Camelia, the Perl 6 bug

IRC log for #bioclipse, 2007-12-05

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

All times shown according to UTC.

Time Nick Message
02:23 kaskelot joined #bioclipse
08:26 olas joined #bioclipse
08:37 steinbeck joined #bioclipse
09:14 jonalv joined #bioclipse
09:40 steinbeck left #bioclipse
10:25 egonw joined #bioclipse
10:35 egonw_ joined #bioclipse
12:00 CIA-19 bioclipse: jonalv * r4397 /branches/bioclipse2/bioinformatics/pl​ugins/net.bioclipse.biojava.business/ (5 files in 4 dirs): Working with getting Spring-osgi to run. For the moment it's not going great...
12:06 masak joined #bioclipse
12:07 EskilA joined #bioclipse
12:15 jonalv olas: ping
13:56 masak hmmm
13:56 masak I'm running into a dependency problem again in bc2
13:57 masak let me try to reason it out here, and maybe it will be clearer
13:57 masak net.bioclipse.core contains core things and not UI things
13:57 masak that must be considered sane
13:58 masak among the core things of interest here are the different scripting sessions: a js session, a groovy session etc
13:58 masak these are all instantiated through net,bioclipse.core.Activator
13:59 masak so far so good
14:00 masak the different consoles JsConsole, GroovyConsole etc lie in the package net.bioclipse.ui
14:00 masak and they all extend ScriptingConsoleView in net.bioclipse.ui.views
14:00 masak that _also_ feels sane
14:00 masak today I started thinking about adding logging to the consoles
14:01 masak the idea was that apart from a js session, the JsConsole would also have a 'logging session', so that output could come from two directions into the console
14:02 masak so my first idea was to make Logger derive ScriptingEnvironment...
14:02 masak now that feels slightly wrong, because the Logger isn't a ScriptingEnvironment
14:03 masak second problem: the _idea_ of the Logger is that it's asynchronous
14:03 masak i.e. it could send a logging message to the console at any time
14:04 masak that's not how the ScriptingEnvironment classes work
14:04 masak they know nothing about consoles, they just send a response to whichever object call them
14:05 masak the Logger needs to know a bunch of client consoles that are interested in logging
14:06 masak ...and here comes the third and fatal problem: that list of clients should intuitively be a List<ScriptingConsoleView>
14:06 masak but this requires a dependency of core on ui
14:06 masak and that's a big no-no
14:11 * masak doesn't know where to go from here
14:52 olas masak: an extension point?
14:52 olas if UI is present, then use it
14:53 olas masak: want to talk more about it today or maybe tomorrow?
14:53 * olas is tired
14:57 masak olas: jonalv++ suggested using a Listener
14:57 masak that seems to work
14:58 olas aah
14:58 * jonalv is helping masak. We will figure this out...
14:58 masak :)
14:58 * olas is very tired
14:58 * masak 2
14:58 * olas ++
14:58 * jonalv is also tired. I think I got tioered by seeing how tiered you were... :)
14:59 olas I am not tiered
14:59 masak he meant "tyred"
14:59 olas nor tioered
14:59 olas :-)
14:59 jonalv *puts fingers in ears'
14:59 jonalv 'lalala'
15:02 EskilA the tires on the bus go round round round...
15:05 masak it's almost like it's Friday afternoon...
15:06 EskilA yay :)
15:14 thomas_ku joined #bioclipse
15:42 olas left #bioclipse
16:20 EskilA left #bioclipse
16:39 CIA-19 bioclipse: biocoder * r4398 /trunk/plugins/net.bioclipse.chart/ (9 files in 5 dirs): Implemented selection from MatrixGridEditor to Chart plot
16:40 EskilA joined #bioclipse
16:43 CIA-19 bioclipse: biocoder * r4399 /trunk/plugins/net.bioclipse.statistics/src/​net/bioclipse/editors/MatrixGridEditor.java: Implemented selection from MatrixGridEditor to ChartUtils
16:56 EskilA left #bioclipse

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