On 05/03/2011 09:40 AM, Vincent Massol wrote:
>
> On May 3, 2011, at 9:39 AM, Vincent Massol wrote:
>
>> Hi Andy,
>>
>> On Apr 27, 2011, at 1:25 PM, andyp wrote:
>>
>>> Hi all,
>>>
>>> I'm trying to install the XWiki war file on a Ubuntu virtual machine (KVM)
>>> but as yet with no success. I've installed and configured java, jboss and
>>> mysql using all the information I could find from your website:
>>
>> [snip]
>>
>>> 01:36:10,087 WARN [ClassLoaderManager] Unexpected error during load
>>> of:org.apache.jackrabbit.core.query.lucene.WildcardQuery$WildcardQueryWeight:
>>> java.lang.IncompatibleClassChangeError: Implementing class
>>
>> [snip]
>>
>>> 01:36:10,936 WARN [ClassLoaderManager] Unexpected error during load
>>> of:org.apache.jackrabbit.core.query.lucene.MatchAllWeight:
>>> java.lang.IncompatibleClassChangeError: Implementing class
>>
>> [snip]
>>
>> As you can see in the error above the problem is an incompatible version. JBoss uses Jackrabbit in a given version and XWiki users another version of Jackrabbit.
>>
>> It's the role of the appserver to isolate classloader so that a webapp can use whatever version it wants in WEB-INF/lib.
>>
>> You need to configure JBoss to provide this isolation. You need to check Jboss's doc for that.
>
> I just googled it quickly and found
> http://community.jboss.org/wiki/ClassLoadingConfiguration
Should we create a jboss-web.xml configuration file as well?
We recently added a sun-web.xml file to fix another classloading
problem, this time for GlassFish.
> Hope it helps,
> -Vincent
>
>>
>> Alternatively you could remove xwiki'ds jackrabbit version from WEB-INF/lib (make sure the version used by jboss is greater than the one used by xwiki).
>>
>> Thanks
>> -Vincent
>>
>>> Can anyone see what I'm doing wrong? All the information on the web seems a
>>> little old – I'm using JBoss 6. I would be grateful if anyone could help
>>> shed any light on this issue.
>>>
>>> Cheers,
>>> Andy
--
Sergiu Dumitriu
http://purl.org/net/sergiu/
Hi,
Not sure I'm using the right channel but here is the deal. We were looking for a Tomcat-based Wiki and XWiki seemed to fit the bill (and in addition work with PostgreSQL that we also use). Unfortunately, this product, although I'm sure it's serious and works under certain strict contingencies, seems to have a very blocking bug under some other circumstances, so we will have to dig a bit more for something else, at least until your XWiki is more sturdy. But at least, having put a couple hours on it without results, we will report on our test: we installed version 3.1 (and then did the same test with version 3.0 with the same outcome). We installed the .war on Tomcat as documented in your web pages and configured it accordingly. In the end, that seemed to work in the sense that when we asked for the default page in a browser, we got an answer and also we saw that 41 tables were created in the database. As suggested, we then downloaded the official .xar file (we tried version 3.0 and then version 2.7 also) and uploaded it in the Import page. Then when we clicked on the link of the filename at the lower-left, the upper-right displayed: Archive xwiki-enterprise-wiki-3.0.xarDescription My packageVersion 1.0.0Author XWikiLicence LGPLselect none, all
as expected, BUT THE GREY INNER FRAME THAT SHOULD DISPLAY A LIST OF CONTENT IN THE XAR IS... EMPTY, YES TOTALLY EMPTY. So clicking on the Import button will yield a message about selecting something but there's nothing to select. We tested the same in Firefox, IE and Chrome. There are no error message in HTML or JavaScript, nor anything special in any of the Tomcat logs or PostgreSQL log. The .xar files are valid in the sense that they can be fully decompressed by WinZip/WinRAR. Our hypothesis is that some XWiki code (probably the compression library) silently traps an exception and is so far incompatible with either Tomcat version 7 or Java version 6 or Windows 7 or Intel 64-bit. Sorry for you and good luck. A. R.
Hi devs,
This is a thread to discuss what kind of title behavior we want in the future. I've listed some topics I consider open below. Could you give me your opinion about them?
1) Should we force titles to be mandatory by default
2) Should we merge page names and titles in the UI (ie only expose titles and compute page names based on it)
3) Do we want titles to be generated from content if the title is empty or not. If 1) is agreed then there'll always be a title set so this is probably not needed
4) Should we allow wiki syntax in the title field and make that field a textarea rather than a single line string (to allow scripts to generate titles, this is useful for applications for example). Note that currently the title field can only contain velocity and no wiki syntax AFAIK.
Any other topic to discuss related to titles?
Thanks
-Vincent
Hi,
First of all, I am very excited that Xwiki Community accepted my proposal
"Auto Completion in Content Editors", and thanks for everyone that helped me
a lot during the application time.
Second, I will talk about something I have done these days, and then talk
about the problems that confuse me.
1. I have setted up an Linux(Ubuntu 11.04)+Git+maven(2.2.1) environment for
developing Xwiki projects. Everything goes well in this step.
2. And I have downloaded the source code of xwiki-enterprise from github,
build it up, and got xwiki-enterprise-jetty-hsqldb-3.1-SNAPSHOT.zip file
under distribution folder, unzip it and run the start_xwiki.sh, and visit
http://localhost:8080, it is successful to see the xwiki page.
3. I have searched some issues related to WYSIWYG editors, and I finnally
chose two for optionals:http://jira.xwiki.org/jira/browse/XWIKI-4471 and
http://jira.xwiki.org/jira/browse/XWIKI-4179.
Problems:
1.Building XE takes a long time about more than half an hour, if I am
working on XE, for example fix some bugs of it, and want to compile and run
it as soon as possible to see the results, have an hour would be a very long
time for waiting each time. I know maybe the second time for building XE
will be much faster than the first time, but it still take some time to
build the whole codes, it must slow down the efficiency to debug my code.
So anyone can tell me how should I do, or is there any best prictices for
debuging and compling the code based on XE.
2. About the selected two issues, one is to add a small new feature, and the
other one is to fix a bug, I want to ask for some advices about which issue
would be ok, and could you give me some more informations for these two
issues.
--
Best wishes,
许凌志(Jame Xu)
MOE KLINNS Lab and SKLMS Lab, Xi'an Jiaotong University
Department of Computer Science and Technology, Xi’an Jiaotong University
Hi,
First of all I would like to thank XWiki for selecting me to the
Google Android Client for XWiki project through the GSOC. As a
beginning I'm currently studying on building XWiki public repositories
like xwiki-enterprise, xwiki-manager, xwiki-platform etc. so that I
can get the domain knowledge. I hope to fix some jira issues in these
repositories. Can anybody tell me some other way to get proper
understanding in XWiki appropriate to this Android project?
Thank you,
Chamika Weerasinghe
Hi devs,
I have a small issues with the comments on github: they are invisible to the xwiki dev team right now. We used to comment on commit mail diffs which made the comments available to everyone (including non developers) following the notifications/devs lists. Right now we have started using github for commenting but this means I don't see anymore when stuff get commented. I have to go to github specifically and look out for comments.
Is there a better way combining our lists + github to keep the open collaboration aspect of reviews?
Thanks
-Vincent
fyi: http://feedly.com/k/fZWfhP
Exceprt from the release notes:
Highlights include:
• Fixed numerous bugs that break the compressor and/or the resulting minified files.
• Added documentation on what exactly the minifier does and also which CSS hacks it tolerates.
• There’s a JavaScript port of CSS min in case it’s more suitable for your build process.Here’s also a test web UI that uses the JavaScript port, where you can experiment with the minifier.
• A significant number of new tests added (but you can add even more).
• Safe handling of some CSS features that are getting more adoption such as media queries and CSS3 transforms.
We could upgrade maybe.
Thanks
-Vincent