- From: Alexander Anokhin <ava@vaz.ru>
- Date: Tue, 13 Dec 2005 15:53:49 +0400
- To: John Boyer <boyerj@ca.ibm.com>
- CC: www-forms@w3.org, www-forms-request@w3.org
- Message-ID: <439EB64D.90100@vaz.ru>
John Boyer wrote: > > In the next day or two, the new 1.1 working draft will appear. > In it, you'll see that we've scrapped duplicate and destroy in favor of > simple modifications of insert and delete that cover the functionality > of duplicate and destroy. > > The main problems for insert were how to identify a nodeset that > could become empty and how to get a prototypical subtree from > someplace else. > > By simply adding the new attributes to insert rather than inventing a > new action, we found we could achieve the desired results. We also > found that insert's separation of the target location into nodeset, at > and position made it easier than was the case with duplicate to > express the most common use case of inserting a new homogeneous > collection item into a repeat after the currently indexed repeat item. > Also, an insert with a target of the root document element of the > instance results in replacing the instance. > > Details and examples to be published shortly. > Thanks for quick reply, John. Completly agree, but the problem is to achieve such functionality in XForms 1.0 only, since very few client-side processors implements 1.1 WD (afaik FormsPlayer only?). As i see for now it can't be done without using of Javascript to make FireFox XForms extension behave like in 1.1 WD. Main problem of client-side XForms for me is poor portability. Sutuation seems like in HTML world while ago - we've one spec and many specialists, whose primary job is to make HTML-pages looks similar in different browsers. So articles like "How to make crossbrowser XForms" makes me worry for the future of this technology ;). and sorry for casual english. -- Alexander Anokhin AVTOVAZ JSC email: ava@vaz.ru icq: 123275798
Received on Tuesday, 13 December 2005 11:54:29 UTC