>>>>> On Wed, 26 Sep 2012 19:46:21 +0200, Frank Caputo <frank_at_frankcaputo.de> said:
FC> 1) vdldocs/facelets/jsf/tld-summary.html has still the old example
FC> for the non TagDecorator approach.
Frank, I think your most recent patch fixed this. Can you please
confirm by checking that [1] is as you expect?
FC> 2) The javadoc for TagDecorator says: "If no matching entry is
FC> found, let jsf:element be the value of targetTag". But the
FC> implementation still has the default behavior, which would map
FC> <input type="unknown" ...> to h:inputText. So we should add the
FC> mapping localName=input, selector attribute empty, target tag
FC> h:inputText.
This also was addressed in your most recent patch, but I will ammend the
text in the "selector attribute" column for that row to include this
(this means that if none of the other input types match, this row will apply)
FC> 3)
FC> renderkitdocs/HTML_BASIC/javax.faces.Inputjavax.faces.passthrough.Input.html
FC> says: "Look in the passthrough attribute map for a key given by the
FC> value of the submittedValueAttributeName entry. If found, use that
FC> value to look in the requestMap.". I think the key for the
FC> requestMap must be the clientId.
I have taken your suggestion to repurpose and rename that renderer to
just be PassthroughRenderer, so this statement no longer applies.
Ed
--
| edward.burns_at_oracle.com | office: +1 407 458 0017
| homepage: | http://ridingthecrest.com/
[1] https://maven.java.net/service/local/repositories/snapshots/archive/javax/faces/javax.faces-api/2.2-SNAPSHOT/javax.faces-api-2.2-20121031.083757-87-javadoc.jar/!/vdldocs/facelets/jsf/tld-summary.html