Camelia, the Perl 6 bug

IRC log for #bioclipse, 2009-06-30

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

All times shown according to UTC.

Time Nick Message
05:15 sneumann joined #bioclipse
06:40 Gpox joined #bioclipse
06:41 Gpoks joined #bioclipse
06:46 Gpox joined #bioclipse
07:26 olass joined #bioclipse
07:27 olass moin
07:27 zarah ni hao olass
07:47 olass @pz 1374
07:47 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1374
07:50 olass @pz 1374 test
07:50 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1374
07:51 olass test @pz 1374 test
08:01 Gpox awk '{if(i==3276) print $1 }/^\${4}/{i++}' ChEBI_complete.sdf
08:08 egonw joined #bioclipse
08:13 steinbeck joined #bioclipse
08:15 shk3 joined #bioclipse
08:54 CIA-52 bioclipse: goglepox * r11531 /bioclipse2/trunk/plugins/net.bioclipse.cdk​.ui.sdfeditor/src/net/bioclipse/cdk/ui/sdfe​ditor/business/MoleculeTableManager.java:
08:54 CIA-52 bioclipse: Ignore empty lines that are not part of any property
08:54 CIA-52 bioclipse: It logs a warning indicating a position in the file
08:54 CIA-52 bioclipse: from where it started reading the properties for the current molecule.
08:54 CIA-52 bioclipse: Solves bug #1374
08:54 zarah bug #1374 | http://tinyurl.com/lzwwrc
08:55 CIA-52 bioclipse: goglepox * r11532 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.jchempaint/src/net/bioclipse/cdk/jchempa​int/widgets/JChemPaintEditorWidget.java:
08:55 CIA-52 bioclipse: Do not set diry if generating 2D coordinates
08:55 CIA-52 bioclipse: when opening file.
08:55 CIA-52 bioclipse: Fixes bug #1372
08:55 zarah bug #1372 | http://tinyurl.com/mo4axz
09:05 steinbeck joined #bioclipse
09:44 steinbeck joined #bioclipse
09:46 CIA-52 bioclipse: goglepox * r11533 /bioclipse2/trunk/plugins/net.bioclipse​.cdk.jchempaint.view/src/net/bioclipse/​cdk/jchempaint/view/SWTRenderer.java:
09:46 CIA-52 bioclipse: Added debug messages to visit method
09:46 CIA-52 bioclipse: relates to bug #905
09:46 zarah bug #905 | http://tinyurl.com/c9ql3g
09:46 steinbeck joined #bioclipse
09:50 CIA-52 bioclipse: shk3 * r11534 /bioclipse2/trunk/plugins/net.bioclipse​.seneca/src/net/bioclipse/seneca/job/St​ochasticStructureElucidationJob.java: if cancel is done, result so far is still displayed
09:58 edrin1 joined #bioclipse
09:58 edrin1 hm
09:58 edrin1 there is an edrin ghost
10:00 edrin1 ok
10:00 edrin1 left #bioclipse
10:01 edrin joined #bioclipse
10:01 Topic for #bioclipseis now An open source workbench for chem- and bioinformatics - http://www.bioclipse.net/ | BioclipseWiki => http://wiki.bioclipse.net/ !! | logs => http://colabti.org/irclogge​r/irclogger_logs/bioclipse & http://moritz.faui2k3.org/irc​log/out.pl?channel=bioclipse || Bioclipse2 beta released! See http://bioclipse.blogspot.com/
10:02 edrin hi
10:02 zarah privet edrin
10:46 CIA-52 bioclipse: shk3 * r11535 /bioclipse2/trunk/plugins/net.bioclip​se.cdk.jchempaint/doc/molecules.html: adopted jcp help to new "new file" mechanism
10:56 CIA-52 bioclipse: shk3 * r11536 /bioclipse2/trunk/plugins/net.bioclipse.sene​ca/src/net/bioclipse/seneca/job/UserConfigur​ableStochasticStructureElucidationJob.java: result contains values for score etc.
10:56 CIA-52 bioclipse: shk3 * r11537 /bioclipse2/trunk/plugins/net.bioclipse.da​ta.seneca/data/hko1/hko1usersettings.sjs: generator was not active
11:00 olass joined #bioclipse
11:02 * olass has a new (old) laptop
11:05 olass @pz 1374
11:06 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1374
11:07 CIA-52 bioclipse: shk3 * r11538 /bioclipse2/trunk/plugins/net.bioc​lipse.bibtex/src/net/bioclipse/bib​tex/business/IBibtexManager.java: corrected comment
11:10 CIA-52 bioclipse: shk3 * r11539 /bioclipse2/trunk/plugins/net.biocl​ipse.specmol/src/net/bioclipse/spec​mol/business/ISpecmolManager.java: corrected comment
11:12 CIA-52 bioclipse: shk3 * r11540 /bioclipse2/trunk/plugins/net.biocli​pse.spectrum/src/net/bioclipse/spect​rum/business/ISpectrumManager.java: corrected comment
11:17 CIA-52 bioclipse: shk3 * r11541 /bioclipse2/trunk/plugins/net.bioclipse.cdk.jchem​paint/src/net/bioclipse/cdk/jchempaint/business/ (2 files): removed the . from @PublishedClass. It seems it makes the " included in the generated help and no other managers have the .
11:17 shk3 olass: please note my latest commit, it's minor, but sort of core stuff.
11:18 shk3 if you are not happy with it, do whatever you like.
11:18 olass ok
11:18 olass sounds good
11:18 olass will be an RC5 anyway
11:18 shk3 ok
11:19 shk3 in a way, it's a bug that comments with . show up in help as blah."
11:19 olass file a bug report
11:20 shk3 btw, I am still missing help pages
11:20 shk3 like there seems to be nothing about balloon, but this is part of core bc
11:20 shk3 ie the distribution
11:20 shk3 or did I oversee it?
11:22 masak joined #bioclipse
12:09 olass egonw: we solved the inchi problem
12:10 * masak does his daily rebase of his local branch
12:10 egonw olass++
12:10 egonw Gpox++
12:11 egonw masak: ping
12:13 masak egonw: pong
12:15 olass shk3: bug 1211
12:15 shk3 ah
12:17 masak egonw, shk3: I have Bioclipse running with commit r11541 not yet applied.
12:17 masak I cannot reproduce the bug you're talking about.
12:17 masak perhaps I'm going about it wrong.
12:18 shk3 when I open the help
12:18 shk3 via help->help contents
12:19 shk3 there is scripting langage documentation
12:19 masak oh!
12:19 shk3 this has a section jcp
12:19 shk3 the header reads
12:19 shk3 Manager for the JChemPaintEditor scripting language."
12:19 shk3 including "
12:19 * masak has a look
12:20 shk3 it looks like it depends on the .
12:20 egonw ok, that was pretty crucial information
12:20 shk3 but this is an assumption
12:20 egonw btw, please do revert the patch
12:20 egonw it's a work around for a problem (bug) somewhere else
12:20 masak must be the script that picks out the documentation, then.
12:20 shk3 have a look at bug #1376
12:20 masak egonw: are you asking me or shk3?
12:20 shk3 it's all in thee
12:20 egonw asking shk3
12:21 shk3 well, I think the . is not necessary anyway
12:21 shk3 no other section has it
12:21 egonw I think it is
12:21 shk3 to be consistent, i would remove it
12:21 egonw to be consistent, I'd add periods for the other managers
12:21 shk3 well, then we need to add it everywhere else, if you wan tit
12:21 shk3 go ahead if you think so
12:22 egonw olass: as Bioclipse Oracle :)... please decide
12:22 egonw but I still like to patch reverted, being a workaround for a problem somewhere else
12:22 masak shk3: could you be more specific about where in the help system the offending line can be found?
12:22 shk3 ooh
12:22 shk3 Bioclipse Scripting
12:22 egonw masak: it's one of our perl scripts to create docs
12:22 masak egonw: aye.
12:23 shk3 Scripting Language Documentation
12:23 egonw shk3: did you file a bug report about the " showing up in the HTML docs?
12:23 masak I still don't recall special-casing full stops.
12:23 shk3 section jcp
12:23 shk3 yes
12:23 shk3 bug #1376
12:23 masak shk3: I see it. thanks.
12:23 masak that's my script. :/
12:23 egonw @pz 1376
12:23 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1376
12:23 shk3 all the crucial information you requested is in there
12:23 shk3 I am for a meeting
12:23 egonw masak: ok, you take that bug?
12:24 masak egonw: yes, right away.
12:24 shk3 if anything is unclear, I will be back
12:24 masak this has got me interested.
12:24 shk3 ok, see you later
12:24 egonw please do revert the patch
12:38 CIA-52 bioclipse: egonw * r11542 /bioclipse2/trunk/plugins/net.bioclipse.cdk.jchem​paint/src/net/bioclipse/cdk/jchempaint/business/ (2 files):
12:38 CIA-52 bioclipse: Revert "removed the . from @PublishedClass. It seems it makes the " included in the generated help and no other managers have the ."
12:38 CIA-52 bioclipse: This reverts commit c895a62bf3ffc44ca65305df09a9c0f4f4c9c2b2.
12:38 CIA-52 bioclipse: The patch does not solve a problem, but merely works around a bug elsewhere (see #1376).
12:40 masak Gpox found the bug. :P
12:40 egonw Gpox++
12:40 masak actually, it's still kind of a mystery why it occurs, or why shk3's patch would fix it.
12:40 masak but I think I have a patch that mildens the flaw. let's apply it and see.
12:41 masak by they way, this would never happen if we had a full Java parser. :)
12:41 masak (or even a subset that recognized annotations)
12:41 egonw see above
12:41 egonw that patch did not fix nothing
12:42 egonw it was a mere workaround
12:42 masak yes, but it was not clear even why the workaround worked.
12:42 egonw ic
12:42 masak I still don't see why the '.' character has anything to do with it.
12:43 masak more likely it was the presence of whitespace after the final "
12:47 egonw http://chem-bla-ics.blogspot.com/2009/06​/jchempaint-hack-session-at-uppsala.html
12:47 zarah egonw's link is also http://tinyurl.com/mrp64q
12:48 masak no, that did not seem to fix the problem.
12:48 masak going into deep debug mode.
12:51 masak by the way, I see a lot of things are called 'scripting language' in the documentation despite not being scripting languages.
12:51 masak I'll need to rectify that. soon as I track down this bug.
12:52 masak also, the documentation generator gives of at least one warning that must be fixed.
12:56 olass masak++ //for tending to these things
12:56 masak one could argue it falls under 'cleaning up my own mess' :P
13:03 CIA-52 bioclipse: ospjuth * r11543 /bioclipse2/trunk/plugins/net.bioclips​e.cdk.business/src/net/bioclipse/cdk/d​omain/CDKMoleculePropertySource.java: Fixed CDKMoleculePropertySource to return SMILES, and InChI properties from the CDKMolecule, and not used cached values.
13:04 CIA-52 bioclipse: ospjuth * r11544 /bioclipse2/trunk/plugins/net.bioclipse.​cdk.jchempaint.view/src/net/bioclipse/cd​k/jchempaint/view/JChemPaintWidget.java: Decrease font size of changed/generated labels in JCP to 16.
13:05 masak ah hah haa.
13:06 masak egonw: the new 'doi' annotation attribute has broken the simplistic documentation generator.
13:06 masak egonw: not your fault, mine.
13:06 egonw ah
13:06 egonw egonw--
13:06 egonw file me a bug report if you like
13:06 masak again, problem would not occur if I didn't cheat so much with pure-text parsing.
13:06 masak egonw: no, I need to fix this.
13:07 egonw ok
13:07 masak there's nothing wrong with the annotations, it's just that the generator can't handle them being there.
13:08 egonw ah
13:09 egonw please restore my karma then :)
13:09 egonw ah, never mind
13:09 masak egonw++
13:09 egonw leave it like that would have been fine
13:09 egonw for jumping to conclusions
13:09 masak masak--
13:09 masak man, this is not good.
13:10 masak (also, one could argue that now, with the 'doi' attribute, having the description attribute be called 'value' is pretty crappy.)
13:17 CIA-52 bioclipse: ospjuth * r11545 /bioclipse2/trunk/plugins/net.bioclipse.cdk​.ui.sdfeditor/src/net/bioclipse/cdk/ui/sdfe​ditor/business/IMoleculeTableManager.java: Removed MolTableManager from Javascript environment by removing all publishedclass/publishedmethods.
13:21 CIA-52 bioclipse: ospjuth * r11546 /bioclipse2/trunk/plugins/ (2 files in 2 dirs): Commented out SMILES folder from sample data and extended sample data since context menu options for SMILES files has been postponed to 2.1.
13:21 egonw masak: :) so much for defaults ...
13:22 masak remember, kids: don't parse computer languages with ad-hoc regexes. that way lies Pain with a capital P.
13:23 masak basically, we haven't had working manager documentation generation since 'doi' was introduced -- and we haven't noticed until now. there's something wrong with our error reporting. will look into that as soon as I'm done fixing this bug.
13:24 * egonw mumbles... auto building with ant calling the perl script and returning fail ...
13:24 * egonw egonw mumbles 'crossed red... that good'
13:25 egonw mmm
13:25 * egonw 's Yoda impersonation did not quite work out :)
13:26 masak no, decidedly not. :)
13:26 egonw well, 'get funny, or die trying' ? :)
13:26 egonw (the latter clearly applies)
13:27 masak :)
13:33 masak next parsing surprise: people use nested generics, and the render-as-html script doesn't think to look for those and escape the '<' and '>' characters. :/
13:34 masak ooooooh! they are replaced, but only once-per-string. :/
13:34 * masak fixes
13:35 egonw hahahaha
13:35 masak masak--
13:35 masak who wrote this crap?
13:37 masak ok, everything works now.
13:39 masak here comes a few fixes.
13:39 CIA-52 bioclipse: carl_masak * r11547 /bioclipse2/trunk/plugins/net.biocli​pse.doc/parse-file-for-annotations:
13:39 CIA-52 bioclipse: [parse-file-for-annotations] fixed whitespace bug
13:39 CIA-52 bioclipse: Turns out string quotes ('"') are not properly removed if they are
13:39 CIA-52 bioclipse: followed by whitespace. Fixed by making the rules match whitespace also.
13:39 CIA-52 bioclipse: Does not fix #1376.
13:39 CIA-52 bioclipse: carl_masak * r11548 /bioclipse2/trunk/plugins/net.biocli​pse.doc/parse-file-for-annotations:
13:39 CIA-52 bioclipse: [parse-file-for-annotations] better parsing of method declarations
13:39 CIA-52 bioclipse: This patch allows 'public SomeType' to occur on one line, and the function
13:39 CIA-52 bioclipse: name to occur on the next. Also documented the regexes a bit.
13:39 CIA-52 bioclipse: carl_masak * r11549 /bioclipse2/trunk/plugins/net.biocli​pse.doc/parse-file-for-annotations:
13:39 CIA-52 bioclipse: [parse-file-for-annotations] takes 'doi' into account
13:39 CIA-52 bioclipse: There's a new 'doi' attribute in the @PublishedClass annotation, and
13:39 CIA-52 bioclipse: the ad-hoc annotation parser needs to correctly remove those.
13:40 CIA-52 bioclipse: carl_masak * r11550 /bioclipse2/trunk/plugins/net.​bioclipse.doc/render-as-html:
13:40 CIA-52 bioclipse: [render-as-html] made escapings global
13:40 CIA-52 bioclipse: There should be a /g flag on all the HTML escape substitutions.
13:40 masak there.
13:40 masak the documentation generator is now returned to its former 'glory'.
13:41 egonw masak++
13:41 masak I'll see if I can strict up the error reporting some from the run-the-whole-thing script.
13:43 olass egonw: I need your help with aromaticity. Could you swing by after your call?
13:43 masak 'confused aromaticity' :P
13:48 CIA-52 bioclipse: ospjuth * r11551 /bioclipse2/trunk/plugins/n​et.bioclipse.jmol/toc.xml: Renamed "Scripting Language Reference" to "Jmol Scripting Reference" since it is conflicting with Bioclipse scripting language.
14:05 masak seems we'll have a rainy weekend. maybe even some thunderstorms.
14:09 CIA-52 bioclipse: carl_masak * r11552 /bioclipse2/trunk/plugins/net.bi​oclipse.doc/run-the-whole-thing:
14:09 CIA-52 bioclipse: [net.bioclipse.doc] better error handling
14:09 CIA-52 bioclipse: More exactly, the run-the-whole-thing script now has two points where
14:09 CIA-52 bioclipse: it checks for errors: (1) after parse-file-for-annotations, and
14:09 CIA-52 bioclipse: (2) after running tidy on the render-as-html output. The first will catch
14:09 CIA-52 bioclipse: specific parsing-related errors; the second will catch more overarching
14:09 CIA-52 bioclipse: stuff.
14:11 shk3 joined #bioclipse
14:12 CIA-52 bioclipse: ospjuth * r11553 /bioclipse2/trunk/plugins/net​.bioclipse.balloon.business/ (9 files in 3 dirs): Added help pages for Balloon. Solves 1211.
14:12 shk3 any problems with that scripting help stuff I can comment on?
14:14 shk3 did you decide on if fullstops or not?
14:21 masak not really, but I do agree on the value of consistency.
14:22 shk3 I can revert my commit
14:22 shk3 no problem with that
14:22 shk3 I really don't care
14:22 shk3 If you tell me what minimizes the risk of me being blamed for whatever, I will do it.
14:22 shk3 ;-)
14:33 CIA-52 bioclipse: ospjuth * r11554 /bioclipse2/trunk/plugins/ (7 files in 5 dirs): Extracted basic scripting cheat sheet from cdk scripting cheat sheet into a separate for the scripting.ui plugin.
14:34 masak shk3: your commit was reverted shortly after you left.
14:35 CIA-52 bioclipse: ospjuth * r11555 /bioclipse2/trunk/plugins/net.bi​oclipse.ui/intro/tutorials.xml: Removed link to getting started guide since it does not exist; we have cheat sheets for that.
14:46 CIA-52 bioclipse: ospjuth * r11556 /bioclipse2/trunk/ (43 files in 43 dirs): Set version to 2.0.0.RC5 for release candidate 5.
14:46 CIA-52 bioclipse: ospjuth * r11557 /bioclipse2/trunk/ (2 files in 2 dirs): Updated update sites for RC5
14:47 CIA-52 bioclipse: ospjuth * r11558 /bioclipse2/trunk/plugins/ (4 files in 4 dirs): Set version to 2.0.0.RC5 for release candidate 5.
14:48 CIA-52 bioclipse: ospjuth * r11559 /bioclipse2/trunk/plugins/ (10 files in 10 dirs): Set version to 2.0.0.RC5 for release candidate 5.
14:49 CIA-52 bioclipse: ospjuth * r11560 /bioclipse2/trunk/features/ (13 files in 13 dirs): Updated feature versions to 2.0.0.RC5 for release candidate 5.
14:55 masak olass: this might be it: org/openscience/cdk/aromaticity/C​DKHueckelAromaticityDetector.java
14:55 olass masak: no, we should the the CDKManager for that
14:55 olass use
14:55 masak indeed.
14:56 masak cdk.perceiveAromaticity, then.
15:00 masak hm, that manager method calls the mis-spelled percieveAtomTypesAndConfigureAtoms method in AtomContainerManipulator...
15:01 masak I'd report the type as a CDK bug, but it might be one of those cases when changing something leads to more trouble than keeping it. :)
15:02 masak anyway, no Hueckel aromaticity detection seems to be done in that method.
15:02 masak shame that egonw left. :/ I'd like to ask him how it all fits together...
15:03 CIA-52 bioclipse: ospjuth * r11561 /bioclipse2/trunk/plugins/ (3 files in 3 dirs): Updated feature versions to 2.0.0.RC5 for release candidate 5.
15:04 CIA-52 bioclipse: ospjuth * r11562 /bioclipse2/trunk/ (12 files in 12 dirs): Updated DS versions to 2.0.0.RC5.
15:09 olass masak: file it as a bug
15:09 olass and let egonw sort it out
15:10 masak no, I think the CDK developers already know about it.
15:10 olass ok
15:10 CIA-52 bioclipse: ospjuth * r11563 /bioclipse2/trunk/plugins/ (net.bioclipse.ds.signsig/ net.bioclipse.ds.signsig.bursi/): Removed from Bioclipse SVN since not working for the general audience.
15:10 masak there were a lot of other occurrences of the word in the vicinity, all correctly spelled.
15:10 olass I see
15:11 masak only the method name was wrong, and they can't change that, because they can't change the API.
15:35 masak olass: first test passes! \o/
15:35 olass weehoow!
15:36 masak I had to basically throw away everything and start over, but I think that was the right decision.
15:37 masak maybe the code already written used a previous version of BioJava or something -- it turns out that instead of like 100 lines of code, one could do what we want in 5.
15:37 masak anyway, time to write more tests. :)
15:39 olass masak: that is 95% less
15:40 olass not bad
15:40 masak could have been 1 line of code, but there was some exception handling also.
16:08 edrin left #bioclipse
16:22 CIA-52 bioclipse: ospjuth * r11564 /bioclipse2/trunk/features/ne​t.bioclipse.platform_feature/ (bioclipse.product feature.xml): Set version to 2.0.0.RC5 for release candidate 5.
16:26 masak olass: ok, found it. I had just missed the line in question before.
16:26 masak CDKHueckelAromaticityDetector.​detectAromaticity(todealwith);
16:54 CIA-52 bioclipse: shk3 * r11565 /bioclipse2/trunk/plugins/net.bioclipse.seneca/ (6 files in 6 dirs): more work on the ga implementation
18:37 edrin joined #bioclipse
18:39 sneumann_ joined #bioclipse
19:39 Topic for #bioclipseis now An open source workbench for chem- and bioinformatics - http://www.bioclipse.net/ | BioclipseWiki => http://wiki.bioclipse.net/ !! | logs => http://colabti.org/irclogge​r/irclogger_logs/bioclipse & http://moritz.faui2k3.org/irc​log/out.pl?channel=bioclipse || Bioclipse2 beta released! See http://bioclipse.blogspot.com/ || http://www.youtube.com/watch?v=ZUBQknWUEYU
20:46 olass joined #bioclipse
20:46 olass @seen egonw
20:46 zarah egonw was last seen 7 h 5 m 8 s ago in #bioclipse saying 'masak++'
20:47 olass g'night
22:18 rojasm joined #bioclipse

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