Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 22 Next »

[09:57:37 CDT(-0500)] <EricDalquist> so the wrong stylesheet descriptor id is ending up in the portlet entity table

[09:59:27 CDT(-0500)] <EricDalquist> when I call layout.json I'm seeing the minimized state stored in the db for the DLMXHTML stylesheet

[10:24:38 CDT(-0500)] <EricDalquist> ok athena I found the bug

[10:24:44 CDT(-0500)] <EricDalquist> though I'm not sure how I'll fix it yet (tongue)

[10:30:40 CDT(-0500)] <athena> which bug is this? the minimization one?

[10:30:57 CDT(-0500)] <athena> glad we actually know what the problem is now (smile)

[10:31:01 CDT(-0500)] <EricDalquist> yeah

[10:31:04 CDT(-0500)] <athena> can we fix that before the 4.0.6 release?

[10:31:16 CDT(-0500)] <EricDalquist> so the problem is the json pipeline forces a specific theme to be used

[10:31:26 CDT(-0500)] <EricDalquist> but the portlet entity registry looks at the user's profile

[10:31:40 CDT(-0500)] <EricDalquist> to figure out the theme

[10:31:43 CDT(-0500)] <EricDalquist> which resolves to XHTML

[10:32:09 CDT(-0500)] * EricDalquist really needs to start coding on the new user/profile/layout stuff

[10:32:39 CDT(-0500)] <athena> oh interesting - does this only happen if you access layout.json from a desktop/tablet device?

[10:32:56 CDT(-0500)] <EricDalquist> maybe

[10:33:03 CDT(-0500)] <athena> that explains why i couldn't reproduce, potentially

[10:33:04 CDT(-0500)] <athena> good to know

[10:33:26 CDT(-0500)] <athena> but hmm, that does sound harder to fix

[10:33:48 CDT(-0500)] <athena> is there a way to not have layout.json save minimization state?

[10:34:54 CDT(-0500)] <EricDalquist> yes

[10:35:03 CDT(-0500)] <EricDalquist> but the entity registry needs to know it is the json layout being used

[10:37:22 CDT(-0500)] <EricDalquist> then end goal in 4.1 is that portlet entity data becomes part of the layout data

[10:37:26 CDT(-0500)] <EricDalquist> which is based on the profile being used

[10:37:42 CDT(-0500)] <EricDalquist> I might see if we can somehow layout.json to force a specific profile to be used for that request

[10:37:50 CDT(-0500)] <EricDalquist> but I'm afraid of breaking things

[10:37:57 CDT(-0500)] <EricDalquist> since that profile code is so scarry

[10:38:00 CDT(-0500)] <athena> yeah, that is worrisome

[10:38:08 CDT(-0500)] <athena> problematic because we can't really leave it as is, either

[10:38:17 CDT(-0500)] <EricDalquist> right

[11:02:37 CDT(-0500)] <EricDalquist> blarg

[11:07:01 CDT(-0500)] <EricDalquist> I hate this profile code

[11:11:21 CDT(-0500)] <athena> yeah it's awful (sad)

[11:14:50 CDT(-0500)] <EricDalquist> here is a question

[11:15:04 CDT(-0500)] <EricDalquist> hrm

[11:15:06 CDT(-0500)] <EricDalquist> nevermind

[11:20:40 CDT(-0500)] <EricDalquist1> so my only idea so far is to add an IUserProfile wrapper

[11:20:49 CDT(-0500)] <EricDalquist1> that lets us override profile information via a ThreadLocal

[11:20:53 CDT(-0500)] <EricDalquist1> its gross

[11:20:59 CDT(-0500)] <EricDalquist1> but I'm not sure what else to do

[11:21:02 CDT(-0500)] <EricDalquist1> (sad)

[11:21:47 CDT(-0500)] <EricDalquist1> oh ... hrm

[11:21:52 CDT(-0500)] <EricDalquist1> maybe I can do it via a request attribute

[11:35:58 CDT(-0500)] <EricDalquist1> yeah I think that will work

[11:36:16 CDT(-0500)] <EricDalquist1> I can tweak the code to set the current theme/struct stylesheet descriptors as request attributs

[11:36:26 CDT(-0500)] <EricDalquist1> and then use that in the portlet registery to do the association

[11:38:54 CDT(-0500)] <athena> excellent (smile)

[12:00:23 CDT(-0500)] <athena> EricDalquist: we still accepting fixes for 4.0.6? i'm imagining you still have some work to do given the profile issues

[12:00:31 CDT(-0500)] <EricDalquist> yup

[12:00:35 CDT(-0500)] <athena> ok, cool :0

[12:00:36 CDT(-0500)] <athena> (smile)

[12:00:40 CDT(-0500)] <EricDalquist> I'll send out a note for the actual code freeze

[12:00:48 CDT(-0500)] <EricDalquist> and really with git

[12:00:51 CDT(-0500)] <EricDalquist> it isn'

[12:00:54 CDT(-0500)] <EricDalquist> t a worry

[12:00:58 CDT(-0500)] <EricDalquist> since I can choose when to release

[12:01:06 CDT(-0500)] <EricDalquist> and just not include commits

[12:01:08 CDT(-0500)] <athena> quite true

[12:55:33 CDT(-0500)] <drewwills> hey there EricDalquist and athena – to your knowledge, is there a wiki page (or other online document) that outlines the uPortal practices for tagging & branching in SCM?

[12:56:11 CDT(-0500)] <athena> all your tagging should be done automatically by the mvn release plugin, probably

[12:56:25 CDT(-0500)] <athena> i can't actually look anything up right now because my internet is being completely dysfuncitonal

[12:56:56 CDT(-0500)] <athena> branching is probably documented in the git workflow page, which is an excellent resource

[12:59:15 CDT(-0500)] <drewwills> good point about the mvn release plugin... we're looking for resources we can link wrt SSP

[13:11:51 CDT(-0500)] <athena> i'm assuming everyone else can get to github just fine?

  • No labels