So, again, what if the cancel button (the patch that started this whole thread) was patched to allow the proposed behavior, but didn't require javascript? Meaning, it used javascript if possible, and if the browser didn't use it provided the old/default method?
The javascript in question is cross-platform, I use it in IE, Mozilla (windows/linux/mac) and Safari, so that really isn't an issue. The problem is Struts provides this tag, and it doesn't jive with a good user interface paradigm (hitting enter in a form should submit it, not cancel it ~ all browsers support this). Struts needs to be clear in it's goals. If it is to be just a controller, then it should be just a controller. For the most part everything can be handled by JSTL without headaches. This however, is an instance where the view returns a value to the controller (hitting the button sets a form value to true) signalling if the form was cancelled. So, this is a very struts specific issue (the value it sets is: org.apache.struts.taglib.html.CANCEL) and this is the propery that Action looks for. On Wed, 23 Jun 2004 11:16:34 -0700, Wendy Smoak <[EMAIL PROTECTED]> wrote: > > > From: Edgar P Dollin [mailto:[EMAIL PROTECTED] > > 2) JSTL is a waste of time. The reason I say this, not counting the > > non-java people, is if you can write x number of lines of > > useful code per > > hour, with jstl that is reduced by a factor greater than 1 due to type > > checking, refactoring support, testing difficulties, etc. > > Speak for yourself, I use Struts-EL and JSTL quite happily. One day I > would like to generate XML for the "view" and use XSLT to transform it > into HTML to which a CSS is applied, but I don't yet know how. > > > 4) I am still wondering why struts became the standardized > > html nazis. This makes absolutely NO sense. > > It would be chaos otherwise, as people introduced non-standard stuff > that works in one browser but breaks another. Struts also requires > adherence to the JavaBeans specification, nobody complains about that. > > > Any tag that has a clean interaction with a struts object or > > bean should be considered for tag libs assuming either very general > usage or > > real benefit in specialized situations. > > Why the insistence on getting things immediately accepted into the core? > If you put it out there, and enough people like it and use it, it will > become a defacto standard. > > I don't care where the tags live. I do care that they generate clean > HTML and don't require JavaScript. If you want JavaScript and > non-standard HTML, then extend the Struts tags or write your own, > release them, and see what happens. > > -- > Wendy Smoak > Application Systems Analyst, Sr. > ASU IA Information Resources Management > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
