=== wasabi [~wasabi@c-67-174-81-195.hsd1.tx.comcast.net] has joined #ubuntu-java === oz__ [~oz@c211-30-61-51.belrs1.nsw.optusnet.com.au] has joined #ubuntu-java === oz___ [~oz@c211-30-61-51.belrs1.nsw.optusnet.com.au] has joined #ubuntu-java === oz__ [~oz@c211-30-61-51.belrs1.nsw.optusnet.com.au] has joined #ubuntu-java === neo_ [~neo@82.Red-80-24-160.pooles.rima-tde.net] has joined #ubuntu-java === wasabi_ [~wasabi@207.55.180.100] has joined #ubuntu-java === mgalvin [~mgalvin@host-66-202-95-170.spr.choiceone.net] has joined #ubuntu-java [05:30] hey guys, does tomcat run on top of gcj/gij [05:31] if so will there be a tomcat package for breezy [05:32] None of the Free VMs have a working security manager. [05:32] So the problem is that it *could* work, but you don't have any of the usual safety guarantees of Java. [05:34] ah i c [05:35] thnx [05:44] that's not what's holding it back [05:44] We just need tomcat 5 packages. [05:44] it runs fine (- security manager) [05:44] tomcat4 makes use of com.sun classes which are not in free JVMs [05:49] wasabi_ are there plans for getting tomcat5 packages into breezy? [05:51] I doubt we'll make breezy. [05:51] In fact, we won't. [05:51] Probably breezy +1 though === oz___ [~oz@c211-30-61-51.belrs1.nsw.optusnet.com.au] has joined #ubuntu-java [06:02] wasabi_: Well, I wasn't mentioning the state of the packaging so much as why you shouldn't do it even if you can right now. [06:37] ahh, the security manager really isn't that big of a deal except for ISVs [06:38] How do you figure? [06:38] The security manager protects you from code you yourself choose to run. [06:38] So does running as another user. [06:39] Or... more than likely, you should run code you don't trust anywa.s [06:39] Can tomcat easily run different apps as different users now? Last I checked it was a big monolith. [06:39] Well, that is true. [06:40] But it's not something the security manager is going to help with, except in the case where you are running untrusted software [06:40] (ISV) [06:40] maybe I mean ASP instead. [06:40] (ASP) [06:40] IT people shouodn't trust their vendor's software either. [06:40] I mean, php doesn't even have a CONCEPT of a security manager [06:40] and people get by fine with it [06:40] Or cgi. [06:40] Well, you at least have safe_mode with php [06:41] oh, also what about the firefox java plugin? right now the sun jre 1.5 is required for that. is there an open source alt for this? if not will gjc/gij and jre co-exists peacefully? [06:41] And most ISPs I know don't do cgi anymore because they got tired of getting hacked. [06:41] heh. [06:41] mgalvin: gcjwebplugin - see above warnings. [06:41] yeah the security manager matters big time for the web plugin [06:42] thanks guys! [06:43] Don't get me wrong, I'm not trying to downplay the security maanger for tomcat. [06:43] It's just not what's holding us back from packaging it. [06:43] Right. We're talking different sides of the problem. [06:43] avdyk has tomcat 5 packages in experimental now I believe. [06:43] But they don't run. [06:43] They need some work. [06:43] I haven't mentioned packages anywhere here. =) === doko [~doko@a130-233-5-210.debconf5.hut.fi] has joined #ubuntu-java [10:46] https://wiki.ubuntu.com/EclipsePlatform [10:46] I need to relearn wiki formatting and stuff unfortunatly. === jbailey looks [10:47] Just trying to shine some light on the pieces, where they live. [10:47] Eventually I will get packaging policy up [10:47] For plugins and other components. [10:47] I spent a bit of time this morning and will spend more later looking at doing the jar splitting on ppc [10:47] Jar splitting? [10:47] Whatcha mean? [10:47] What needs to be split? [10:56] To work around a ppc bug with gcj [10:57] Their aot-compile-rpc will split the jars automatically for compiling on pppc [10:57] s/rpc/rpm [11:07] what's wrong with it on ppc? [11:07] Is it just that one of the jars don't work on ppc? [11:08] Could just exclude it. I've done that to a few that don't work. No biggy. [11:13] It's in your backlog on #gcj from a couple days ago. There's a bug in gcc. [11:13] One of the RTL passes doesn't do what it should do.