You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@struts.apache.org by "Lemke, Volker" <vo...@infor.de> on 2000/09/14 19:16:32 UTC

Internet-Explorer and cache

Hello everybody,

I have recognized a problem with the 'wonderful' MS-Internet-Explorer:
Due to the fact that some of my JSP-Pages contain dynamic content (which is
included via 'jsp:include page..') I recognized that the 
MS-IE aborts the datatransfer to tomcat and takes a cached version. I
switched my ActionServlet to nocache=true.
Netscape works fine with this adjustment but the MS-IE still takes the
outdated version from its cache !
To have a workaround I changed my testapplication to append always a
timestamp to the response-URL, which makes it
impossible for the IE to find the page in its cache.
However this works, I would prefer a cleaner solution to handle this..

If you have any ideas to support me you can mail me at:
mailto:volker.lemke@infor.de 
:-)

Freundliche Grüße
Volker Lemke
Entwicklung Dorsten
Development department Dorsten


infor business solutions AG
Niederlassung Dorsten
Barbarastraße 70
46282 Dorsten
Tel.: 	+49 (0)2362/ 973-0
Fax: 	+49 (0)2362/ 973-111
Mobil 	+49 0170 3349 260

mailto:volker.lemke@infor.de
http://www.infor.de

Vorstand: Hubert Becker (Vorsitzender), Andreas Reinicke (stv.
Vorsitzender), Günter Löchner, Ludwig Augustin, Markus Kießling
Aufsichtsratsvorsitzender: Prof. Dr. Joachim Hertel
Sitz der Gesellschaft: Friedrichsthal/Saar. Handelsregister: Amtsgericht
Saarbrücken. HRB 11470   


BeanUtils

Posted by Ranjan Dasgupta <ra...@ps.net>.
Hello

In the struts framework's sample implementation of the forms, the validations
(String [] validate) has code like:

if ((firstName == null) || (firstName.length() < 1))
         add to error vector
.
.
.
Could it be done by something like:

if (BeanUtils.getScalarProperty(pif, "firstName") == null .......)
    add to error vector

By doing this we can have a list of required fields (their matching getters),
iterate thru that list and generate the errors accordingly.

Is there a reason for not doing so? I mean bean introspection is slow,
concurrency, or something?

Thanks.
-Ranjan