Hi,
I can't find any information on the relevance scoring/weighting in XWiki's
Lucene Search. Haven't I looked close enough? From what I understand about
Lucene and XWiki, each page and attachment is indexed with several fields,
such as page title, page name, contents, tags, parent, not creator..
The relevance scoring doesn't show how it's computed. I figure the
percentage of hits per field is multiplied with a weight per field (eg: a
hit in the title is scored higher than a hit in the content, and if the
title 100% matches it scores higher than a longer title where the query
only occurs in one of 10 words).
When title or pagename match the search query, the relevance always seems
100%.
Are my assumptions correct?
What fields are taken into account and by what weight?
Between results that have 100% relevance, how is sorting done?
Thanks for explaining!
Joris
Hi,
In XWiki 1.7 the following code was created; it creates a html table with
links generated by #pagedViewLinks()
It now displays the #pagedViewLinks() but no longer the links
What is the current way of doing this (migrated to XWiki 3.5.1)?
<table id="cdlsTable"
class="grid sortable filterable doOddEven"
cellpadding="0"
cellspacing="0"
border="0">
<tr class="sortHeader">
<th>$msg.get('Subject')</th>
<th>$msg.get('lst_modified')</th>
</tr>
#foreach ($faq in $faqlist)
<tr>
#set ($faqdoc = $xwiki.getDocument($faq))
#set ($faqobjs =
$faqdoc.getObjects("CdlsatdCode.CdLS_ATD_PublicFAQClass"))
#foreach ($faqobj in $faqobjs)
#if($faqobj.language == $lang)
<td>[[$faqobj.subject>>$faq]]</td>
<td>$xwiki.formatDate($faqdoc.ContentUpdateDate, "dd/MM/yyyy")</td>
#end
#end
</tr>
#end
</table>
#pagedViewLinks($itemsPerPage $totalPages $crtPageNumber)
Hi,
Because of memory problems with our (small) server, I've decided to
switch from hibernate storage to file storage for attachments.
When I launch the filesystemAttachmentPorterScript-2.0.xwiki20 groovy
script, the script prints :
"This is your storage directory, when filesystem attachments are enabled
you have to back this up as part of your data backup routine.
/usr/apache-tomcat-5.5.25/logs/data/storage"
The value is the result of
"Utils.getComponent(FilesystemStoreTools.class).storageDir.getAbsolutePa
th()" command (tomcat was started from logs directory).
The xwiki doc says :
"The directory in which the attachments are stored in the filesystem is
defined with the parameter xwiki.work.dir in the xwiki.cfg file. "
But the value in my xwiki.cfg is :
xwiki.work.dir=/usr/apache-tomcat-5.5.25/webapps/xem/work
If I try the command below in a groovy script, the value is what I want,
but not the value
"Utils.getComponent(FilesystemStoreTools.class).storageDir.getAbsolutePa
th()" prints :
xwiki.getXWiki().getConfig().getProperty('xwiki.work.dir') =>
/usr/apache-tomcat-5.5.25/webapps/xem/work
It seems to me that the value of xwiki.work.dir is not used, only the
default value.
That's a bug ?
Thank you
Nicolas CG
hi all,
id like to use the Space-Index macro to list all pages on the WebHome page of every Space. With {{spaceindex count="0"/}} i get a list which is sorted by the time the page was created (or last modified?), but i'd like a list which is sorted in alphabetical order. Is that possible?
thanks in advance,
zae.
On 05/21/2012 03:08 AM, Moritz Hesse (EA GmbH) wrote:
> Hi,
>
>
>
> I have created two applications, let's say one for managing clients and one
> for managing projects. How can I fetch the clients name and use it in my
> list of projects?
>
>
>
> To speak in database terms I want my project's table to have a foreign key
> which refers to the client primary key so that I can access all client's
> attributes.
>
>
>
> Would be thankful for inspiration or helpful links to existing(?) tutorials.
>
>
>
> Thanks and bests,
>
Maybe this can help you:
http://www.xwiki.org/xwiki/bin/FAQ/HowToCreateAOneToManyRelationshipInXWiki
--
Sergiu Dumitriu
http://purl.org/net/sergiu/
Hello,
I received a mail from a xwiki user who request xar of my macro
because it was zero sized in extension page.
After login and updating it, I see in my recent pages there are two
pages for same macro:
- Xslt Macro: http://extensions.xwiki.org/xwiki/bin/view/Extension/Xslt+Macro
- XSLT Macro: http://extensions.xwiki.org/xwiki/bin/view/Extension/XSLT+Macro
I don't remember create it twice, so may be there are some page name
rule apply when migrating to extensions pages and someone duplicated
it.
Which page should I delete?
Regards,
Arnaud.
Hi Moritz,
On Mon, May 21, 2012 at 10:08 AM, Moritz Hesse (EA GmbH)
<moritz.hesse(a)ea-gmbh.de> wrote:
> Hi,
>
>
>
> I have created two applications, let's say one for managing clients and one
> for managing projects. How can I fetch the clients name and use it in my
> list of projects?
>
>
>
> To speak in database terms I want my project's table to have a foreign key
> which refers to the client primary key so that I can access all client's
> attributes.
>
>
>
> Would be thankful for inspiration or helpful links to existing(?) tutorials.
>
Each application has a class that describes the type of data managed
by the application (the client class and the project class in your
case). By default a live table can display information taken from the
instances of a single class. If you want a live table that aggregates
('join' in database language) data from instances of multiple classes
then you have to write a custom live table results page. Live tables
are fed with JSON generated by a wiki page. By default this page is
XWiki.LiveTableResults but you can create your own results page. It's
not trivial though..
Useful links:
http://extensions.xwiki.org/xwiki/bin/view/Extension/Livetable+Macrohttp://extensions.xwiki.org/xwiki/bin/view/Extension/Query+Module
Hope this helps,
Marius
>
>
> Thanks and bests,
>
> Moritz
>
> _______________________________________________
> users mailing list
> users(a)xwiki.org
> http://lists.xwiki.org/mailman/listinfo/users
Hello,
On the xwiki 3.2, we have decided to use the virtual xwiki (multi-tenant) feature as it should be better for scalability and performance
http://platform.xwiki.org/xwiki/bin/view/AdminGuide/Virtualization
"Each wiki has its own database schema, thus allowing for a complete data separation between the different wikis in the same instance.
The main advantage of multi-tenancy is performance optimization."
Problem:
The main wiki has a relative good level of performance - with a response time around 2.5 second to browse a page
So we are in the xwiki database
But when we are in a virtual wiki (another schema ) - the performance are very poor.
The average response time is around 5 seconds per page !!! So at list 2 times slower.
We are using MySQL and glassfish 3.1 - and we have done tuning on the MySQL database and glassfish.
We have used a profiling tool and see that the request send to the virtual environment are raising a big amount of exception ...
The problem is not on the database part (even if they are many SQL calls to visualize only one page)
The problem seems really to be on the virtual environment configuration OR in a file corruption ...
I have attached the exception list . (exception_list.xls)
>From this is it possible to know if we have an installation issue ??
Does someone have experience of very good performance with xwiki ? ( response time #1second per page)
In that case
Best regards,
Olivier
_______________________________________________
Olivier Merlin
RnD Tuning/profiling community Leader
Tel: + 33.4.42.36.59.87
olivier.merlin(a)gemalto.com<mailto:frederic.olier-EXTERNAL@gemalto.com>
_______________________________________________
Hi,
I have created two applications, let's say one for managing clients and one
for managing projects. How can I fetch the clients name and use it in my
list of projects?
To speak in database terms I want my project's table to have a foreign key
which refers to the client primary key so that I can access all client's
attributes.
Would be thankful for inspiration or helpful links to existing(?) tutorials.
Thanks and bests,
Moritz
We have a small Confluence Wiki that is approaching its user limit. We have the option of either purchasing additional licenses or moving to XWiki. One huge advantage of XWiki is that we could vastly increase the number of users with access to our Wiki even if we did not allow those users write access.
In order to make such a conversion feasible, we would first need to import our existing Confluence Wiki. In searching around I have found references to some Google Summer of Code proposals < http://dev.xwiki.org/xwiki/bin/view/GoogleSummerOfCode/ImportExportfromanyo…>. I have been less successful in locating implementations.
Is there a (relatively) straightforward way to import my Confluence content into an XWiki? How might I go about doing this?
Thanks
Michael Ottati