[12:49] *** Mindreef has joined #ApacheAxis [12:49] *** Mindreef is now known as MarkEricson [12:53] *** Russell-lunch is now known as RussellButek [13:00] *** Glen-confcall is now known as GlenDaniels [13:00] <GlenDaniels> DONG [13:00] <tjordahl> Ding [13:00] <GlenDaniels> clock on the wall sez 1PM [13:01] <GlenDaniels> Let's wait a minute for stragglers, then begin? [13:02] <RichScheu> Here. [13:02] <RichScheu> Russell ? [13:03] <AlanGordie> Hello all [13:03] <RichScheu> Hello [13:03] <RussellButek> yo [13:04] *** rickr-badconnection has quit IRC (Ping timeout) [13:04] <GlenDaniels> OK - shall we? [13:04] <GlenDaniels> Do we want to dive right into the bugzilla list, or talk a bit first? [13:04] <RussellButek> other topics: TODO.txt, attachment sample. Anything else? [13:05] <RichScheu> Change compile debug default to on. [13:05] <GlenDaniels> Hehehe [13:05] <GlenDaniels> That's gone back 'n forth like three times now [13:06] <RichScheu> It helps with bugzilla stack traces... [13:06] <RussellButek> It's become a real nuisance having axis-user problems reported without line numbers. [13:06] <GlenDaniels> During the beta, I'm fine with debug being the default [13:06] <RussellButek> I don't see a problem with it being the default. [13:06] <RussellButek> Even beta 2 could be built in debug. [13:06] <GlenDaniels> +1 [13:06] <GlenDaniels> But when we go 1.0 I think we shouldn't. [13:07] <RussellButek> Certainly when we go 1.0 we should turn it off. [13:07] <GlenDaniels> hehehe [13:07] <GlenDaniels> jinx [13:07] <tjordahl> Makes things bigger, but I guess it doesn't affect the running code like C/C++ -g [13:07] <RichScheu> Okay, I'll do that right now. +1 to switching at 1.0 [13:07] <RussellButek> Good. [13:07] <tjordahl> THAT was easy. :-) [13:07] <RussellButek> We lost Rick, and he's the one we want for the attachment sample discussion, so lets go to bugzilla. [13:07] <GlenDaniels> What, like any decisions we make AREN'T easy, Tom? :) [13:08] <GlenDaniels> Bugzilla! Bugzilla! [13:08] <GlenDaniels> 7373 is a dup of 6872, I think, which is mine [13:08] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7373 [13:09] <GlenDaniels> HTTPSender isn't doing HTTPS proxies right, which I'll try to fix today. [13:09] <RussellButek> OK. Does 6872 go into beta 2? [13:09] <GlenDaniels> I think it should [13:10] <GlenDaniels> +1s, -1s? [13:10] <RussellButek> (I'm the release manager - all my opinions will be "NO", so I'll stay out of voting) [13:10] <RussellButek> Did https ever work? [13:10] *** Glyn has joined #ApacheAXIS [13:10] <RussellButek> Hi, Glyn! We're going through bugzilla. Just started. [13:10] <GlenDaniels> I think so, but not through proxies, apparently [13:11] <RichScheu> Will this hold up the release ? [13:11] <RussellButek> If it used to work but doesn't, that's a real good candidate for beta 2 even WITH my release manager hat on. [13:11] <Glyn> Russell: Hi! Thanks. Got delayed sorting out children. [13:11] <GlenDaniels> :) Glyn [13:11] <tjordahl> It doesn't sound like it every worked... [13:11] <RussellButek> Glyn, we're talking about 6872. [13:11] <Glyn> Russell: k [13:11] <GlenDaniels> and 7373 [13:12] <tjordahl> Glen, would you use the code provided? [13:12] <GlenDaniels> Yup, looks pretty straightforward. [13:12] <RichScheu> Russell: I'm making a list of the bugzilla defects and status as we are discussing them...will post to axis-dev after chat [13:13] <RussellButek> Rich, I'm doing the same! If you wanna do it, that's fine. Frees up some of my ink. [13:13] <tjordahl> I say do it, the worst that will happen is Glen break proxy stuff which is already not working. :-) [13:13] <GlenDaniels> I can't really test this is the problem, in that I don't know of a secure proxy [13:13] <GlenDaniels> But I think putting in the fix and making sure nothing breaks should be sufficient [13:14] <GlenDaniels> Then the bug submitters can test [13:14] <tjordahl> Dims added some proxy code the WSDL2Java, maybe we can get him to test. [13:14] <GlenDaniels> sure [13:14] <RussellButek> Personally, I agree (if I don't wear my rm hat). As long as nothing else breaks. [13:14] <GlenDaniels> Russell, I think you need to agree WHILE wearing the RM hat.... [13:14] <RussellButek> Let me cock it a bit... [13:14] <RussellButek> <rm> agree </rm> [13:14] <GlenDaniels> ok. :) [13:14] <GlenDaniels> 7373/6872 : FIX FOR BETA2 [13:15] <GlenDaniels> 7407 is next in the all-axis query [13:15] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7407 [13:15] <GlenDaniels> Including wsdl2java/java2wsdl ant tasks [13:15] <RussellButek> Huh? I get 7132. [13:15] <tjordahl> Mee too. 7132 doesn't sound like a bug to me. [13:15] <RussellButek> How many bugs are in your list? I have 16. [13:16] <GlenDaniels> 15 now [13:16] <GlenDaniels> I'm reading top down 7373, 7407, 7469, 7722.... [13:16] <RussellButek> 7132 probably isn't a bug. Not for beta 2. [13:16] <GlenDaniels> do you see a different order? [13:16] <tjordahl> Anyway, I don't think it is a bug at all. Aren't xmlns attributes not really attributes? [13:16] <GlenDaniels> OK, I just ordered by ID [13:17] <RussellButek> Mine are ordered by number. [13:17] <RussellButek> Tom, correct. [13:17] <Glyn> Regardless of whether it's a bug, 7132 is really pending more information from the raiser. [13:17] <tjordahl> So we should cancel this one. Why wait? [13:17] <GlenDaniels> No, it's a bug, I think. [13:17] <GlenDaniels> It's due to the special-casing we do for namespace declarations [13:17] <tjordahl> ?? [13:17] <RussellButek> 7267? The severity is "minor" [13:18] <GlenDaniels> The solution is a getAllAttributes() method or something like that [13:18] <tjordahl> Oh, WE do the special case. In that case. [13:18] <GlenDaniels> right [13:18] <tjordahl> I thought that XML parser would not pass xmlns as an attribute [13:18] <GlenDaniels> nope [13:18] <GlenDaniels> it does [13:18] <GlenDaniels> we eat it [13:18] *** rickr-badconnection has joined #ApacheAxis [13:19] <RichScheu> 7132 NOT BETA2 [13:19] <RussellButek> agreed. [13:19] <tjordahl> 7267 I would say not for B2 either [13:19] <GlenDaniels> ok by me. [13:19] <GlenDaniels> Can we go one at a time pleasE? [13:19] <tjordahl> as Glen said "We've got big problems" [13:20] <tjordahl> We are, its just that Russell and I are moving to the next one. [13:20] <GlenDaniels> Let's discuss / vote / resolve each one first, though. [13:20] <RussellButek> So it's agreed that 7132 is not for beta 2? [13:20] <tjordahl> [13:18] <RichScheu> 7132 NOT BETA2 [13:20] <tjordahl> Try and keep up :-) [13:20] <GlenDaniels> That was Rich jumping the gun [13:20] <GlenDaniels> did we vote? [13:20] <RussellButek> +1 for 7132 NOT in beta 2. [13:20] <GlenDaniels> (no offense, Rich, I just want things to be clear) [13:21] <tjordahl> +1 for 7132 NOT in B2 [13:21] <GlenDaniels> +1 [13:21] <Glyn> +1 for 7132 NOT in beta 2 [13:21] <RussellButek> Any -1? [13:21] <RussellButek> going... [13:21] <RussellButek> going... [13:21] <GlenDaniels> ok ok :) [13:21] <RussellButek> gone. [13:21] <RussellButek> 7267. It's a severity of "minor". [13:21] <GlenDaniels> I propose not for beta2, but soon thereafter [13:22] <RussellButek> +1 for 7267 NOT in beta 2. [13:22] <Glyn> (Since I commented on 7132, I'll comment on it again that we accept the bug but not for beta 2.) [13:22] <GlenDaniels> Also, the larger issue is, I think, major. [13:22] <GlenDaniels> i.e. not dealing correctly with different schemas [13:22] <RichScheu> +1 7267 NOT BETA2 [13:22] <tjordahl> +1 for a defer. Can we fix this minor issue and open a new item for the whole shebang? [13:22] <GlenDaniels> We currently give lip service to 1999 schema support but don't really do it [13:22] <Glyn> +1 7267 NOT beta 2 [13:23] <RussellButek> Any -1 for 7267 not in beta 2? [13:23] <GlenDaniels> Tom: Let's leave it but open a new one anyway. [13:23] <RussellButek> going ... [13:23] <RussellButek> goingGone [13:23] <GlenDaniels> 7373 is a dup of 6872 [13:23] <tjordahl> 7373 dup? [13:23] <GlenDaniels> Will fix for beta [13:24] <RussellButek> yes. [13:24] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7407 [13:24] <GlenDaniels> ant tasks [13:24] <RichScheu> Defer ant tasks until after beta 2 [13:24] <tjordahl> +1 to defer [13:24] <Glyn> +1 [13:24] <GlenDaniels> +0 [13:24] <RussellButek> +1 to defer. [13:24] <RussellButek> 7407 NOT in beta 2. [13:25] <RichScheu> goingGone [13:25] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7469 [13:25] <tjordahl> Anyone have a clue about 7469? I started looking in to it but I don't see what the problem is. [13:25] <GlenDaniels> Looks like a classpath issue? [13:25] <GlenDaniels> parser jar problems [13:26] <tjordahl> Class Cast exception.... [13:26] <tjordahl> Cancel it? [13:26] <GlenDaniels> I don't think we need do anything about this for beta-2, but we should assign it to someone to work on with the submitter. [13:26] <RussellButek> I don't even see AXIS in the stack trace. [13:26] <RussellButek> +1 7469 NOT in beta 2. [13:26] <Glyn> +1 7469 NOT in beta 2 [13:26] <RichScheu> +1 and +1 to cancel [13:27] <tjordahl> Should we pursue this kind of bug? +1 to cancel... [13:27] <RichScheu> Send to level 3 support... [13:28] <GlenDaniels> Yes we should pursue it, at least to point them to axis-user and classpath as the likely culprit. [13:28] <RussellButek> Any volunteers? [13:28] <tjordahl> I think Glen just did, right? :-) [13:28] <GlenDaniels> Sure. [13:28] <tjordahl> hehe [13:29] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7718 [13:29] <RussellButek> OK. We're done with 7469. 7718? [13:29] <Glyn> Well volunteered Glen1 There *may* also be some advice we can add to the docs to avoid it cropping up frequently. [13:29] <tjordahl> 7718 I just assigned to myself. Looks like a reasonable addition with low risk - just doc impact. [13:29] <GlenDaniels> Two line fix, seems OK to me. [13:29] <RichScheu> But do we have an allowedInheritedMethods option ? [13:30] <RichScheu> oh is this a wsdd parameter thingee. [13:30] <GlenDaniels> right [13:30] <RussellButek> If it's just a doc fix, +1 for beta 2. [13:30] <GlenDaniels> +1 [13:30] <RussellButek> Do we need a test for it? [13:30] <tjordahl> No code fix [13:30] <GlenDaniels> It's both [13:30] <tjordahl> No, there is a code fix I mean to say [13:30] <GlenDaniels> right [13:31] <GlenDaniels> It's a two line code fix and a doc fix and we should add a test, yes. [13:31] <Glyn> What effect would the fix have on the externals? A 'static' option or an option on the ?wsdl invocation? [13:31] <AlanGordie> can we also add a "published" parameter as well, not necessarily for beta 2, but we need a way to determine if we should publish metadata (via gui, WSIL, etc) for each service [13:32] <RussellButek> If there's a code change I vote -0. [13:32] <GlenDaniels> It's a service option, Glyn, so it's in the WSDD. [13:32] <GlenDaniels> This allows the deployer to specify it so the ?wsdl will pick it up [13:32] <Glyn> ok. -0 from me [13:33] <GlenDaniels> The ServiceDesc should notice this as well and not gen operationDescs if its set, but that can happen later. [13:33] <RussellButek> I lost count. Do we have 2 -0 and 2 +1? [13:33] <RichScheu> -0 [13:34] <GlenDaniels> Any -1s? [13:34] <RichScheu> 7718 Owner=tom Beta2=maybe [13:34] <RussellButek> No -1's. [13:34] <RussellButek> (it APPEARS innocuous) [13:34] <tjordahl> ok, lets leave it out. -1 for Beta 2 [13:34] <GlenDaniels> Can we tack RESOLVE: onto the resolution lines, btw, just so reading the log is easier? [13:35] <GlenDaniels> And Russell is the only one who can type those, I think. [13:35] <RussellButek> What can I do? [13:35] <GlenDaniels> So now you're -1, Tom? [13:35] <GlenDaniels> i.e. "RESOLVE: 7718 - MAYBE BETA2, OWNER=Tom [13:35] <Glyn> Also, is someone going to update each defect to say whether or not it will be fixed in beta 2. Nice for raisers to get some feedback. I'll do it based on a summary posted to axis-dev if no-one else can be bothered. [13:36] <tjordahl> I don't think its that important to worry about it for B2 [13:36] <RussellButek> OK. So 7718 NOT in beta 2. [13:36] <GlenDaniels> ok [13:36] <RussellButek> agreed? [13:36] <tjordahl> +1 7718 NOT in beta 2 [13:37] <RussellButek> OK. 7722? [13:37] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7722 [13:37] <RichScheu> Cancel 7722. The code has changed and there are not attachments. [13:37] <GlenDaniels> Um - we should look into it at least, IMHO. [13:38] <GlenDaniels> Do we know this works? [13:38] <RussellButek> Give the opener a few days to respond. [13:38] <tjordahl> Rich, this is a ClassRep bug. [13:38] <GlenDaniels> I guess there's the inheritance test.... [13:38] <GlenDaniels> We don't use ClassRep there anymore, Tom. [13:38] <RichScheu> ClassRep is not used anymore [13:38] <tjordahl> We should extend the test to do more interfaces. [13:38] <RussellButek> +1 extending the test. [13:38] <tjordahl> Then we would know if this (and the other ClassRep bug) is still a problem. [13:38] <RussellButek> So, +1 7722 NOT in beta 2. [13:38] <GlenDaniels> I need to finish merging in the ClassRep stuff so I can clean up the code. [13:39] <Glyn> +1 7722 NOT in beta 2 [13:39] <GlenDaniels> ACTION: Glen to finish ClassRep/metadata merge and clean up wsdl.fromJava code [13:40] <Glyn> (Is that a post-beta 2 action?) [13:40] <RussellButek> (I hope so!) [13:40] <GlenDaniels> I'd prefer to do it pre, but whatever. [13:40] <Glyn> Why not just update the TODO list pre to get the item on the books. [13:40] <GlenDaniels> Releasing code we *know* is going to change seems a little weird. [13:41] <Glyn> LOL! Most code changes eventually. [13:41] <GlenDaniels> Yes but this is more like code that is in the middle of changing. [13:42] <RichScheu> Usually we try not to make changes hours before a release. [13:42] <Glyn> Depends how dictatorial the release manager is feeling ;-) [13:42] <RussellButek> This stuff doesn't seem like just a couple hours work. When do you expect beta 2 to go out? [13:42] <RussellButek> I'm hoping for this week. [13:42] <GlenDaniels> Did we decide on an actual date? [13:42] <tjordahl> Question: what is our target release date? [13:42] <GlenDaniels> hehehe [13:42] <Glyn> (See!) [13:42] <GlenDaniels> So Friday? [13:42] <RussellButek> We tentatively said this week last week. [13:42] <RichScheu> Which Friday :-) [13:42] <GlenDaniels> This Friday :) [13:42] <RussellButek> I don't like releasing things on Fridays. [13:43] <RussellButek> Either Thursday or Monday. [13:43] <GlenDaniels> OK - I vote Monday, and I would like to get this change in. [13:43] <RussellButek> (if we say Thursday it'll probably be Friday anyway!) [13:43] <RichScheu> +1 Monday...but lets refrain from pushing lots of stuff in over the weekend. :-) [13:44] <RussellButek> We're talking 7722, Glen? [13:44] <RussellButek> I'll vote -0. [13:44] <GlenDaniels> Not 7722 specifically, just the cleanup/merge which is already in progress. [13:44] <RussellButek> does it subsume 7722? [13:44] <GlenDaniels> Get rid of ClassRep after having merged its functionality. [13:44] <GlenDaniels> Probably - we don't know if 7722 is a real issue [13:45] <GlenDaniels> I think interfaces work, and they were probably fixed after April 3, so this is likely fixed. [13:45] <RussellButek> Basically tell the guy to try again once you do your work, right? [13:45] <GlenDaniels> I think it's fixed now, but we should test to make sure. [13:45] <GlenDaniels> I'll take this one. [13:45] <RussellButek> Oh, Rich already did that. [13:45] <tjordahl> BTW, It looks like the inheritence case already covers interfaces [13:45] <RussellButek> So what's the vote for Glen's cleanup? -0 from me, +1 from Glen. [13:46] <tjordahl> +1 on Glen's cleanup [13:46] <RichScheu> +1 from me iff he can get it done by Thurs so I can review it. [13:46] <GlenDaniels> will do my best, Rich. [13:46] <Glyn> If cleanup means no new or changed externals, +0. Otherwise -1. [13:46] <GlenDaniels> Should be doable. [13:46] <RichScheu> +1 then [13:46] <RussellButek> OK. Go for it, Glen. Back to bugzilla. 8001. [13:47] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8001 [13:47] <GlenDaniels> Minor issue, we should fix it but not for beta-2 [13:47] <RussellButek> +1 8001 NOT for beta 2. [13:47] <Glyn> +1 8001 NOT for beta 2. [13:48] <RichScheu> +1 8001 NOT FOR BETA 2 [13:48] <tjordahl> If its minor shouldn't we just fix it? [13:48] <RussellButek> Gone! [13:48] <RussellButek> Not for beta 2, Tom. [13:48] <tjordahl> huh. [13:49] <RichScheu> ouch [13:49] <GlenDaniels> fixing it involves checking for the dir when we need it, which could be a little complicated. Post-beta-2. [13:49] <tjordahl> We have gotten lots of reports about this attachements directories. People complain about it [13:49] <RussellButek> Minor in this context means it's not that important. [13:49] <RichScheu> The RM has spoken...we tremble. [13:49] <Glyn> (Russell's got this release manager role off to a tee. Pointy hair and all.) [13:49] <GlenDaniels> hm. Well, how about this. If it gets fixed by Friday, then great. [13:50] <RussellButek> You're not worried that someone might actually be depending on the current behavior? [13:50] <RussellButek> We'd regress that poor soul. [13:50] <GlenDaniels> I.e. if someone wants to fix and test it, and everything works, I for one will be psyched it's fixed. [13:50] <GlenDaniels> How could this break anyone? [13:50] <GlenDaniels> The fix is to just create the attachments directory when you need it, not before. [13:51] <RussellButek> Someone might be assuming that directory WILL exist. Always. Why, I haven't a clue. I'm just wearing that pointy haired hat. [13:51] <GlenDaniels> If anyone relies on the dir being there, they're so deep into our whitebox that they should be prepared for breakage at any point. [13:51] <Glyn> And the fix shifts the point at which we detect we can't create an attachments directory in some circumstances. Possible problem lurking there. [13:52] <RussellButek> Glen/Tom are you voting -1 for "8001 NOT for beta 2"? [13:52] <GlenDaniels> I don't think that's a big deal, Glyn. [13:52] <GlenDaniels> Yes, I think it should be simply left pending. [13:52] <GlenDaniels> If it gets fixed in time, great. But it's not a must-fix. [13:52] <Glyn> Me neither. Just think we should defer fixes we can reasonably defer and then fix it ASAP after beta 2. [13:53] *** rubys has joined #ApacheAXIS [13:53] <Glyn> Once we get out of the "cut and run" frame of mind, the date will slip and slip. [13:53] <GlenDaniels> I'm just saying if someone has the time and inclination to fix it, they should go for it and not feel restricted. [13:53] *** Disconnected Session Close: Tue Apr 23 13:53:51 2002
Session Start: Tue Apr 23 13:54:49 2002 [13:54] *** Now talking in #ApacheAxis [13:54] *** Topic is 'Free-form conversation for Axis (http://xml.apache.org/axis) developers.' [13:54] *** Set by GlenDaniels on Wed May 23 13:55:17 [13:54] <RussellButek> Hi, Sam! We're going over the bugzilla list deciding what goes into beta 2. [13:54] <GlenDaniels> sorry, lost connection [13:55] <RussellButek> Did you see my last comment, Glen? [13:55] <GlenDaniels> Probably not [13:55] <RussellButek> [Tue 12:53] <RussellButek> <RM> ONLY do stuff for beta 2 that we agree on. DON'T say 'maybe'. Opens up too many doors. </RM> [13:55] <rubys> When is beta 2 targetted for? [13:56] <GlenDaniels> Monday release [13:56] <tjordahl> last check-ins Friday I would say. [13:56] <RussellButek> +1, Tom. [13:57] <RussellButek> Which means I COULD release beta 2 during the weekend, but I'll be out of town. [13:57] <rubys> How about a RC on Friday? [13:57] <GlenDaniels> I think we should have the following priorities : 1) MUST fix for beta-2 (showstopper), 2) MAY fix for beta-2 (if done and tested by Friday AM, let's say), 3) DEFER (definitely not in beta-2), 4) PUNT (not going to fix / not an issue) [13:57] <GlenDaniels> I think this falls in the (2) category [13:58] *** rubys is now known as SamRuby [13:58] <RussellButek> I prefer yes/no for beta 2. [13:58] <RichScheu> I disagree. We should not think about it for beta 2...and concentrate on other beta 2 issues. [13:58] <RichScheu> +1 Russell yes/no for beta 2 [13:59] <RussellButek> There's a lot of documentation work that we could be doing. So if a bug fix isn't required, then spend time on docs instead. [13:59] <GlenDaniels> ok, whatever. I'm not married to this, just like the flexibility. Onwards! [14:00] <RussellButek> 8001 is out for beta 2. 8296? [14:00] <RussellButek> This doesn't look like a bug. [14:00] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8296 [14:00] <RussellButek> M Shue has an issue with XML itself. [14:00] <RichScheu> 8296 appears to be a user problem to me. I have looked at this already [14:01] <GlenDaniels> So he wants to define a type in the "" namespace? [14:01] <Glyn> +1 to 8296 NOT in beta 2 [14:01] <RussellButek> I'll respond. Essentially, "this is the way XML works." [14:01] <RichScheu> +1 8296 NOT IN BETA 2 [14:02] <GlenDaniels> I don't understand the problem... [14:02] <RichScheu> I can respond to the problem. He doesn't understand prefixing in xml...among other things. [14:02] <GlenDaniels> Ok [14:02] <RussellButek> ok, rich. [14:03] <tjordahl> sounds like a cancel to me. +1 [14:03] <RussellButek> +1. [14:03] <RussellButek> Onward! 8313. [14:03] <RichScheu> +1 to cancel [14:03] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8313 [14:03] <tjordahl> Whats up with this one? [14:04] <GlenDaniels> Is this a JDK1.4 thing? [14:04] *** hreynolds2 has joined #ApacheAxis [14:05] <GlenDaniels> *crickets chirp* [14:05] <RussellButek> Does anyone have JDK1.4 javadocs handy? [14:05] <hreynolds2> Yes [14:05] <RussellButek> If that's not the problem, I don't understand this one. [14:06] <RussellButek> Could you look at java.util.Properties? [14:06] <RussellButek> Does it still have getProperties? [14:06] <RussellButek> I mean getProperty? [14:06] <GlenDaniels> Yes [14:07] <hreynolds2> Yes... [14:07] <hreynolds2> public String getProperty(String key) [14:07] <hreynolds2> public String getProperty(String key, [14:07] <hreynolds2> String defaultValue) [14:07] <GlenDaniels> looks the same to me [14:07] <tjordahl> But we are calling get() [14:07] <GlenDaniels> It looks like we're using it like it was a hashtable, though [14:07] <tjordahl> the fix is to change it to getProperty() I think. [14:08] <Glyn> We are calling keys() too. [14:08] <GlenDaniels> oh it is a Hashtable :) [14:08] <GlenDaniels> I don't understand why this wouldn't work then. [14:08] <RussellButek> Oh. Yeah. Hmmm... I don't understand the problem. We need more input. +1 for 8313 NOT in beta 2. [14:08] <GlenDaniels> Throw it back to the submitter, I think. [14:08] <tjordahl> I assume they wouldn't just change this. [14:08] <RussellButek> I'll take this one on. [14:09] <RussellButek> Glen, right. [14:09] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8377 [14:09] <tjordahl> Hasn't this code been working for everyone else in the work already [14:09] *** NelsonM has joined #apacheaxis [14:09] <GlenDaniels> This should work now, I think - so we're waiting on more info from the submitter? [14:10] <tjordahl> +1 to fixing 8377 if it really still is a bug. It sounds serious (the whole Date/Calendar mishmash) [14:11] <RussellButek> I've been talking with this fellow. [14:11] <RussellButek> I can't reproduce it. If he gives me a concrete example of the failure that I can reproduce it should be fixed for beta 2. [14:11] <RussellButek> If he gives me an example by Friday, that is! [14:12] <Glyn> +1 to Russell's suggestion and deadline [14:12] <GlenDaniels> +1 [14:12] <RichScheu> +1 [14:12] <tjordahl> +1 [14:13] <RussellButek> OK, 8377 for beta 2. Last one, 8395 [14:13] <GlenDaniels> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8419 [14:13] *** AlanGordie is now known as AlanG_grabbed4mtg [14:13] <GlenDaniels> oops, I mean http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8395 [14:14] <tjordahl> Sounds like a .NET compiler problem with the generated code. I can look in to this, I have .NET. [14:14] <RussellButek> Should we look into it for beta 2? [14:15] <GlenDaniels> gotdotnet? [14:15] <tjordahl> I am not a huge fan of responding to 'bugs' like this, since there isn't a bug defined. [14:15] <RussellButek> I took a look at his WSDL and couldn't find any problems. [14:15] <Glyn> (Evening calls. Gotta go. Good work guys!) [14:15] *** Glyn has quit IRC (Quit: ) [14:15] <tjordahl> Its a 'help me' bug. [14:15] <GlenDaniels> But it might also be an "interoperability problem" bug, no? [14:16] <RussellButek> Yes, it might. But it's a pretty basic WSDL file he's got. [14:17] <tjordahl> HE doesn't get his program (with I assume .NET generated files) to compile. [14:17] <tjordahl> HE -> He [14:17] <GlenDaniels> So build the .NET stub, compile it, and if it works set the bug to WORKSFORME... [14:18] <RussellButek> +1 8395 NOT in beta 2. [14:19] <GlenDaniels> Oh noooooooooo - more bugs coming in as we speak! :) [14:19] <RichScheu> +1 8395 NOT IN BETA 2 [14:19] <RussellButek> If it IS an interop problem, it'll take longer than the next couple days to resolve it between us and .NET. [14:20] <RichScheu> Can we continue with the next bug [14:20] <RussellButek> The vote's not complete, yet. What do the rest of you say? [14:20] <GlenDaniels> Just a note folks - if there are serious problems (i.e. if we were actually generating bad WSDL) I would hope we would hold the release to fix them.... [14:20] <GlenDaniels> I don't think this will be the case here, but just wanted to make the point. [14:21] <RussellButek> Glen, I agree. [14:21] <RussellButek> I agree with both your points. We should hold it for certain cases, but I don't think this is one of them. [14:21] <GlenDaniels> cool, yup. [14:21] <tjordahl> +! 8395 NOT in Beta 2 [14:21] <RussellButek> So, barring Tom finding something really nasty, +1 for 8395 not in beta 2? [14:22] <GlenDaniels> right [14:22] <RussellButek> OK. 8395 is not in beta 2. Now look at the new ones: 8419. [14:22] <RichScheu> Just responded to 8419 requesting the wsdl document. [14:22] <tjordahl> 8395 assigned to me [14:22] <tjordahl> Rich, I was just adding the WSDL! [14:22] <GlenDaniels> hehehe [14:22] <GlenDaniels> good timing, gents :) [14:23] <tjordahl> It is there now. This is from one of our QA guys. [14:23] <tjordahl> .NET is producing ucky WSDL... [14:23] <RichScheu> Is DataSet something we should support ? [14:23] <GlenDaniels> Not for beta-2, we should discuss how to deal with it later [14:23] <GlenDaniels> (IMHO) [14:23] <RussellButek> +1 8419 NOT for beta 2. [14:23] <SamRuby> DataSet = ADO? [14:24] <tjordahl> I moved it to enhancement. +1 NOT for B2 (maybe not for 1.0) [14:24] <GlenDaniels> Think so, Sam. [14:24] <RussellButek> ADO? [14:24] <hreynolds2> ActiveX Data Object [14:24] <RussellButek> Gotcha. Thanks. [14:24] <RichScheu> +1 8419 NOT FOR BETA 2 [14:24] <GlenDaniels> Who was that masked man? :) [14:25] <GlenDaniels> (hreynolds, who are you? :)) [14:25] *** rickr-badconnection has quit IRC (Ping timeout) [14:25] <hreynolds2> Harris Reynolds....I am wanting to get involved with Axis [14:25] <GlenDaniels> awesome - welcome aboard! [14:25] <hreynolds2> thx [14:26] <RussellButek> OK. 8419 is NOT in beta 2. 8423? [14:26] <GlenDaniels> Hm. [14:27] <GlenDaniels> AxisClassLoader is no longer used, but their point about the context classloader is correct. [14:27] *** rickr-badconnection has joined #ApacheAxis [14:28] <GlenDaniels> Personally, I think any container which doesn't set the context classloader to something reasonable is broken. [14:28] <RussellButek> A beta 2 candidate? They think it's critical. [14:28] <GlenDaniels> We can talk to them about it. I'll take it. [14:29] <RussellButek> beta 2, though? My vote is -0. [14:29] <tjordahl> ClassLoaders are scary. -0 to changing it for B2. [14:29] <GlenDaniels> If we can work out a solution that gets us working with iPlanet, that would be good. [14:29] <GlenDaniels> Not a showstopper though [14:30] <tjordahl> Glen, is this the same thing as what Clement was talking about the other day? [14:31] <GlenDaniels> Yup [14:31] <GlenDaniels> And I continue to think that the context classloader really should be set to something reasonable. :) [14:32] <tjordahl> Someone posted a response to this message on axis-user, talks about java.endorsed.dirs [14:33] <tjordahl> Subject thread "problem loading class" [14:33] <GlenDaniels> I see it [14:34] <RussellButek> So it's a setup problem? Not an AXIS problem? [14:34] <GlenDaniels> this is a reasonable 1.4 solution as long as you don't have some funky classloader (like Clement) [14:34] <RussellButek> (sounds like a FAQ candidate, to me) [14:34] <GlenDaniels> Classloaders are tricky [14:34] <GlenDaniels> definitely a FAQ [14:34] <GlenDaniels> I'll write it up [14:35] <RussellButek> Thanks, Glen. [14:35] <GlenDaniels> The problem is I'm not sure exactly what the solution should be for iPlanet [14:35] <RichScheu> So what's the verdict on 8423 [14:35] <GlenDaniels> Pending. I'm going to investigate. [14:35] <tjordahl> investigate for B2, add FAQ if possible. [14:35] <GlenDaniels> If we find something we can do for beta-2, we'll consider it. [14:36] <tjordahl> +1 [14:36] <RussellButek> OK. Let's defer the decision on this one for beta 2. Once you have more info, let's get back on this chat. [14:37] <RichScheu> Ok I am going to send out the defect summary to axis-dev [14:37] <GlenDaniels> cool-o, thanks Rich [14:37] <GlenDaniels> I'll post the log [14:37] <RussellButek> 8426? [14:38] <RussellButek> Last one! [14:38] <RichScheu> Darn I just sent the summary :-) [14:39] <GlenDaniels> we need more info [14:39] <GlenDaniels> If he's using DII, maybe he's not setting SOAPAction [14:39] <GlenDaniels> If he's using Stubs, we might have a problem [14:39] <GlenDaniels> (though likely don't) [14:39] <GlenDaniels> Looks like DII to me [14:39] <RussellButek> Done with bugzilla? [14:40] <GlenDaniels> from the stack trace [14:40] <RussellButek> Rick, you there? Want to talk about the attachment sample? [14:40] <GlenDaniels> (I'll brb - bio-break) [14:40] *** rickr-badconnection has quit IRC (Ping timeout) [14:41] <tjordahl> Sylvain just posted a note about the Date Calendar - looks like we can cancel. [14:41] <MarkEricson> Greetings all, I'm new. Watching activities for now, hope to get more involved with Axis development. I met Sam & Glen at the last SoapBuilders. [14:41] *** rickr-badconnection has joined #ApacheAxis [14:42] <rickr-badconnection> Any time for some quick questions? [14:42] <RussellButek> Shoot, Rick. [14:42] <rickr-badconnection> The way you map serializers in WSDD seems to have changed and now you need a operation element to assoc. the parameter to its typeqname. (I think) Is there an example on how to assoc the return? [14:43] <RussellButek> Tom, I see Sylvain's note. I'll cancel. [14:43] <GlenDaniels> "map serializers" means <typeMapping> to me, which hasn't changed. I assume you mean map types to parameters? [14:43] <rickr-badconnection> Yup [14:44] <GlenDaniels> the return type is specified in the <operation>, i.e. <operation name="foo" returnType="xsd:string"...> [14:44] <GlenDaniels> the name of the return element can be specfied as returnQName="myNS:name" [14:44] <rickr-badconnection> Thanks [14:45] <rickr-badconnection> I have a case where for some reason the operationDesc has a method that is null. This gives me a null pointer exceptin at RPCProvider 229 any hints whats gong haywire here? [14:45] <GlenDaniels> Shouldn't ever get an operationdesc with a null method - though perhaps we should bulletproof this a bit tighter. [14:46] <GlenDaniels> You want to send your service class + WSDD? [14:46] <rickr-badconnection> Well I'll keep hunting for why this seems to be the case for me. Probably something I'm doing wrong in the WSDD. [14:47] <RussellButek> This is the attachment sample, right Rick? [14:47] <hreynolds2> I have a quick Q [14:47] <rickr-badconnection> Yup. [14:47] <GlenDaniels> Are you building the <operation>s yourself in the WSDD? [14:48] <rickr-badconnection> Yup. [14:48] <GlenDaniels> I can add some bulletproofing to complain if we can't map an OperationDesc to a method. [14:48] <GlenDaniels> Then you'll know earlier what's wrong [14:48] <GlenDaniels> Send the WSDD + impl file to me and I'll see if I can suss out the problem? [14:49] <GlenDaniels> Shoot, Harris! [14:49] <rickr-badconnection> Curious should old WSDD still work? [14:49] <hreynolds2> What version of Ant to I need to do a clean local build of Axis...I always have problems with the optional.jar file? [14:49] <GlenDaniels> Yes, rick [14:50] <GlenDaniels> Have you installed the optional.jar file in your ant/lib? [14:51] <hreynolds2> Yes [14:51] <GlenDaniels> I'm using Ant 1.4 [14:51] <MarkEricson> I just found I needed to go to the nightly build of Ant because of a bug with Ant and JDK1.4 [14:51] <GlenDaniels> ah - are you using 1.4, Harris? [14:52] <hreynolds2> I have tried a couple versions...1.4 and 1.5alpha [14:52] <hreynolds2> the alpha is a 4/12 build...should I get a later version? [14:53] <GlenDaniels> alas, I dunno [14:54] <MarkEricson> Q: Functional Test failure (posted to Apache-Dev) for some reason test.wsdl.refattr is failing, apparantly name mangling issue. Is this perhaps related to which XML parser is used for build? What is recommended? [14:58] <GlenDaniels> We've tested with crimson and xerces (up to Xerces 2.0) [14:58] <MarkEricson> OK, what is the position on Axis with JDK1.4 ? [14:58] <GlenDaniels> I haven't tested it myself - others? [14:59] <hreynolds2> I have used it for simple examples, but nothing heavy duty [15:00] <MarkEricson> OK... nobody has built with it? Probably my mistake! [15:10] *** MarkEricson has quit IRC (Quit: ) [15:14] *** tjordahl has quit IRC (Ping timeout) [15:15] *** NelsonM has quit IRC (Client closed connection) [15:20] <hreynolds2> Glen...not sure if you are still there...I was able to build axis, but only after I commented out the to condition tasks in the build.xml file [15:32] <hreynolds2> quit [15:32] *** hreynolds2 has left #ApacheAxis [15:41] *** RussellButek is now known as Russell-away
