[ https://issues.jboss.org/browse/RF-13235?page=com.atlassian.jira.plugin.s... ] Lukáš Fryč updated RF-13235: ---------------------------- Description: The row key generation for Tree stopped to work: *got:* <div id="j_idt106:tree:org.richfaces.model.SequenceRowKey[0, 0, 0]:j_idt111" class="rf-tr-nd rf-tr-nd-lf"> *expected:* <div id="j_idt106:tree:0.0.0:j_idt111" class="rf-tr-nd rf-tr-nd-lf"> ---- The cause was that rowKeyConverter resolves to null, so plain toString on rowKey is used. I was wondering why rowKeyConverter isn't resolved for Tree in Showcase - the problem is that CDK generates this attribute and thus its implementation in AbstractTree (which is the correct way of rowKeyResolution) isn't used at all. I'm pretty sure this is caused by refactoring from tree-model-props.xml to use Java interfaces TreeProps and TreeCommonProps: https://github.com/richfaces/richfaces/commit/62ab0fe3f00b44a166b522ecb71... However it's not clear why the rowKeyConverter wasn't generated in RF 4.3 when it wasn't marked as generate=false: https://github.com/richfaces4/components/blob/master/iteration/ui/src/mai... Note that in RF 5 it is generated because UIAdaptor has @Attribute which defaults to generate=false: https://github.com/richfaces/richfaces/blob/master/framework/src/main/jav... Workaround is marking rowKeyConverter as generate=false in AbstracTree (or maybe RowKeyConverter), but we should rather investigate why this issue happens. ---- This issue has bigger potential since during refactoring of properties, [~bleathem] could work with the same assumption as I did ^. was: The row key generation for Tree stopped to work: *got:* <div id="j_idt106:tree:org.richfaces.model.SequenceRowKey[0, 0, 0]:j_idt111" class="rf-tr-nd rf-tr-nd-lf"> *expected:* <div id="j_idt106:tree:0.0.0:j_idt111" class="rf-tr-nd rf-tr-nd-lf"> ---- I was wondering why rowKeyConverter isn't resolved for Tree in Showcase - the cause is that CDK generated this attribute and thus its implementation in AbstractTree isn't used at all. I'm pretty sure this is caused by refactoring from tree-model-props.xml to use Java interfaces TreeProps and TreeCommonProps: https://github.com/richfaces/richfaces/commit/62ab0fe3f00b44a166b522ecb71... However it's not clear why the rowKeyConverter wasn't generated in RF 4.3 when it wasn't marked as generate=false: https://github.com/richfaces4/components/blob/master/iteration/ui/src/mai... Note that in RF 5 it is generated because UIAdaptor has @Attribute which defaults to generate=false: https://github.com/richfaces/richfaces/blob/master/framework/src/main/jav... Workaround is marking rowKeyConverter as generate=false in AbstracTree (or maybe RowKeyConverter), but we should rather investigate why this issue happens. ---- This issue has bigger potential since during refactoring of properties, [~bleathem] could work with the same assumption as I did ^. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira