From svn-return-4259-apmail-forrest-svn-archive=forrest.apache.org@forrest.apache.org Sat Nov 12 00:27:36 2005 Return-Path: Delivered-To: apmail-forrest-svn-archive@www.apache.org Received: (qmail 48875 invoked from network); 12 Nov 2005 00:27:35 -0000 Received: from hermes.apache.org (HELO mail.apache.org) (209.237.227.199) by minotaur.apache.org with SMTP; 12 Nov 2005 00:27:35 -0000 Received: (qmail 23952 invoked by uid 500); 12 Nov 2005 00:27:34 -0000 Delivered-To: apmail-forrest-svn-archive@forrest.apache.org Received: (qmail 23892 invoked by uid 500); 12 Nov 2005 00:27:34 -0000 Mailing-List: contact svn-help@forrest.apache.org; run by ezmlm Precedence: bulk list-help: list-unsubscribe: List-Post: Reply-To: "Forrest Developers List" List-Id: Delivered-To: mailing list svn@forrest.apache.org Received: (qmail 23881 invoked by uid 99); 12 Nov 2005 00:27:33 -0000 Received: from asf.osuosl.org (HELO asf.osuosl.org) (140.211.166.49) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 11 Nov 2005 16:27:33 -0800 X-ASF-Spam-Status: No, hits=-9.4 required=10.0 tests=ALL_TRUSTED,NO_REAL_NAME X-Spam-Check-By: apache.org Received: from [209.237.227.194] (HELO minotaur.apache.org) (209.237.227.194) by apache.org (qpsmtpd/0.29) with SMTP; Fri, 11 Nov 2005 16:27:26 -0800 Received: (qmail 48616 invoked by uid 65534); 12 Nov 2005 00:27:12 -0000 Message-ID: <20051112002712.48596.qmail@minotaur.apache.org> Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: svn commit: r332682 - /forrest/events/forrest-friday/20051111-log.txt Date: Sat, 12 Nov 2005 00:27:07 -0000 To: svn@forrest.apache.org From: crossley@apache.org X-Mailer: svnmailer-1.0.5 X-Virus-Checked: Checked by ClamAV on apache.org X-Spam-Rating: minotaur.apache.org 1.6.2 0/1000/N Author: crossley Date: Fri Nov 11 16:27:00 2005 New Revision: 332682 URL: http://svn.apache.org/viewcvs?rev=332682&view=rev Log: Update Modified: forrest/events/forrest-friday/20051111-log.txt Modified: forrest/events/forrest-friday/20051111-log.txt URL: http://svn.apache.org/viewcvs/forrest/events/forrest-friday/20051111-log.txt?rev=332682&r1=332681&r2=332682&view=diff ============================================================================== --- forrest/events/forrest-friday/20051111-log.txt (original) +++ forrest/events/forrest-friday/20051111-log.txt Fri Nov 11 16:27:00 2005 @@ -462,3 +462,131 @@ Nov 12 06:34:50 * diwaker is away: school Nov 12 06:34:58 i'll leave the bot running and logs collecting Nov 12 06:34:59 for later +Nov 12 09:59:26 good morning, i will be back after breakfast +Nov 12 09:59:59 Good morning David, not sure how long I am around now - got an early start wih my son tomorrow +Nov 12 10:00:23 hi Ross, how long? +Nov 12 10:00:46 At least an hour +Nov 12 10:01:13 okay, back after porridge. +Nov 12 10:01:29 Hey, you spelt it right :-) +Nov 12 10:27:39 backreading was easy, not much chat, but i see from svn logs that you been busy Ross +Nov 12 10:28:15 Yeah, I took our discussion about how best to use FT/FF to heart and just got on with it ;-) +Nov 12 10:28:31 i think so too ... +Nov 12 10:29:01 perhaps let each other know what we are up to from time-to-time +Nov 12 10:29:14 Sure... +Nov 12 10:29:34 I just posted an important comment to FOR726 (plugins use locationmap) - please check it out +Nov 12 10:30:10 looking. +Nov 12 10:30:35 I've been working from the top of my plugins directory listing downwards, just done OOo and starting PhotoGallery +Nov 12 10:31:14 okay i will work backwards +Nov 12 10:32:08 re: FOR-726 note, yes i made the same pencil note when reading svn logs this morning. +Nov 12 10:32:34 I think we need to wrtite a release process for this +Nov 12 10:35:25 Do we need to make every resource in the plugins use the locationmap +Nov 12 10:35:26 or just the ones that will obviously be used from outside? +Nov 12 10:35:26 I wonder if it is inefficient to make absolutely everything +Nov 12 10:35:26 use the locationmap. +Nov 12 10:35:57 I've not really considered this to be honest... +Nov 12 10:36:16 you are right it does introduce a bottleneck +Nov 12 10:36:50 Perhaps we should just do the obvjously useful as you suggest +Nov 12 10:37:11 later we can do the rest +Nov 12 10:37:24 +1 - I'll start doing that from now on +Nov 12 10:37:43 the trouble with leaving some out is inconsistency ... +Nov 12 10:38:18 people might expect the same naming convention. +Nov 12 10:38:26 ... +Nov 12 10:38:51 i suppose that improved error handling for locationmap will assist. +Nov 12 10:39:38 error handling is the biggest problem at present, it is hard to debug locationmaps :-( +Nov 12 10:39:56 very. +Nov 12 10:40:01 keeping the local resources our of the LM actually makes it easier to build the LM +Nov 12 10:40:46 We can discuss the naming convention onlist (consistency is good from an ease of developer perspective) +Nov 12 11:00:41 So Ross, regarding cocoon-docs, i think that we are saying no chance yetto add the Changes processing, because Daisy plugin intercepts everything. Is that right? +Nov 12 11:02:35 As it stands right at this minute yes, but let me commit the simplified-docbook plugin, get a whiskey and I will contemplate it, more to come... +Nov 12 11:04:18 Otherwise, i will try to get the cocoon-site build to generate it into the 2.1 directory. Better for the website, but still doesn't get it into cocoon-docs.zip +Nov 12 11:04:45 I think there should be a way to work around it... +Nov 12 11:05:12 previous to the insistence on sticking with the same url space I used *.daisy.html to indicate daisy docs +Nov 12 11:05:14 Mmm, i had better get some Saturday morning medicine then, and join you. +Nov 12 11:05:21 that way we just had *.html for other docs +Nov 12 11:05:35 I think the answer will be something like only matching 2.1/* docs +Nov 12 11:06:04 hope that it is that easy. +Nov 12 11:06:30 I don't think so, that's why I'll need the whiskey ;-) +Nov 12 11:24:22 My experiments have not been encouraging... +Nov 12 11:24:57 The problem is that Daisy returns a document even if it down't exist (i.e. it doesn't give a 404, it gives an XML error document) +Nov 12 11:25:02 (stupid design!) +Nov 12 11:25:10 Anyway... +Nov 12 11:25:25 This means we can't use a selector, since a document s returned +Nov 12 11:25:28 so... +Nov 12 11:26:31 The answer is to create special matchers in the daisy-to-docs module +Nov 12 11:26:45 rather than do the 2.1/** one +Nov 12 11:27:02 we should do changes.html etc +Nov 12 11:27:06 what do you think? +Nov 12 11:28:09 Yes, i was wondering about that too ... +Nov 12 11:29:21 but didn't/don't know how to effect that. The projectInfo plugin needs to match on "changes" and "todo". +Nov 12 11:29:45 yes, thinking about it there will be multiple matches for none 2.1 daisy docs... +Nov 12 11:30:00 it will be easier to do it the way I first suggested (2.1/*) - experimenting... +Nov 12 11:37:51 ross, see forrest-user list reply. +Nov 12 11:38:14 is this our plugin version numbering inconsistencies? +Nov 12 11:38:44 Could well be, good thought +Nov 12 11:38:56 I'll respond onlist +Nov 12 11:39:26 i did deploy some last night, how will we fix that? +Nov 12 11:41:52 if the version number of the plugin was updated then there is no problem +Nov 12 11:42:06 we just need to tell users to specificy a specific version number (as with daisy) +Nov 12 11:42:15 however, if it wasn't updated +Nov 12 11:42:21 it probably wasn't, i will go find which ones that i deployed. +Nov 12 11:42:44 we will need to CO the plugin prior to the revision +Nov 12 11:42:47 deploy it +Nov 12 11:42:52 co head +Nov 12 11:43:02 update version number +Nov 12 11:43:13 wait for a coordinated release of all plugins post locationmap +Nov 12 11:43:52 okay, i will attend to that. Grrr, i should have known. +Nov 12 11:44:13 I very nearly did the same thing, just managed to CTRL-C in time +Nov 12 11:44:37 howverm there are lots of inconsitensies in my build.xml files and status.xml files now :-( +Nov 12 11:45:31 does 'svn cleanup' help? +Nov 12 11:46:14 I don't think so, I've only ever used that to solve the "repository is locked" errror, what else does it do? +Nov 12 11:46:31 not sure +Nov 12 11:46:43 :-) - you have the same style as me ;-) +Nov 12 11:50:30 where is status.xml for the 2.1 docs? +Nov 12 11:50:41 (I'm having some success :-) +Nov 12 11:51:02 ... +Nov 12 11:51:23 http://svn.apache.org/repos/asf/cocoon/branches/BRANCH_2_1_X/status.xml +Nov 12 11:51:27 thanks +Nov 12 11:52:04 there is a locationmap in http://svn.apache.org/repos/asf/cocoon/site/src/documentation/content/locationmap.xml +Nov 12 11:52:32 temporary ... workaround option c) +Nov 12 11:57:36 the good news is that I have changes.xml (etc.) working in the 2.1 docs... +Nov 12 11:57:48 The problem is that it breaks the daisy plugin... +Nov 12 11:58:02 but not for the cocoon-docs +Nov 12 11:58:07 hooray; oh dear +Nov 12 11:58:13 I can fix the daisy plugin, but I must go to bed +Nov 12 11:58:23 I'll send you a mail tellin gyou how to get it working +Nov 12 11:58:40 you can do what you need to do and we';ll coordinate again later +Nov 12 11:59:16 good night, i will pop back in from time-to-time ... out to the sunny garden soon. +Nov 12 11:59:54 I'll send that mail now, small changes, if you can get it working before I go to bed then I can sleep soundly ;-) +Nov 12 12:04:12 regarding that issue with the deployment of plugins and version numbers: This shows that we need to add some forrestbot builds for the release branch ... continuous integration. +Nov 12 12:04:25 +1 +Nov 12 12:09:07 I sent the mail on how to get cocoon-docs working with changes.xml... +Nov 12 12:09:21 thanks +Nov 12 12:09:28 going to get ready for bed, have a read and let me know if you understand +Nov 12 12:09:43 will chack back in 10 mnutes +Nov 12 12:19:07 Does it make sense for you David? +Nov 12 12:20:01 the first part yes, not sure that i understand the [1] note about locationmap yet. +Nov 12 12:21:10 don't worry, i will try it (not yet looked into navigation-to-locationmap.xsl) +Nov 12 12:21:29 another quick question before you go ... +Nov 12 12:21:40 I think you will be OK with it, it's farly simple +Nov 12 12:21:45 GO ahead with your qquestion +Nov 12 12:21:47 What is performance like for you with the cocoon-daisy-to-docs build? +Nov 12 12:21:47 For me, it takes ages to start (presumably building the navigation), +Nov 12 12:21:47 then takes about 45 seconds per document. Whereas on the zone +Nov 12 12:21:47 it takes 1.5 seconds per document. +Nov 12 12:22:25 Performance locally is awful, similar to what you experience +Nov 12 12:22:37 I think there is a massive amount of network traffic +Nov 12 12:22:56 on the zones it is al within the local netowrk so very fast +Nov 12 12:22:57 i wondered that too ... +Nov 12 12:23:18 and ... yes local to zones. +Nov 12 12:23:59 I've not done any optimisation of the daisy plugin yet +Nov 12 12:24:08 i exaggerated ... 30 seconds per doc. +Nov 12 12:24:15 erk. +Nov 12 12:24:18 my plan is to have a daisy generator that only requests head most of the time +Nov 12 12:24:28 aha. +Nov 12 12:24:58 okay, sweet dreams. +Nov 12 12:25:33 Enjoy your day in the sun, winter has truly arrived here :-( +Nov 12 12:25:37 good night +Nov 12 12:25:58 <-- rgardler has quit ("Chatzilla 0.9.68.5 [Firefox 1.0.7/20050915]") +Nov 12 12:26:11 thanks goodness that we have the project:start-uri parameter in forrest.properties (project:start-uri=changes.html) +Nov 12 12:26:25 builds that doc first. +Nov 12 12:26:31 good night.