ExtendedDataTable column width policy

[prabhat.jha]

Following is how I am using ExtendedDataTable:

<rich:extendedDataTable value="#{bean.messages}" var="msgs" id="table">
 <rich:column label="Name">
 <f:facet name="header">
 <h:outputText value="Name" />
 </f:facet>
 <h:outputText value="#{msgs.senderScreenName}" />
 </rich:column>
 <rich:column label="Message">
 <f:facet name="header">
 <h:outputText value="Message" />
 </f:facet>
 <h:outputText value="#{msgs.text}" />
 </rich:column>
 </rich:extendedDataTable>

Ideally this should be rendered as table with 2 columns. Please note that I have not explicitly specified any default dimension.

This table is rendered as table with three columns out of which 2 columns have data and third one is empty. Looks like default width specified by renderer does not split available spaces equally. Of course, column separator is draggable so you could drag it to fix your need.

I have discussed this with Nick and he thinks this can be treated as a bug. I wanted to throw it out in the wild before I create a jira.

[jbalunas]

Seems list a reasonable bug to me :-)

[ilya_shaikovsky]

yes.. we used empty column in scrollable and extended table in order to reserve free space for resizing functionality. So seems good time to revise this. Customers also suffering from "unnecessary" column.

[prabhat.jha]

There is a concern for backward compatibility as 3.3.2 is a minor release and we don't want to change default behaviour in a minor release. I would like to change the fix version for https://jira.jboss.org/jira/browse/RF-7611 to 4.0. Please reply if you have any objection otherwise you can keep your fingers shut. ;-)

[jbalunas]

Agree - should not change default behavior in a point release.

[nbelaevski]

Moved the issue to 4.0.0.BETA1.