34 egonw the ones with ".git" indicate current development branches
15:34 egonw those without are fixed released versions
15:34 egonw following some links, you can get to this page:
15:34 egonw http://pele.farmbio.uu.se/nightl​y-1.4.x/test/result-smiles.html
15:34 egonw there you can see that xtestQuinone() is known to fail
15:35 egonw and that you can use as reference
15:37 konditorn ok...
15:37 egonw Gpox++ (for the isBridge() pointer! looks interesting!)
15:48 konditorn mm, two of the constructor used in the test are marked as deprecated…
15:54 egonw it is old code :)
15:54 egonw seriously... the CDK is a very large code base, and maintenance is a problem
15:55 konditorn I think I've heard something about that… :)
15:58 egonw :)
15:58 egonw I have to catch my train now
15:59 egonw please collect all questions, and email
15:59 egonw I'm online tomorrow too, and have few scheduled meetings :)
15:59 egonw ttl
15:59 konditorn yes, I am.
16:00 egonw ok!
16:00 egonw maybe we can talk about SMILES and MDL molfiles then :)
16:00 konditorn yes. =)
16:17 konditorn left #cdk
17:48 zaetnick joined #cdk
18:31 egonw joined #cdk
18:40 jonalv joined #cdk
18:40 CIA-121 cdk: Egon Willighagen cdk-1.4.x * r32d58d6 / src/test/org/openscience/cdk/cove​rage/CoverageAnnotationTest.java : (log message trimmed)
18:40 CIA-121 cdk: Bridge methods are methods introduced by compilers in relation to methods that use generics. Various renderer related method trigger test methods for just bridge methods, but are never really implemented.
18:40 CIA-121 cdk: For example, the below interface code causes the coverage test before
18:40 CIA-121 cdk: this patch to expect the bridge generate(IChemObject object, RendererModel model)
18:40 CIA-121 cdk: method to be tested:
18:40 CIA-121 cdk: public interface IGenerator<T extends IChemObject> {
18:40 CIA-121 cdk:  public IRenderingElement generate(T object, RendererModel model);
18:42 egonw that's arvid's tip
18:51 CIA-121 cdk: Egon Willighagen cdk-1.4.x * r47d65be / src/main/org/openscience/cdk/sm​iles/DeduceBondSystemTool.java :
18:51 CIA-121 cdk: Split up into ring systems before running the AllRingsFinder to speed up the algorithm (patch by Kevin Lawson)
18:51 CIA-121 cdk: Change-Id: I03bca9e4803f88217271e6e4acb440405f6aafe8
18:51 CIA-121 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://git.io/4AQqdg
18:51 CIA-121 cdk: Egon Willighagen cdk-1.4.x * rd9c147d / src/test/org/openscience/cdk/smil​es/DeduceBondSystemToolTest.java :
18:51 CIA-121 cdk: Unit test with the SMILES reported by Kevin Lawson
18:51 CIA-121 cdk: Change-Id: I2a3c6cfec082a82ce48434ab0120ae4ac8fd2bd6
18:51 CIA-121 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://git.io/dWPEVA
18:51 CIA-121 cdk: Egon Willighagen cdk-1.4.x * r596258a / src/test/org/openscience/cdk/smil​es/DeduceBondSystemToolTest.java :
18:51 CIA-121 cdk: The Bioclipse test case of failing bond order assignment
18:51 CIA-121 cdk: Change-Id: I86f9bc709bfd3b3e1a3b816914e2dc32bbad6df0
18:51 CIA-121 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://git.io/FflpOQ
18:52 CIA-121 cdk: Egon Willighagen cdk-1.4.x * re9c3f4b / AUTHORS :
18:52 CIA-121 cdk: Added Kevin Lawson to the list of AUTHORS
18:52 CIA-121 cdk: Helper methods now return results, rather than change parameter content, and uses interfaces as variable types rather than implementations (patch by Kevin Lawson).
18:52 CIA-121 cdk: Change-Id: Iba28472b2e1837e44667ffb14aa44f523c3831fc
18:52 CIA-121 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://git.io/OiC1rQ
18:55 slyrus joined #cdk
19:02 egonw hi slyrus
19:02 slyrus hey egonw!
19:03 slyrus I'm trying to get my old structure drawing code-using example to play nice with a newish cdk...
19:03 egonw ah, yes, a challenge :)
19:03 slyrus so far no good
19:05 egonw what errors do you get?
19:09 slyrus error: java.lang.RuntimeException: java.lang.IllegalAccessError: You requested the active parameter of type org.openscience.cdk.renderer.gene​rators.BasicSceneGenerator$Scale, but it has not been registered yet. Did you make sure the IGeneratorParameter is registered, by registring the appropriate IGenerator? Alternatively, you can use getDefault() to query the default value for any parameter...
19:09 slyrus ...on the classpath.
19:09 egonw ah, did you add the BasicSceneGenerator?
19:09 egonw to your renderer instance?
19:11 slyrus nope
19:12 egonw this is why:
19:13 egonw those rendering parameters are provided by generators...
19:13 egonw the scale is provided by the scene generator
19:13 egonw that's all
19:13 egonw that's one down :)
19:14 slyrus indeed! thanks!
19:18 CIA-121 cdk: Egon Willighagen cdk-1.4.x * r8d6a349 / build.props :
19:18 CIA-121 cdk: Bumped version number, anticipating the release
19:18 CIA-121 cdk: Change-Id: I0f6ae07b9052bcc2ad2aac61b62010c120b0ab5f - http://git.io/pSptlQ
19:18 sneumann joined #cdk
19:19 CIA-121 cdk: Egon Willighagen cdk-1.4.x * r2e05c1a / build.props :
19:19 CIA-121 cdk: Open for more bug fixing
19:19 CIA-121 cdk: Change-Id: I2482db1e78fb65bb2a0a8050a86fad58afd1b639 - http://git.io/f6lFow
19:32 slyrus egonw: I see the SDG still doesn't get E-Z across double bonds right :(
19:32 egonw no, correct
19:32 egonw long standing issue
19:32 slyrus I was hoping that i I went away long enough, that would magically get fixed :)
19:32 egonw :)
19:32 egonw stupid thing is, it
19:33 egonw it's probably not too difficult to fix :(
19:33 egonw well, that is...
19:33 egonw we now have a data structure to properly store this data...
19:33 egonw actually, not even that is completely true...
19:33 egonw if only we have more people working on the foundations of the CDK, rather than applications :/
19:36 egonw but things do improve over time...
19:36 egonw just not as fast as all would like to :)

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