Camelia, the Perl 6 bug

IRC log for #cdk, 2009-03-06

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

All times shown according to UTC.

Time Nick Message
06:18 jbrefort joined #cdk
08:00 Gpox joined #cdk
08:00 Gpoks joined #cdk
08:01 Gpox joined #cdk
09:10 shk3 joined #cdk
09:37 CIA-62 cdk: shk3 * r14360 /cdk/branches/jchempaint-primary/src/​main/org/openscience/cdk/controller/ (3 files): some changes to undo/redo which become necessary when adopting it to bioclipse
10:20 jonalv joined #cdk
10:42 egonw joined #cdk
11:10 CIA-62 cdk: shk3 * r14361 /cdk/branches/jchempaint-primary/src/​main/org/openscience/cdk/controller/ (9 files): moved undo/redo to hub
11:15 CIA-62 cdk: shk3 * r14362 /jchempaint/trunk/ (6 files in 4 dirs): moved undo/redo to hub
11:34 CIA-62 cdk: shk3 * r14363 /cdk/branches/jchempaint-primar​y/src/main/org/openscience/cdk/ (controller/ControllerHub.java io/ReaderFactory.java): synchronized changes in the bc modules
11:37 CIA-62 cdk: shk3 * r14364 /cdk/branches/jchempaint-primary/src/main/​org/openscience/cdk/io/ReaderFactory.java: this should not get committed
11:46 carsten joined #cdk
13:18 CIA-62 cdk: egonw * r14365 /cdk/branches/cdk-1.2.x/src/test/org/o​penscience/cdk/io/MDLWriterTest.java: Fixed testing of the correct writer
13:44 egonw shk3: ping
13:55 shk3 hi
13:55 zarah ni hao shk3
13:56 egonw hej
13:56 egonw shk3: I'm writing a patch to clean up the code in 1.2 for writing MDL files, molfiles and SD files
13:56 egonw can you have a look at it later?
13:57 shk3 ye
13:57 shk3 s
13:57 egonw and a question about it right now
13:57 egonw I'm splitting the code up
13:58 egonw in a SDFWriter and a MDLWriter
13:58 egonw the latter for a single IMOlecule
13:58 egonw the former for SD files
13:58 shk3 aren't those separate?
13:58 egonw this makes the API more consistent, and the user can write these files without using dedicated method
13:58 egonw no, not yet
13:59 shk3 no, indeed
13:59 egonw the SD fields now should be set specifically...
13:59 egonw is it OK to use the IMOlecule.getProperties() instead?
13:59 egonw or would that break any of your use cases (beyond the obvious changes to make)?
13:59 egonw that is...
13:59 egonw are you using setSDFields() for things other than IMOelcule.getProperties()?
14:00 shk3 no, properties should be right
14:00 egonw ok, good
14:00 shk3 was wondering about this myself recently
14:00 shk3 why properties are not used
14:00 egonw no idea
14:00 shk3 just bad design
14:01 egonw I think I remember the author had a reason for it...
14:01 egonw but don't remember what
14:02 shk3 might be you need to clean properties before writing
14:02 shk3 and later set them again
14:03 shk3 or you get stuff you don't want in the sdf
14:03 egonw yes, indeed...
14:03 egonw I think that was the reason indeed
14:04 shk3 we could have a switch for using/not using fields
14:04 egonw yes, an IOOption
14:04 shk3 so if you want them, you need to switch on and set your props
14:04 egonw thingy
14:04 egonw will do that
14:05 shk3 and the mdlwriter shouldn't use that at all, obviously
14:05 shk3 only the sdf one
14:06 egonw yes
14:06 egonw shk3: how about this:
14:07 egonw if people do not want to write certain properties...
14:07 egonw it's sort of up to them to remove them first, not?
14:07 egonw I mean, giving the LEGO design of the CDK, it's up to the user to prepare what needs to be serialized, not?
14:08 egonw and indeed... it should not be too difficult for users to setProperties(null)
14:08 shk3 what I worry about
14:08 egonw or IMOlecule.getProperties().set("someKey"
14:08 egonw or IMOlecule.getProperties().set("someKey", null)
14:08 shk3 some cdk classes are using the properties internally
14:08 shk3 without the user knowing about it
14:09 egonw yes
14:09 shk3 on the other hand
14:09 egonw should it filter those out
14:09 egonw ?
14:09 egonw there are not so many
14:09 shk3 these should not be needed after running
14:09 egonw yes, that makes sense
14:09 shk3 so no problem when they are unset
15:10 egonw shk3: https://sourceforge.net/tracker2/?func=detail&am​p;aid=2668537&group_id=20024&atid=320024
15:10 egonw you can find the patch at that link
15:10 egonw please review
15:12 shk3 will get a coke and then do it
15:13 egonw thanx!
16:07 |egonw| joined #cdk
16:27 CIA-62 cdk: egonw * r14366 /cdk/branches/cdk-1.2.x/src/ (5 files in 3 dirs): Split up SDF writer from the MDL molfile writer
16:28 CIA-62 cdk: egonw * r14367 /cdk/branches/cdk-1.2.x/src/main/org/ope​nscience/cdk/io/formats/SDFFormat.java: Point to the new writer
17:10 egonw joined #cdk
18:48 egonw joined #cdk
19:01 timvdm joined #cdk
19:48 CIA-62 cdk: egonw * r14368 /cdk/branches/cdk-1.2.x/src/​main/org/openscience/cdk/io/ (6 files in 2 dirs): Fixed the use of the iterating readers for Jython, by reverting an patch r10907. Problem was reported on http://chem-bla-ics.blogspot.com/20​09/02/cdk-12-release-candidate.html
19:48 zarah CIA-62's link is also http://tinyurl.com/dmgwt4
20:04 bag joined #cdk

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