[00:03:50] *** davidstrauss has joined #hudson
[00:07:42] *** Lewisham has quit IRC
[00:10:48] *** kfu_ has quit IRC
[00:31:46] *** ircuser999 has joined #hudson
[00:33:06] *** Lewisham has joined #hudson
[00:33:15] *** Baraius has quit IRC
[00:50:31] *** decriptor is now known as decriptor|out
[01:02:47] *** wsmoak has quit IRC
[01:03:21] <Lewisham> is there a good place to see how to write a plugin to create a new page in Hudson?
[01:03:37] <Lewisham> I have JDepend reports being generated, and now just need a way of showing them next to a build
[01:04:26] *** abayer has quit IRC
[01:04:59] *** abayer has joined #hudson
[01:05:24] <Lewisham> looks like something to do with Stapler?
[01:08:47] <Lewisham> or Jelly?
[01:09:00] <Lewisham> all the HTML is all ready to go, I just need to have a button appear...
[01:09:22] <Lewisham> which then makes a new Hudson page which I can inject my HTML into
[01:13:04] *** ircuser999 has quit IRC
[01:15:41] <Lewisham> no-one? :(
[01:46:17] *** BigAllan has quit IRC
[01:56:11] <mrooney> Lewisham: you can use my plug-in!
[01:56:27] <Lewisham> which one is that?
[01:56:29] <mrooney> it is a generic HTML archiver, basically
[01:57:12] <Lewisham> ok
[01:57:29] <Lewisham> where's the source?
[01:57:36] *** mrooney1 has joined #hudson
[01:57:40] *** mrooney has quit IRC
[01:58:30] *** mrooney1 is now known as mrooney
[01:58:35] <mrooney> I really need to figure out how to release it
[01:58:45] *** keshureddyp has quit IRC
[01:59:37] <mrooney> Lewisham: let me get you an .hpi
[02:00:10] <Lewisham> can you decompile a .hpi?
[02:01:05] <mrooney> if you need the source it is already hosted on hudson
[02:01:17] <mrooney> I just haven't figured out how to get it on the list in Manage Hudson
[02:01:50] <Lewisham> yeah, just the source is good
[02:01:56] <Lewisham> what's the plugin called?
[02:02:42] *** keshureddyp has joined #hudson
[02:05:25] <calculus> that list in Manage Hudson is generated by a script (possibly a cron-job) on kohsuke's machines
[02:06:02] <kohsuke> "list in Manage Hudson"?
[02:06:13] <kohsuke> You mean the update center metadata?
[02:06:40] <calculus> I believe so
[02:06:45] <calculus> mrooney: ^^
[02:07:24] <mrooney> kohsuke: yeah I have tried so many times to get a pom.xml which lets me release
[02:07:27] <mrooney> and failed always
[02:07:35] <abayer> I've never had a problem
[02:07:47] <mrooney> clearly other people have been successful
[02:07:49] <kohsuke> what's the error that you get?
[02:08:10] <mrooney> let me see
[02:09:45] * Lewisham managed to get a thingy appearing in the sidebar
[02:09:47] <Lewisham> PROGRESS!
[02:10:10] * Lewisham is using kohsuke 's Emma plugin as a guideline
[02:10:19] <mrooney> Lewisham: I can send you my html archiver which already works momentarily :)
[02:10:36] <mrooney> we should just have one generic one that works instead of one for each use case which does the same exact thing with a different name
[02:10:36] <Lewisham> OK :)
[02:14:20] * Lewisham is going AFK
[02:14:26] <mrooney> Lewisham: http://wiki.hudson-ci.org/download/attachments/37749341/NCover.hpi, check out http://wiki.hudson-ci.org/display/HUDSON/NCover+Plugin for the overview
[02:14:31] <Lewisham> sweet
[02:14:59] <mrooney> currently it still has unimportant references to ncover such as the name and URL you see, but you just give it a directory relative to workspace and which files in it to display and it will do it per build or project
[02:15:49] <Lewisham> the only problem is an hpi is byte-code compiled :)
[02:15:53] <Lewisham> so I can't read the source ;)
[02:15:57] <Lewisham> not to matter!
[02:16:02] <Lewisham> thanks for the help
[02:16:17] <Lewisham> I'm going AFK, my RSI is twinging, which means I've been hacking on this too long ;)
[02:17:20] <mrooney> the source is already at plugins/ncover
[02:17:23] <mrooney> in hudson
[02:17:47] <Lewisham> cool
[02:17:56] <mrooney> https://svn.dev.java.net/svn/hudson/trunk/hudson/plugins/ncover to be specific
[02:17:57] <Lewisham> I'll check it out of SVN and have a peek tomorrow
[02:17:58] <Lewisham> thanks mate
[02:18:16] <mrooney> I need a more generic name for, perhaps HTML Reporter or something
[02:18:20] <mrooney> no problem let me know how you find it :)
[02:18:32] * Lewisham is reminded to commit to SVN and follow those Google Code instructions for converting Git to SVN
[02:18:53] <Lewisham> http://stackoverflow.com/questions/661018/pushing-an-existing-git-repository-to-svn
[02:19:06] <mrooney> the first mistake was using git to begin with
[02:19:09] * mrooney much prefers bzr
[02:19:13] <mrooney> ;)
[02:22:58] <mrooney> oh no I can't remember my Hudson SVN password
[02:23:28] <mrooney> oh there we go, it was the one for java.net not hudson-ci.org
[02:27:02] *** admc has quit IRC
[02:27:06] *** admc has joined #hudson
[02:27:52] <mrooney> whoa hey maybe it worked?
[02:31:07] <mrooney> kohsuke: what is the best way to see if it was actually released?
[02:31:27] <kohsuke> if the mvn release process completed without an error, it was actually released.
[02:35:24] <mrooney> neato burrito!
[02:39:15] <mrooney> kohsuke: now I just need to figure out why the DirectoryBrowserSupport eats javascript in iframes, does anything ring a bell with that?
[02:39:53] <mrooney> I have JS in the page "hosted" by the DBS and it works fine, but within that page I have an iframe with JS which doesn't work
[02:39:59] <kohsuke> DBS just writes out what's in the directory
[02:40:11] <mrooney> if I make that directory accessible via apache and view the same file it works fine
[02:40:21] <kohsuke> It doesn't do anything remotely intelligent like scraping JavaScript.
[02:40:26] <mrooney> hmm
[02:40:55] <mrooney> I can try to make a simple test case which reproduces the issue, perhaps a simple alert() in an iframe
[02:43:30] <kohsuke> sure
[02:44:58] <kohsuke> I think it'll be more productive if you use something like FireBug to look at it from the browser side
[02:49:59] <Lewisham> kohsuke: Does an Action also handle the page it points to, or is that supposed to be a different file?
[02:50:21] <kohsuke> It can point to a different URL, but it's normally the same.
[02:50:33] <kohsuke> See JavadocAction for example.
[02:52:52] <Lewisham> cool, just making sure my separation of concerns was right
[02:57:00] <Lewisham> oh dear
[02:57:09] <Lewisham> looks like getting results from builds is not easy
[02:57:11] <Lewisham> ack
[02:57:21] <Lewisham> the code for the Emma plugin is pretty scary
[02:57:46] <kohsuke> "getting results from builds"?
[02:59:12] <Lewisham> so I have a Recorder that runs
[02:59:25] <Lewisham> and generates a JDepend object which has all the data in it
[02:59:36] <Lewisham> and I need to retrieve that back
[02:59:49] <Lewisham> I think this is something to do with adding build actions?
[02:59:56] * Lewisham is scanning through the Emma code
[03:00:18] <Lewisham> final EmmaBuildAction action = EmmaBuildAction.load(build,rule,healthReports,localReport); build.getActions().add(action);
[03:00:56] *** alexlod has quit IRC
[03:01:06] <Lewisham> I think BuildAction is what I want
[03:01:21] <Lewisham> I'm tripping over the naming ("It's a result of the build, right?")
[03:04:25] * Lewisham just had a ray of shining light come down
[03:04:29] <Lewisham> *I SEE*
[03:04:52] <Lewisham> ProjectAction is a sidebar button at the project level, BuildAction is a sidebar button on builds
[03:04:56] <Lewisham> man, I feel so stupid now
[03:06:38] <mrooney> haha
[03:06:41] <Lewisham> is there a difference between summary.jelly and index.jelly?
[03:07:02] <mrooney> did you ever see if the plugin I sent does what you want? not that there is anything wrong with learning :)
[03:07:20] <Lewisham> yeah, it's good
[03:07:25] <Lewisham> I'm sort of working with all of them
[03:07:29] <Lewisham> trying to find the common threads
[03:08:01] <mrooney> yeah cool, let me know if it seems to be missing something
[03:09:42] <mrooney> Lewisham: I'm going to eventually add two more options that let you set the build/project text that currently is hard-coded to "Code Coverage" and the URL that appears as "ncover"
[03:10:01] <mrooney> then it should be pretty generic for archiving arbitrary html reports and displaying them
[03:10:17] <Lewisham> oh sweet
[03:10:32] <Lewisham> yeah
[03:10:35] <Lewisham> the reason I'm learning
[03:10:39] <Lewisham> is that I'm a PhD student
[03:10:51] <Lewisham> and this plugin was the "easy" jump off to a much more complex Hudson plugin
[03:10:59] <Lewisham> for some research I'm being paid to do :)
[03:11:52] <Lewisham> so I can't take the easy way out, as it'll just punish me when I get to the really hard one ;)
[03:16:36] <mrooney> haha yeah, that's funny
[03:16:43] <mrooney> that was about why I wrote this plugin too
[03:16:44] *** keshureddyp has quit IRC
[03:16:53] <mrooney> eventually I need to port phpundercontrol to Hudson
[03:16:54] *** keshureddyp has joined #hudson
[03:17:30] *** jieryn-w has joined #hudson
[03:34:08] *** mrooney has quit IRC
[03:39:32] *** rromanchuk has quit IRC
[03:46:12] <Lewisham> kohsuke: Is there any way to inject custom stylesheets into the <head> using Jelly?
[03:46:44] <kohsuke> Yes.
[03:46:57] <kohsuke> Do you want to just do it for specific pages or do you want to do it for all the pages?
[03:48:20] <Lewisham> a specific one
[03:48:42] <Lewisham> the JDepend HTML output requires some CSS to actually look right, otherwise it's all messed up
[03:48:54] <Lewisham> it's expecting some Maven CSS
[03:48:58] <kohsuke> then use <l:head>
[03:49:05] <Lewisham> cool, thanks
[03:53:11] <Lewisham> kohsuke: http://hudson.pastebin.com/m21822d78
[03:53:19] <Lewisham> that's not working, I'm missing something obvious
[03:53:31] <Lewisham> (the hardcoded CSS links also naughty ;) )
[03:54:13] <kohsuke> sorry, it was <l:header>
[03:54:17] <kohsuke> and it has to be in <l:layout>
[04:02:27] <Lewisham> ah, perfect
[04:02:47] <Lewisham> now to trim those down so they're don't mess with the hudson styles
[04:02:51] <Lewisham> and I should be golden
[04:02:54] <Lewisham> you're a star kohsuke
[04:03:02] <kohsuke> thanks!
[04:03:28] <Lewisham> I should really make the trip up from Santa Cruz for that hackathon
[04:03:43] <kohsuke> Yes, please
[04:03:43] <Lewisham> I'm not the only person at UC Santa Cruz that owes you and the Hudson team a beer at this point :)
[04:03:54] *** wsmoak has joined #hudson
[04:04:03] <abayer> Bring us slugs!
[04:04:20] <Lewisham> :D
[04:11:00] *** admc has quit IRC
[04:38:53] *** rromanchuk has joined #hudson
[04:44:19] *** alexlod has joined #hudson
[05:01:47] *** akostadinov has joined #hudson
[05:17:16] *** waz has quit IRC
[05:23:15] *** glassy has joined #hudson
[05:26:46] <glassy> good evening everyone!
[05:37:54] *** waz has joined #hudson
[06:51:04] *** waz has quit IRC
[07:09:16] *** wsmoak has quit IRC
[07:17:42] *** admc has joined #hudson
[07:22:38] <akostadinov> glassy: evening where? You should be punished by your local time forfeit
[07:29:17] <glassy> :-(
[07:29:24] <glassy> arizona...although now it is night :)
[07:34:32] *** alexlod has quit IRC
[07:50:55] *** akostadinov has quit IRC
[07:55:08] *** davidstrauss has quit IRC
[08:11:51] *** davidstrauss has joined #hudson
[08:14:26] *** davidstrauss has quit IRC
[08:38:02] *** khmarbaise has joined #hudson
[08:45:22] *** jieryn-w has quit IRC
[09:11:59] *** akostadinov has joined #hudson
[09:15:38] *** admc has quit IRC
[10:01:17] *** akostadinov has quit IRC
[11:01:48] *** dvrzalik has joined #hudson
[11:30:36] *** khmarbaise has quit IRC
[11:31:33] *** EggM4n has joined #hudson
[11:33:19] <EggM4n> Hey all, I'm trying to find some documentation on Hudson's Alpha feature for 'multi-configuration' projects - can anyone help please?
[11:37:57] *** khmarbaise has joined #hudson
[12:32:51] *** duncanmv has joined #hudson
[12:40:23] *** duncanmv has joined #hudson
[12:52:39] *** akostadinov has joined #hudson
[12:56:26] *** EggM4n has quit IRC
[13:02:22] *** wdk_ has joined #hudson
[13:02:22] *** wdk has quit IRC
[13:02:24] *** wdk_ is now known as wdk
[13:04:39] *** velo has joined #hudson
[13:09:14] *** timp_ has joined #hudson
[13:10:38] *** wdk_ has joined #hudson
[13:15:39] *** davidstrauss has joined #hudson
[13:25:03] *** wdk has quit IRC
[13:35:31] *** silentNinja has joined #hudson
[13:35:45] <silentNinja> l.o.
[13:57:26] *** waz has joined #hudson
[14:02:07] *** brent has joined #hudson
[14:09:06] <btQuark> hello all
[14:09:14] <btQuark> i've found a nice bug
[14:09:24] <btQuark> with ProxySCM the svn tagging function does not work against it
[14:09:35] <btQuark> ends in a "cannot cast ProxySCM to SubversionSCM" exception
[14:16:24] *** khmarbaise has quit IRC
[14:25:09] *** khmarbaise has joined #hudson
[14:34:58] *** jieryn-w has joined #hudson
[14:45:05] *** jdolan_ has joined #hudson
[14:48:19] <btQuark> has anyone thought of placing those view tabs on the side instead on the topß
[14:48:20] <btQuark> ?
[14:54:45] * duncanmv can't just figure out what the git plugin does when adding more than one repository. It does not check out each one in a subdirectory as subversion does.
[14:57:32] <duncanmv> anyone understands the logic?
[14:57:42] <duncanmv> it seems to add them as origin, origin1
[15:04:08] *** glassy has quit IRC
[15:05:22] *** abayer_ has joined #hudson
[15:08:36] *** abayer_ has left #hudson
[15:13:40] *** davidstrauss_ has joined #hudson
[15:14:41] *** davidstrauss has quit IRC
[15:20:19] *** waz has quit IRC
[15:23:26] *** abayer has quit IRC
[15:28:22] *** ircuser999 has joined #hudson
[15:40:51] *** waz has joined #hudson
[15:43:20] *** abayer has joined #hudson
[15:49:48] *** ircuser999 has quit IRC
[16:17:01] *** rromanchuk has quit IRC
[16:29:38] *** waz has quit IRC
[16:49:45] *** silentNinja has quit IRC
[16:50:39] *** rromanchuk has joined #hudson
[16:51:59] *** Baraius has joined #hudson
[17:00:04] *** Baraius has quit IRC
[17:40:34] *** mindless has joined #hudson
[17:55:34] *** ircuser999 has joined #hudson
[17:55:34] *** decriptor|out is now known as decriptor
[17:59:38] <rtyler> duncanmv: i didn;'t know you could add multiple repos for the git plugin o_O
[18:02:28] <duncanmv> is that plugin actively developed?
[18:02:55] <rtyler> yes and no
[18:03:03] <rtyler> yes there's plenty of flamewars about it
[18:03:07] <rtyler> no commits aren't often :P
[18:03:32] <rtyler> duncanmv: you should join our SF hackathon next saturday >:D
[18:07:03] <duncanmv> what would be the workflow to add features and test changes in the source code?
[18:07:58] <rtyler> duncanmv: from what I've observed tracking the development of the Git plugin:
[18:08:23] <rtyler> 1. Email the dev list either: proposing $CHANGE, asking why $CHANGE doesn't exist, etc
[18:08:28] <rtyler> 2. Discuss
[18:09:00] <rtyler> 3. Once you have a Syn+Ack, send patch
[18:09:08] <rtyler> 4. All parties involved Ack, commit
[18:09:18] <rtyler> there is a dude "responsible" for the plugin now IIRC
[18:09:38] <rtyler> but there's a lot of folks dependendent on it and use it in different ways, so there's a lot of push back on breaking changes
[18:15:38] *** ircuser999 has quit IRC
[18:18:17] *** akostadinov has quit IRC
[18:22:02] *** davidstrauss_ has quit IRC
[18:29:52] <duncanmv> rtyler: I mean from the code point of view, I have never developed for hudson
[18:30:21] <rtyler> heh
[18:30:22] <rtyler> fair enough
[18:30:36] <rtyler> http://wiki.hudson-ci.org/display/HUDSON/Extend+Hudson
[18:30:54] <rtyler> http://wiki.hudson-ci.org/display/HUDSON/Plugin+tutorial
[18:31:02] <rtyler> the basics at the bottom of that page I found really helpful
[18:33:36] *** khmarbaise has quit IRC
[18:40:52] *** dvrzalik has quit IRC
[18:43:35] *** davidstrauss has joined #hudson
[19:03:28] *** mrooney has joined #hudson
[19:05:07] *** alexlod has joined #hudson
[19:23:07] *** Lewisham has quit IRC
[19:35:06] <rtyler> kohsuke: FWIW, I've spoken with the Jython folks, at adding Jython onto the Python plugin, or as a core-level scripting language should actually be fairly easy
[19:35:27] <kohsuke> yeah, could be fun to come back to that.
[19:35:41] <kohsuke> We can already do Groovy. Can't see why other languages don't work.
[19:35:55] <rtyler> well, Jython and JRuby should be doable
[19:36:00] <kohsuke> The only tricky part with Groovy was the annotation processing.
[19:36:06] <rtyler> annotation?
[19:36:29] <kohsuke> We need to be able to scan the code either at the compile time or at the run time to list up @Extension and @QueryParameter and etc.
[19:36:35] <rtyler> ick
[19:36:42] *** alexlod_ has joined #hudson
[19:36:46] <kohsuke> Didn't Python has something like that?
[19:36:57] *** alexlod__ has joined #hudson
[19:36:57] *** alexlod has quit IRC
[19:37:05] <kohsuke> we could use the doc comment, too. Was that __doc__  ?
[19:37:27] <rtyler> list up @Extension?
[19:37:30] <rtyler> you're speaking greek!
[19:37:42] <kohsuke> We need to find classes that has @Extension annotation on it
[19:37:50] <rtyler> why?
[19:38:01] <kohsuke> that's how Hudson finds what extensions a given plugin implements
[19:38:35] <rtyler> hate to sound repetitious, why? :P
[19:38:59] <abayer> Because that's how Hudson knows what the hell a plugin does. =)
[19:39:29] <rtyler> right, but why does the scripting language need to know?
[19:39:48] <kohsuke> It's Hudson that needs to know
[19:39:57] <abayer> Well, one would assume you want to actually do something in Hudson with Jython, yes?
[19:39:58] <rtyler> oh oh oh
[19:40:08] <rtyler> you're talking about writing a plugin in Jython
[19:40:28] <kohsuke> Yeah. OK, so that wasn't what you are talking about,  I assume.
[19:40:30] <rtyler> I was talking about scripting build steps, or other portions with Jython
[19:40:39] <kohsuke> OK, yeah, that's very very easy.
[19:40:43] * rtyler nods
[19:40:50] <kohsuke> And I see that Python has decorators --- PEP-318
[19:40:58] <kohsuke> So we can let plugins be written entirely in Python.
[19:41:02] <rtyler> only function decorators
[19:41:08] <rtyler> class decorators come in 2.6
[19:41:09] <kohsuke> not on class?
[19:41:11] <rtyler> which Jython doesn't support
[19:41:12] <kohsuke> oh
[19:41:54] <rtyler> would it not also work to have a ScriptingPluginRegistrationCenter sort of class
[19:42:09] <rtyler> and just have the plugin, once instantiated, call .getInstance().register() hooks?
[19:42:15] <rtyler> instead of using annotations?
[19:42:31] <kohsuke> there are other things annotations are used for, like @DataBoundConstructor
[19:42:40] <kohsuke> oh, but that's on functions
[19:48:45] *** khmarbaise has joined #hudson
[19:55:09] *** alexlod_ has quit IRC
[19:59:15] *** davidstrauss has quit IRC
[19:59:37] *** Lewisham has joined #hudson
[19:59:58] *** Lewisham has quit IRC
[20:31:08] <mrooney> I sure do love Python
[20:31:44] <rtyler> python loves you back
[20:50:07] *** keshureddyp has quit IRC
[20:56:12] *** alexlod__ has quit IRC
[21:01:45] *** alexlod has joined #hudson
[21:12:51] *** jamesvecore has joined #hudson
[21:14:38] <jamesvecore> is this that place to ask a question about an error I'm getting?
[21:14:54] <jamesvecore> or is this just for development?
[21:16:32] <rtyler> try us ;)
[21:17:27] <jamesvecore> k I'm using version 1.315 and I'm just trying to get a simple windows batch script to run. I get the following exception just trying to run "dir"
[21:17:47] *** jamesvecore has quit IRC
[21:18:44] *** jamesvecore has joined #hudson
[21:19:20] <jamesvecore> ok I got booted for flooding, but the exception is: java.lang.StringIndexOutOfBoundsException: String index out of range: -1
[21:19:39] <jamesvecore> trying to run a batch script that just does a dir for now
[21:20:24] <jamesvecore> top two lines of the call stack:
[21:20:34] <jamesvecore> at java.lang.String.substring(Unknown Source)
[21:20:41] <jamesvecore> at org.jvnet.winp.WinProcess.parseCmdLineAndEnvVars(WinProcess.java:136)
[21:21:22] <jamesvecore> any ideas? I have no problem running ant builds and python shells
[21:22:42] *** keshureddyp has joined #hudson
[21:23:31] <jamesvecore> I should note, that this is running on a windows slave node, and seems to work fine locally on the master
[21:23:53] <jamesvecore> but when run on the slave I get the error above
[21:24:18] <rtyler> slashes fubared?
[21:24:48] <jamesvecore> no slashes in my script, my only command is "dir"
[21:25:17] <jamesvecore> and it runs successfully
[21:25:33] <jamesvecore> but something in the clean up of the process exiting seems to be throwing that java error
[21:31:31] *** ircuser999 has joined #hudson
[21:35:29] *** BigAllan has joined #hudson
[21:42:24] *** Lewisham has joined #hudson
[21:45:04] *** waz has joined #hudson
[21:53:27] *** admc has joined #hudson
[21:55:30] *** brent has quit IRC
[21:55:43] <jamesvecore> no takers on my issue? where do I file bug reports?
[22:02:08] <Lewisham> try putting your stuff in Pastie
[22:02:14] <Lewisham> so you don't get kicked
[22:03:25] <rtyler> heh
[22:03:29] <jamesvecore> k
[22:03:30] <rtyler> he wasn't in danger of that
[22:03:43] <jamesvecore> I did get kick with orginal post
[22:04:34] <jamesvecore> http://hudson.pastebin.com/m12483719
[22:05:03] <jamesvecore> thats the full output from a windows batch job for "dir" executing on a remote slave
[22:05:18] *** Lewisham has quit IRC
[22:06:48] <jamesvecore> a "dir" batch job works fine if its set to execute on the local master
[22:07:04] <jamesvecore> the slave is running as windows service under a domain account
[22:07:18] <rtyler> wonder if something's pwning permissions
[22:08:10] <jamesvecore> my actual bat files do get executed when I try a more complex job, but the clean up of the process is java land is what seems to be failing
[22:08:27] <jamesvecore> err "in java land"
[22:09:04] <jamesvecore> which cause the job to fail even through the bat file was executed and succedded
[22:09:22] *** khmarbaise_ has joined #hudson
[22:11:10] *** Lewisham has joined #hudson
[22:15:09] <jamesvecore> any ideas? should I post the problem in the issue tracker?
[22:16:13] *** khmarbaise has quit IRC
[22:18:46] * Lewisham has nothing :/
[22:27:27] *** ircuser999 has quit IRC
[22:32:41] <Lewisham> kohsuke: What's the recommended way to compare FilePaths? == isn't working, nor .equals(). The only success I have had is converting both arguments to URIs and comparing those, but it seems dodgy
[22:37:35] <abayer> Hmm. Yeah, there doesn't seem to be a comparison method. If you open an issue for that, I'll see if I can get that added.
[22:41:51] <abayer> It should be as simple as verifying that fp1.getRemote()==fp2.getRemote(), and fp1.getChannel()==fp2.getChannel().
[22:44:06] *** velo has quit IRC
[22:52:45] <Lewisham> oh, right
[22:52:52] <Lewisham> so I should do that instead of converting to URIs?
[22:52:55] <abayer> But a convenience method couldn't hurt.
[22:52:58] <Lewisham> the URI conversion makes me not confident
[22:53:12] <abayer> I'd go with what I suggested - it's more accurate.
[22:53:13] <Lewisham> I have a function in a plugin I'm writing that copies slave data to the master for processing if need be
[22:53:26] <Lewisham> then attempts to delete that temp folder afterwards
[22:53:33] <abayer> You could have two FilePaths pointing to the same URI but via different channels, for instance.
[22:53:34] <Lewisham> but I'm leaning towards disabling the deletion for safety
[22:53:40] <Lewisham> ack
[22:53:40] <Lewisham> OK
[22:53:46] <abayer> I mean, you probably won't.
[22:53:50] <abayer> But you *could*.
[22:54:06] <Lewisham> anything that could lead to someone's workspace being deleted means I should worry about it :)
[22:54:11] <abayer> =)
[22:54:12] <Lewisham> I think I might disable the functionality
[22:54:20] <Lewisham> and just rely on the OS to clean up it's temp dir
[22:54:28] <abayer> I tend towards that.
[22:54:31] <Lewisham> and place a note in the release notes
[22:55:25] <Lewisham> and put a little warning in the output
[22:55:31] <Lewisham> that should do
[22:55:51] <Lewisham> then I'll just wait for the first person to file a bug report saying their temp directory filled the HD because they were building 900 times a day :)
[22:55:56] <abayer> heehee
[22:56:16] <abayer> If it wasn't for the users, development would be so much simpler!
[22:56:47] <kohsuke> Lweisham: FilePath should already have the equals method. It's a relatively new addition
[22:57:08] <abayer> Very new - it's not in my local source copy, which is like three days old!
[22:58:01] <Lewisham> I'm building against 1.315, and it doesn't seem to work
[22:58:24] <kohsuke> OK, somehow that was put into a branch
[22:58:37] <kohsuke> I'll bring push the change to the trunk, too
[23:00:36] *** alexlod_ has joined #hudson
[23:04:44] <Lewisham> awesome
[23:06:50] *** khmarbaise_ has quit IRC
[23:07:41] <Lewisham> does it work like abayer described?
[23:07:59] <Lewisham> I'll implement it as he said right now, then move to the actual function when 1.316 hits
[23:12:47] *** jieryn-w has quit IRC
[23:37:39] <mrooney> kohsuke: oh, false alarm regarding a successful plugin release, apparently that was only the release:prepare task
[23:37:57] <kohsuke> that explains
[23:38:26] <mrooney> now I'm getting "Error deploying artifact: Unsupported Protocol: 'java-net': Cannot find wagon which supports the requested protocol: java-net" which I remember I was stuck at before
[23:41:19] <mrooney> kohsuke: is that a common problem? have I missed one of the many links littered throughout various wiki pages explaining a step? :)
[23:41:40] <kohsuke> What was your plugin?
[23:41:44] <kohsuke> Your POM is likely broken a bit
[23:44:16] <mrooney> ncover
[23:44:48] <kohsuke> Hmm.
[23:44:56] <mrooney> see any issues?
[23:44:59] <kohsuke> Does "mvn help:effective-pom" lists wagon-svn in <build>/<extension> ?
[23:46:32] <mrooney> it is running now, let me check
[23:47:40] <mrooney> kohsuke: I see "urls[1] = file:/home/michael/.m2/repository/org/jvnet/wagon-svn/wagon-svn/1.8/wagon-svn-1.8.jar"
[23:48:17] <kohsuke> I don't know why your Maven is failing to use it
[23:53:19] <mrooney> kohsuke: here is the full release:perform output http://pastebin.com/d3fbaaed5
[23:53:23] <mrooney> if you happen to have any other ideas
[23:54:53] <kohsuke> I might try "rm -rf /home/michael/.m2/repository/org/jvnet/wagon-svn"
[23:54:58] <kohsuke> but other than that I really don't know.
[23:55:00] *** jamesvecore has left #hudson
[23:55:14] <Lewisham> abayer: Your FilePath comparison didn't work :/
[23:55:21] <abayer> d'oh.
[23:55:31] <kohsuke> I'd need a debugger to understand the root cause.
[23:55:32] <abayer> Good thing Kohsuke implemented it differently, then. =)
[23:55:43] *** alexlod_ has quit IRC
[23:55:43] *** alexlod has quit IRC
[23:55:45] <kohsuke> I'm curious why you want an equality test
[23:55:46] <Lewisham> oh, is it somewhere I can see it?
[23:55:58] *** alexlod_ has joined #hudson
[23:55:59] *** alexlod__ has joined #hudson
[23:56:02] <mrooney> kohsuke: hm still the same error. Would you be able to check it out and release it for me? I'm not sure what else I can do
[23:56:18] <mrooney> I've hit this same thing on two different machines
[23:56:24] <kohsuke> I think you need to send me your release.properties file
[23:56:30] <abayer> Lewis: http://bit.ly/qEBj4
[23:56:37] *** alexlod__ has left #hudson
[23:56:52] <kohsuke> How did you do your past releases?
[23:57:51] <Lewisham> kohsuke: If you remember, I said that JDepend is not nice to work with, and so sending the JDepend computation to a slave isn't viable. I'm thus copying data to the master if need be. In order to figure out at the end of the method if I did a copy, I'd like to do a FilePath comparison, otherwise I can refactor the method to return a tuple with a truth value and the FilePath if it did the copy...
[23:59:28] <kohsuke> It sounds like you just need to check if the build is locally done or remotely done
[23:59:44] <Lewisham> what I do is check if the workspace is remote
[23:59:44] <kohsuke> but anyway, thanks for the clarification.
[23:59:47] <Lewisham> and then copy it if it is
[23:59:59] <Lewisham> but then I want to check if the method really did that copy or not later on

top