[jira] Updated: (FELIX-1959) Move towards unified LF and extended branding support

2009-12-22 Thread Valentin Valchev (JIRA)

 [ 
https://issues.apache.org/jira/browse/FELIX-1959?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Valentin Valchev updated FELIX-1959:


Attachment: jqueryui-vs-default-lnf.png

 Move towards unified LF and extended branding support
 --

 Key: FELIX-1959
 URL: https://issues.apache.org/jira/browse/FELIX-1959
 Project: Felix
  Issue Type: New Feature
  Components: Web Console
Reporter: Valentin Valchev
Priority: Minor
 Attachments: jqueryui-vs-default-lnf.png


 Currently the Web Console uses heavily on the JQuery framework. Using a 
 unified JavaScript framework simplifies development of all plugins and 
 unifies the used approach.
 However, when talking about visual styling, there are number of differences 
 because each plugin developer uses own styles.
 My suggestion is to adopt the JQuery UI . The benefits of using it as unified 
 widget/css framework are:
 - no time to spend on writing widgets already in the library
 - clean CSS visual styling
 - easy way to change the LF by changing the theme (extended branding 
 support!)
 - improved cross-browser support (JQuery UI takes care of CSS differences)
 Using the JQuery UI framework the developer shouldn't care about color but 
 only for layout - components position; and for data being displayed.
 To illustrate the benefits I've saved the Log Service page, modified it to 
 use JQuery UI, took screen-shot, modified the theme CSS only, and again took 
 screen-shot, and finally added the original LF for reference, so you can 
 easily compare the result.
 The attached image contains the combined screen-shots.
 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Created: (FELIX-1959) Move towards unified LF and extended branding support

2009-12-22 Thread Valentin Valchev (JIRA)
Move towards unified LF and extended branding support
--

 Key: FELIX-1959
 URL: https://issues.apache.org/jira/browse/FELIX-1959
 Project: Felix
  Issue Type: New Feature
  Components: Web Console
Reporter: Valentin Valchev
Priority: Minor
 Attachments: jqueryui-vs-default-lnf.png

Currently the Web Console uses heavily on the JQuery framework. Using a unified 
JavaScript framework simplifies development of all plugins and unifies the used 
approach.

However, when talking about visual styling, there are number of differences 
because each plugin developer uses own styles.

My suggestion is to adopt the JQuery UI . The benefits of using it as unified 
widget/css framework are:
- no time to spend on writing widgets already in the library
- clean CSS visual styling
- easy way to change the LF by changing the theme (extended branding support!)
- improved cross-browser support (JQuery UI takes care of CSS differences)

Using the JQuery UI framework the developer shouldn't care about color but only 
for layout - components position; and for data being displayed.

To illustrate the benefits I've saved the Log Service page, modified it to 
use JQuery UI, took screen-shot, modified the theme CSS only, and again took 
screen-shot, and finally added the original LF for reference, so you can 
easily compare the result.

The attached image contains the combined screen-shots.


-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Assigned: (FELIX-1959) Move towards unified LF and extended branding support

2009-12-22 Thread Carsten Ziegeler (JIRA)

 [ 
https://issues.apache.org/jira/browse/FELIX-1959?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Carsten Ziegeler reassigned FELIX-1959:
---

Assignee: Carsten Ziegeler

 Move towards unified LF and extended branding support
 --

 Key: FELIX-1959
 URL: https://issues.apache.org/jira/browse/FELIX-1959
 Project: Felix
  Issue Type: New Feature
  Components: Web Console
Reporter: Valentin Valchev
Assignee: Carsten Ziegeler
Priority: Minor
 Attachments: jqueryui-vs-default-lnf.png


 Currently the Web Console uses heavily on the JQuery framework. Using a 
 unified JavaScript framework simplifies development of all plugins and 
 unifies the used approach.
 However, when talking about visual styling, there are number of differences 
 because each plugin developer uses own styles.
 My suggestion is to adopt the JQuery UI . The benefits of using it as unified 
 widget/css framework are:
 - no time to spend on writing widgets already in the library
 - clean CSS visual styling
 - easy way to change the LF by changing the theme (extended branding 
 support!)
 - improved cross-browser support (JQuery UI takes care of CSS differences)
 Using the JQuery UI framework the developer shouldn't care about color but 
 only for layout - components position; and for data being displayed.
 To illustrate the benefits I've saved the Log Service page, modified it to 
 use JQuery UI, took screen-shot, modified the theme CSS only, and again took 
 screen-shot, and finally added the original LF for reference, so you can 
 easily compare the result.
 The attached image contains the combined screen-shots.
 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Commented: (FELIX-1959) Move towards unified LF and extended branding support

2009-12-22 Thread Carsten Ziegeler (JIRA)

[ 
https://issues.apache.org/jira/browse/FELIX-1959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12793585#action_12793585
 ] 

Carsten Ziegeler commented on FELIX-1959:
-

Looks really cool!

I completly agree that having a common look at and feel across all plugins is 
desirable.
The current implementation of some plugins when it comes to rendering is too 
complex. We generate partial html, have javascript that adds missing html etc. 
Very confusing and error prone.
One idee I discussed briefly with Felix M., was to go one step back and make 
the plugins deliver all html - without styling, without javascript etc. and 
then add the dynamic stuff via jquery.

I haven't done anything with jQuery UI (but will do in the next days). If it 
helps in making the development of plugins easier and more consistent, I'm all 
for it!

Could you show some sample code of how this would look like from the 
development perspective?

 Move towards unified LF and extended branding support
 --

 Key: FELIX-1959
 URL: https://issues.apache.org/jira/browse/FELIX-1959
 Project: Felix
  Issue Type: New Feature
  Components: Web Console
Reporter: Valentin Valchev
Priority: Minor
 Attachments: jqueryui-vs-default-lnf.png


 Currently the Web Console uses heavily on the JQuery framework. Using a 
 unified JavaScript framework simplifies development of all plugins and 
 unifies the used approach.
 However, when talking about visual styling, there are number of differences 
 because each plugin developer uses own styles.
 My suggestion is to adopt the JQuery UI . The benefits of using it as unified 
 widget/css framework are:
 - no time to spend on writing widgets already in the library
 - clean CSS visual styling
 - easy way to change the LF by changing the theme (extended branding 
 support!)
 - improved cross-browser support (JQuery UI takes care of CSS differences)
 Using the JQuery UI framework the developer shouldn't care about color but 
 only for layout - components position; and for data being displayed.
 To illustrate the benefits I've saved the Log Service page, modified it to 
 use JQuery UI, took screen-shot, modified the theme CSS only, and again took 
 screen-shot, and finally added the original LF for reference, so you can 
 easily compare the result.
 The attached image contains the combined screen-shots.
 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Updated: (FELIX-1959) Move towards unified LF and extended branding support

2009-12-22 Thread Valentin Valchev (JIRA)

 [ 
https://issues.apache.org/jira/browse/FELIX-1959?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Valentin Valchev updated FELIX-1959:


Attachment: tb6.jar

Ok, I've added the LogService refactored to use JQueryUI internally. Of course, 
the JQueryUI is exported temporally by plug-in resource, since it is not yet 
exported by the console itself.

You can check the themes in a very simple way:
- set the system property 'webconsole.alternative.theme=true'
- update the bundle
- refresh the web browser (Ctrl+F5)

This is just proof of concept and is just code copied from the original log 
service with some modifications but it's not optimized to run at it's best 
performance. There are a number of optimizations that can be performed, if it 
is supposed to replace the original LogService. As example it can improved by:
- caching of the loaded resource (data.html)
- adding possibility to load the resource using the request locale (load 
data_en.html, data_fr.html, depending on request.getLocales()).

 Move towards unified LF and extended branding support
 --

 Key: FELIX-1959
 URL: https://issues.apache.org/jira/browse/FELIX-1959
 Project: Felix
  Issue Type: New Feature
  Components: Web Console
Reporter: Valentin Valchev
Assignee: Carsten Ziegeler
Priority: Minor
 Attachments: jqueryui-vs-default-lnf.png, tb6.jar


 Currently the Web Console uses heavily on the JQuery framework. Using a 
 unified JavaScript framework simplifies development of all plugins and 
 unifies the used approach.
 However, when talking about visual styling, there are number of differences 
 because each plugin developer uses own styles.
 My suggestion is to adopt the JQuery UI . The benefits of using it as unified 
 widget/css framework are:
 - no time to spend on writing widgets already in the library
 - clean CSS visual styling
 - easy way to change the LF by changing the theme (extended branding 
 support!)
 - improved cross-browser support (JQuery UI takes care of CSS differences)
 Using the JQuery UI framework the developer shouldn't care about color but 
 only for layout - components position; and for data being displayed.
 To illustrate the benefits I've saved the Log Service page, modified it to 
 use JQuery UI, took screen-shot, modified the theme CSS only, and again took 
 screen-shot, and finally added the original LF for reference, so you can 
 easily compare the result.
 The attached image contains the combined screen-shots.
 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Commented: (FELIX-1792) Felix OBR seems to just randomly choose one of the satisifed bundles if more than one bundle meets the requirement

2009-12-22 Thread hehe ji (JIRA)

[ 
https://issues.apache.org/jira/browse/FELIX-1792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12793645#action_12793645
 ] 

hehe ji commented on FELIX-1792:


Hi Richard,

Thanks for the patch. I tried it but I got deadlock problem for the following 
situation:

Bundle71 (in repo) : Depends on Bundle72
Bundle72 v1 (in repo) : depends on bundle73
Bundle72_2 v2 (in eba)  : Depends on Bundle.invalid // This should not be 
picked up
Bundle73 (in eba) : No Dependencies

I expect bundle71, bundle71 v1 and bundle73 are installed in framework.



3XMTHREADINFO  Thread-129 TID:0x0238A400, 
j9thread_t:0x04000ABA1920, state:CW, prio=5
3XMTHREADINFO1(native thread ID:0x6BF9, native priority:0x5, native 
policy:UNKNOWN)
3XMTHREADINFO2(native stack address range from:0x04000B58, 
to:0x04000B54, size:0x4)
4XESTACKTRACE  at 
java/lang/Character.toLowerCase(Character.java:4187(Compiled Code))
4XESTACKTRACE  at 
java/lang/String.compareValue(String.java:450(Compiled Code))
4XESTACKTRACE  at 
java/lang/String.compareToIgnoreCase(String.java:466(Compiled Code))
4XESTACKTRACE  at 
org/apache/felix/bundlerepository/ResourceImpl$1.compare(ResourceImpl.java:55(Compiled
 Code))
4XESTACKTRACE  at java/util/TreeMap.cmp(TreeMap.java:4472(Compiled 
Code))
4XESTACKTRACE  at java/util/TreeMap.get(TreeMap.java:4344(Compiled 
Code))
4XESTACKTRACE  at 
org/apache/felix/bundlerepository/ResourceImpl.getSymbolicName(ResourceImpl.java:107(Compiled
 Code))
4XESTACKTRACE  at 
org/apache/felix/bundlerepository/ResourceImpl.hashCode(ResourceImpl.java:84(Compiled
 Code))
4XESTACKTRACE  at java/util/HashMap.getEntry(HashMap.java:506(Compiled 
Code))
4XESTACKTRACE  at 
java/util/HashMap.containsKey(HashMap.java:432(Compiled Code))
4XESTACKTRACE  at java/util/HashSet.contains(HashSet.java:138(Compiled 
Code))
4XESTACKTRACE  at 
org/apache/felix/bundlerepository/ResolverImpl.resolve(ResolverImpl.java:158(Compiled
 Code))
4XESTACKTRACE  at 
org/apache/felix/bundlerepository/ResolverImpl.resolve(ResolverImpl.java:199(Compiled
 Code))
4XESTACKTRACE  at 
org/apache/felix/bundlerepository/ResolverImpl.resolve(ResolverImpl.java:199)
4XESTACKTRACE  at 
org/apache/felix/bundlerepository/ResolverImpl.resolve(ResolverImpl.java:133)

Please let me know if you need more information.
Thanks
Emily

 Felix OBR seems to just randomly choose one of the satisifed bundles if more 
 than one bundle meets the requirement
 --

 Key: FELIX-1792
 URL: https://issues.apache.org/jira/browse/FELIX-1792
 Project: Felix
  Issue Type: Bug
  Components: Bundle Repository (OBR)
Affects Versions: bundlerepository-1.4.2
 Environment: n/a
Reporter: david small99
Assignee: Richard S. Hall
 Fix For: bundlerepository-1.6.0


 I have one bundle bundle1, which imports a package called com.obr.bundle
 Bundle1's manifest:
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle1
 Bundle-Version: 1.0.0
 Export-Package: com.obr.bundle1
 Import-Package: com.obr.bundle;version=[1.2.0.999,3.2.2.bz]
 There are two bundles in my repositories, bundler2 and bundle 3. Both of them 
 export package com.obr.bundle. Below are their manifest files.
 Bundle2
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle2
 Bundle-Version: 1.0.0
 Bundle-Vendor: xxx
 Import-Package: a.b.c
 Export-Package: com.obr.bundle;version=3.2.2.blah
 Bundle3:
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle3
 Bundle-Version: 1.0.0
 Import-Package: a.b.c
 Export-Package: com.obr.bundle;version=3.1
 As you can see, both bundle2 and bundle3 meet the requirements of bundle1. I 
 hope the highest package version, which is bundle2, is chosen by felix obr. 
 However, sometimes bundle 3 was chosen instead of bundle2. The behaviour is 
 random.
 Am I right to say that the Felix obr runtime just picks the first bundle that 
 meets the requirements and then stop searching for any more eligible bundles?
 Thanks

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Commented: (FELIX-1792) Felix OBR seems to just randomly choose one of the satisifed bundles if more than one bundle meets the requirement

2009-12-22 Thread Richard S. Hall (JIRA)

[ 
https://issues.apache.org/jira/browse/FELIX-1792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12793654#action_12793654
 ] 

Richard S. Hall commented on FELIX-1792:


What is in eba mean?

Could you give me a simplified example of the bundle manifests and a 
description of the steps needed to reproduce the issue? If so, I will look into 
it.

 Felix OBR seems to just randomly choose one of the satisifed bundles if more 
 than one bundle meets the requirement
 --

 Key: FELIX-1792
 URL: https://issues.apache.org/jira/browse/FELIX-1792
 Project: Felix
  Issue Type: Bug
  Components: Bundle Repository (OBR)
Affects Versions: bundlerepository-1.4.2
 Environment: n/a
Reporter: david small99
Assignee: Richard S. Hall
 Fix For: bundlerepository-1.6.0


 I have one bundle bundle1, which imports a package called com.obr.bundle
 Bundle1's manifest:
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle1
 Bundle-Version: 1.0.0
 Export-Package: com.obr.bundle1
 Import-Package: com.obr.bundle;version=[1.2.0.999,3.2.2.bz]
 There are two bundles in my repositories, bundler2 and bundle 3. Both of them 
 export package com.obr.bundle. Below are their manifest files.
 Bundle2
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle2
 Bundle-Version: 1.0.0
 Bundle-Vendor: xxx
 Import-Package: a.b.c
 Export-Package: com.obr.bundle;version=3.2.2.blah
 Bundle3:
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle3
 Bundle-Version: 1.0.0
 Import-Package: a.b.c
 Export-Package: com.obr.bundle;version=3.1
 As you can see, both bundle2 and bundle3 meet the requirements of bundle1. I 
 hope the highest package version, which is bundle2, is chosen by felix obr. 
 However, sometimes bundle 3 was chosen instead of bundle2. The behaviour is 
 random.
 Am I right to say that the Felix obr runtime just picks the first bundle that 
 meets the requirements and then stop searching for any more eligible bundles?
 Thanks

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Commented: (FELIX-1959) Move towards unified LF and extended branding support

2009-12-22 Thread Felix Meschberger (JIRA)

[ 
https://issues.apache.org/jira/browse/FELIX-1959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12793658#action_12793658
 ] 

Felix Meschberger commented on FELIX-1959:
--

This looks very promising -- but also like a big change in various rendering 
areas.

Can we do this step by step aka page by page ?

Anyways, a big +1 for pursuiing this.

... and investigation of internationalization would be a nice thing.

 Move towards unified LF and extended branding support
 --

 Key: FELIX-1959
 URL: https://issues.apache.org/jira/browse/FELIX-1959
 Project: Felix
  Issue Type: New Feature
  Components: Web Console
Reporter: Valentin Valchev
Assignee: Carsten Ziegeler
Priority: Minor
 Attachments: jqueryui-vs-default-lnf.png, tb6.jar


 Currently the Web Console uses heavily on the JQuery framework. Using a 
 unified JavaScript framework simplifies development of all plugins and 
 unifies the used approach.
 However, when talking about visual styling, there are number of differences 
 because each plugin developer uses own styles.
 My suggestion is to adopt the JQuery UI . The benefits of using it as unified 
 widget/css framework are:
 - no time to spend on writing widgets already in the library
 - clean CSS visual styling
 - easy way to change the LF by changing the theme (extended branding 
 support!)
 - improved cross-browser support (JQuery UI takes care of CSS differences)
 Using the JQuery UI framework the developer shouldn't care about color but 
 only for layout - components position; and for data being displayed.
 To illustrate the benefits I've saved the Log Service page, modified it to 
 use JQuery UI, took screen-shot, modified the theme CSS only, and again took 
 screen-shot, and finally added the original LF for reference, so you can 
 easily compare the result.
 The attached image contains the combined screen-shots.
 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Issue Comment Edited: (FELIX-1792) Felix OBR seems to just randomly choose one of the satisifed bundles if more than one bundle meets the requirement

2009-12-22 Thread Richard S. Hall (JIRA)

[ 
https://issues.apache.org/jira/browse/FELIX-1792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12793654#action_12793654
 ] 

Richard S. Hall edited comment on FELIX-1792 at 12/22/09 4:35 PM:
--

What does in eba mean?

Could you give me a simplified example of the bundle manifests and a 
description of the steps needed to reproduce the issue? If so, I will look into 
it.

  was (Author: rickhall):
What is in eba mean?

Could you give me a simplified example of the bundle manifests and a 
description of the steps needed to reproduce the issue? If so, I will look into 
it.
  
 Felix OBR seems to just randomly choose one of the satisifed bundles if more 
 than one bundle meets the requirement
 --

 Key: FELIX-1792
 URL: https://issues.apache.org/jira/browse/FELIX-1792
 Project: Felix
  Issue Type: Bug
  Components: Bundle Repository (OBR)
Affects Versions: bundlerepository-1.4.2
 Environment: n/a
Reporter: david small99
Assignee: Richard S. Hall
 Fix For: bundlerepository-1.6.0


 I have one bundle bundle1, which imports a package called com.obr.bundle
 Bundle1's manifest:
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle1
 Bundle-Version: 1.0.0
 Export-Package: com.obr.bundle1
 Import-Package: com.obr.bundle;version=[1.2.0.999,3.2.2.bz]
 There are two bundles in my repositories, bundler2 and bundle 3. Both of them 
 export package com.obr.bundle. Below are their manifest files.
 Bundle2
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle2
 Bundle-Version: 1.0.0
 Bundle-Vendor: xxx
 Import-Package: a.b.c
 Export-Package: com.obr.bundle;version=3.2.2.blah
 Bundle3:
 Manifest-Version: 1.0
 Ant-Version: Apache Ant 1.7.0
 Bundle-ManifestVersion: 2
 Bundle-Name: Bundle Plug-in
 Bundle-SymbolicName: com.obr.bundle3
 Bundle-Version: 1.0.0
 Import-Package: a.b.c
 Export-Package: com.obr.bundle;version=3.1
 As you can see, both bundle2 and bundle3 meet the requirements of bundle1. I 
 hope the highest package version, which is bundle2, is chosen by felix obr. 
 However, sometimes bundle 3 was chosen instead of bundle2. The behaviour is 
 random.
 Am I right to say that the Felix obr runtime just picks the first bundle that 
 meets the requirements and then stop searching for any more eligible bundles?
 Thanks

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



Re: [VOTE] Release fileinstall 2.0.6

2009-12-22 Thread Chris Custine
This vote is cancelled due to bad signatures on the artifacts. I'll launch a
vote on a 2.0.8 release in a new thread.

Chris
--
Chris Custine
FUSESource :: http://fusesource.com
My Blog :: http://blog.organicelement.com
Apache ServiceMix :: http://servicemix.apache.org
Apache Felix :: http://felix.apache.org
Apache Directory Server :: http://directory.apache.org


On Thu, Dec 17, 2009 at 3:06 AM, Guillaume Nodet gno...@gmail.com wrote:

 Release Notes - Felix - Version fileinstall-2.0.6
 ** Bug
* [FELIX-1775] - fileinstall release archives contains an empty
 load folder which should be removed
* [FELIX-1789] - FileInstall is too verbose
* [FELIX-1851] - FileInstall watched bundles are restarted twice upon
 update
* [FELIX-1861] - FileInstall created temp directories are never deleted
* [FELIX-1862] - fileinstall thread name as printed in log
 messages need to be less verbose
* [FELIX-1928] - File installer starts bundles too early on restart
 ** Task
* [FELIX-1811] - fileinstall should depend on the official osgi
 artifacts

 Staging repository:
 https://repository.apache.org/content/repositories/felix-staging-054/

 You can use this UNIX script to download the release and verify the
 signatures:
 http://svn.apache.org/repos/asf/felix/trunk/check_staged_release.sh

 Usage:
 sh check_staged_release.sh 054 /tmp/felix-staging

 Please vote to approve this release:

 [ ] +1 Approve the release
 [ ] -1 Veto the release (please provide specific comments)

 This vote will be open for 72 hours.

 --
 Cheers,
 Guillaume Nodet
 
 Blog: http://gnodet.blogspot.com/
 
 Open Source SOA
 http://fusesource.com



[VOTE] Release fileinstall 2.0.8

2009-12-22 Thread Chris Custine
Hi All,
Before checking the staged artifacts, please be sure to import my public key
from either http://www.apache.org/dist/felix/KEYS or a public keyserver.

Release Notes - Felix - Version fileinstall-2.0.8
** Bug
   * [FELIX-1775] - fileinstall release archives contains an empty load
folder which should be removed
   * [FELIX-1789] - FileInstall is too verbose
   * [FELIX-1851] - FileInstall watched bundles are restarted twice upon
update
   * [FELIX-1861] - FileInstall created temp directories are never deleted
   * [FELIX-1862] - fileinstall thread name as printed in log messages need
to be less verbose
   * [FELIX-1928] - File installer starts bundles too early on restart
** Task
   * [FELIX-1811] - fileinstall should depend on the official osgi artifacts

Staging repository:
https://repository.apache.org/content/repositories/orgapachefelix-005/

You can use this UNIX script to download the release and verify the
signatures:
http://svn.apache.org/repos/asf/felix/trunk/check_staged_release.sh

Usage:
sh check_staged_release.sh 005 /tmp/felix-staging

Please vote to approve this release:

[ ] +1 Approve the release
[ ] -1 Veto the release (please provide specific comments)

This vote will be open for 72 hours.

Thanks,
Chris

--
Chris Custine
FUSESource :: http://fusesource.com
My Blog :: http://blog.organicelement.com
Apache ServiceMix :: http://servicemix.apache.org
Apache Felix :: http://felix.apache.org
Apache Directory Server :: http://directory.apache.org


Re: [VOTE] Release fileinstall 2.0.6

2009-12-22 Thread Guillaume Nodet
Sorry about that (I'm in holidays until January so haven't had much
time to follow on this vote).  Chris, thx for taking care of that.

On Tue, Dec 22, 2009 at 21:06, Chris Custine chris.cust...@gmail.com wrote:
 This vote is cancelled due to bad signatures on the artifacts. I'll launch a
 vote on a 2.0.8 release in a new thread.

 Chris
 --
 Chris Custine
 FUSESource :: http://fusesource.com
 My Blog :: http://blog.organicelement.com
 Apache ServiceMix :: http://servicemix.apache.org
 Apache Felix :: http://felix.apache.org
 Apache Directory Server :: http://directory.apache.org


 On Thu, Dec 17, 2009 at 3:06 AM, Guillaume Nodet gno...@gmail.com wrote:

 Release Notes - Felix - Version fileinstall-2.0.6
 ** Bug
    * [FELIX-1775] - fileinstall release archives contains an empty
 load folder which should be removed
    * [FELIX-1789] - FileInstall is too verbose
    * [FELIX-1851] - FileInstall watched bundles are restarted twice upon
 update
    * [FELIX-1861] - FileInstall created temp directories are never deleted
    * [FELIX-1862] - fileinstall thread name as printed in log
 messages need to be less verbose
    * [FELIX-1928] - File installer starts bundles too early on restart
 ** Task
    * [FELIX-1811] - fileinstall should depend on the official osgi
 artifacts

 Staging repository:
 https://repository.apache.org/content/repositories/felix-staging-054/

 You can use this UNIX script to download the release and verify the
 signatures:
 http://svn.apache.org/repos/asf/felix/trunk/check_staged_release.sh

 Usage:
 sh check_staged_release.sh 054 /tmp/felix-staging

 Please vote to approve this release:

 [ ] +1 Approve the release
 [ ] -1 Veto the release (please provide specific comments)

 This vote will be open for 72 hours.

 --
 Cheers,
 Guillaume Nodet
 
 Blog: http://gnodet.blogspot.com/
 
 Open Source SOA
 http://fusesource.com





-- 
Cheers,
Guillaume Nodet

Blog: http://gnodet.blogspot.com/

Open Source SOA
http://fusesource.com


Re: [VOTE] Release fileinstall 2.0.8

2009-12-22 Thread Gert Vanthienen
L.S.,

Here's my +1 (non-binding)

Regards,

Gert Vanthienen

Open Source SOA: http://fusesource.com
Blog: http://gertvanthienen.blogspot.com/



2009/12/22 Chris Custine ccust...@apache.org:
 Hi All,
 Before checking the staged artifacts, please be sure to import my public key
 from either http://www.apache.org/dist/felix/KEYS or a public keyserver.

 Release Notes - Felix - Version fileinstall-2.0.8
 ** Bug
   * [FELIX-1775] - fileinstall release archives contains an empty load
 folder which should be removed
   * [FELIX-1789] - FileInstall is too verbose
   * [FELIX-1851] - FileInstall watched bundles are restarted twice upon
 update
   * [FELIX-1861] - FileInstall created temp directories are never deleted
   * [FELIX-1862] - fileinstall thread name as printed in log messages need
 to be less verbose
   * [FELIX-1928] - File installer starts bundles too early on restart
 ** Task
   * [FELIX-1811] - fileinstall should depend on the official osgi artifacts

 Staging repository:
 https://repository.apache.org/content/repositories/orgapachefelix-005/

 You can use this UNIX script to download the release and verify the
 signatures:
 http://svn.apache.org/repos/asf/felix/trunk/check_staged_release.sh

 Usage:
 sh check_staged_release.sh 005 /tmp/felix-staging

 Please vote to approve this release:

 [ ] +1 Approve the release
 [ ] -1 Veto the release (please provide specific comments)

 This vote will be open for 72 hours.

 Thanks,
 Chris

 --
 Chris Custine
 FUSESource :: http://fusesource.com
 My Blog :: http://blog.organicelement.com
 Apache ServiceMix :: http://servicemix.apache.org
 Apache Felix :: http://felix.apache.org
 Apache Directory Server :: http://directory.apache.org



[jira] Commented: (FELIX-1959) Move towards unified LF and extended branding support

2009-12-22 Thread Carsten Ziegeler (JIRA)

[ 
https://issues.apache.org/jira/browse/FELIX-1959?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12793956#action_12793956
 ] 

Carsten Ziegeler commented on FELIX-1959:
-

Yes, I think we should move forward with this!
Maybe we should start by adding the libraries to the console and then migrate 
one plugin after the other.
When we have migrated all plugins we could cleanup the html/css of the console 
itself

And having i18n support would be a real nice thing...but maybe we could do this 
afterwards?

So if you have any patches, Valentin, they're more than welcome :)

 Move towards unified LF and extended branding support
 --

 Key: FELIX-1959
 URL: https://issues.apache.org/jira/browse/FELIX-1959
 Project: Felix
  Issue Type: New Feature
  Components: Web Console
Reporter: Valentin Valchev
Assignee: Carsten Ziegeler
Priority: Minor
 Attachments: jqueryui-vs-default-lnf.png, tb6.jar


 Currently the Web Console uses heavily on the JQuery framework. Using a 
 unified JavaScript framework simplifies development of all plugins and 
 unifies the used approach.
 However, when talking about visual styling, there are number of differences 
 because each plugin developer uses own styles.
 My suggestion is to adopt the JQuery UI . The benefits of using it as unified 
 widget/css framework are:
 - no time to spend on writing widgets already in the library
 - clean CSS visual styling
 - easy way to change the LF by changing the theme (extended branding 
 support!)
 - improved cross-browser support (JQuery UI takes care of CSS differences)
 Using the JQuery UI framework the developer shouldn't care about color but 
 only for layout - components position; and for data being displayed.
 To illustrate the benefits I've saved the Log Service page, modified it to 
 use JQuery UI, took screen-shot, modified the theme CSS only, and again took 
 screen-shot, and finally added the original LF for reference, so you can 
 easily compare the result.
 The attached image contains the combined screen-shots.
 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Updated: (FELIX-1913) All events are processed in a queue

2009-12-22 Thread Carsten Ziegeler (JIRA)

 [ 
https://issues.apache.org/jira/browse/FELIX-1913?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Carsten Ziegeler updated FELIX-1913:


Description: 
The current event admin implementation puts all events into a  queue and 
processes this queue in one thread. This creates a bottleneck when different 
threads send events as they have to wait for other threads to be processed 
first. Events from different threads can be processed in parallel.
In the async mode, event deliver might take a long time as these events are 
processed one after the other as well.

  was:The current event admin implementation puts all events into one single 
queue and processes this queue is in one thread. This creates a bottleneck when 
different threads send events as they have to wait for other threads to be 
processed first. Events from different threads can be processed in parallel.

Summary: All events are processed in a queue  (was: All synchronous 
events are processed in one queue)

 All events are processed in a queue
 ---

 Key: FELIX-1913
 URL: https://issues.apache.org/jira/browse/FELIX-1913
 Project: Felix
  Issue Type: Improvement
  Components: Event Admin
Affects Versions: eventadmin 1.0.0
Reporter: Carsten Ziegeler
Assignee: Karl Pauls
Priority: Minor
 Attachments: ea.patch


 The current event admin implementation puts all events into a  queue and 
 processes this queue in one thread. This creates a bottleneck when different 
 threads send events as they have to wait for other threads to be processed 
 first. Events from different threads can be processed in parallel.
 In the async mode, event deliver might take a long time as these events are 
 processed one after the other as well.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.



[jira] Created: (FELIX-1960) Fine-grained timeout configuration

2009-12-22 Thread Carsten Ziegeler (JIRA)
Fine-grained timeout configuration
--

 Key: FELIX-1960
 URL: https://issues.apache.org/jira/browse/FELIX-1960
 Project: Felix
  Issue Type: Improvement
  Components: Event Admin
Affects Versions: eventadmin 1.0.0
Reporter: Carsten Ziegeler


The timeout setting for event delivery is currentl a global setting - event 
listeners should (recommended by the spec as well) start an async thread if 
they need some time to process the event.
With the timeout setting in place, this creates two threads for each event 
delivery: one for the timeout handling and one for the event processing.

I think it would be nice, if one could optimize this behaviour by configuring 
the event admin to directly call specific event listeners and don't check the 
timeout. An administrator could add well-known event listeners to this 
configuration. This would reduce the server load for well-behaving event 
listeners.

I could imagine something like 
ignore.timeout.for=org.apache.sling.jcr.*,my.special.EventListener  (with a 
better name for the config property) - so you can either add package prefixes 
or specific classes to the configuration.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.