Re: [wtp-dev] JSP Editor: two kinds of validation?

[David M Williams <david_williams@xxxxxxxxxx>]

There's two types in all kinds of editors
... the concept is that "as you type" validation reflects the
*unsaved, unbuilt* contents of the document you are editing. 
So, that's where you see underline.
The problems-view type markers should only reflect the *saved, built* state
of the document. 
In general, all editors should behave
similarly to the Java Editor in this regards. 
So .. given that conceptual orientation,
perhaps you could open specific bugs if/when you see things that could
be improved?

Cameron Bateman <cameron.bateman@xxxxxxxxxx>
Sent by: wtp-dev-bounces@xxxxxxxxxxx

03/27/2006 10:39 PM

Please respond to
"General discussion of project-wide or architectural issues." <wtp-dev@xxxxxxxxxxx>

To	wtp-dev@xxxxxxxxxxx
cc
Subject	[wtp-dev] JSP Editor: two kinds of validation?

There seems to be two types of validation in the JSP
editor.  One kind 
find problems like this:

<foo:bar>

where foo:bar is a tag that doesn't exist.  This kind of validation
is 
normally instant (as you type) but it does not update markers.  The
problem is underlined in the editor but neither the Problems view or the
navigator are decorated to tell you there is a problem even after you do
a clean build.

The second kind catches things like this:

<%@ taglib uri="http://java.sun.com/jsf/html" prefix="h"%>
<%@ taglib uri="http://java.sun.com/jsf/html" prefix="h"%>

Where the same prefix is used twice.  This second case is identifiably
validated using the IValidator framework.

This raises the following questions in my mind:

1) Why are there two modes of validation?
2) Why is most validation is done in this first way that seems less 
satisfactory given that is triggers no markers or build warnings?


--Cam
_______________________________________________
wtp-dev mailing list
wtp-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/wtp-dev