Re: probably a cleaner way... testing for just one user
Hi Frank, Doesn't the sessionCount get decremented regardless of whether the session is being destroyed as the result of a regular logoff or timeout as well as if it was a rejected logon (i.e., max sessions already in use)? I dealt with this problem in the code I'm adding to Java Web Parts tomorrow, but maybe you found a better way than I did :) Yes, the sessionCount is decremented everytime a session is destroyed regardless of regular logoff, 'forced' logoff (max sessions in use) :-)) or timeout. But that's what I wanted. Do you see a problem with it? sessionCount is the number of existing sessions, so I decrement the sessionCount everytime a session is destroyed(no matter why it was destroyed). Also, I was a bit surprised to learn that filters fire BEFORE listeners... it seemed more logical the other way around, but it's not reality. Any problem with what you've done with that in mind? A SessionContextListener's sessionCreated method will execute when a new HttpSession is created. HttpSessions are created by some servlet or jsp code. So if you use a Filter that catches any request, the Filter will fire before any code that could create a session has a chance to execute. If the request is not for the 'special' appInUse.jsp or logout.jsp(by the way I think that it's ok if we only treat appInUse.jsp specially, logout.jsp can be treated like all other resources) then I always get a session in the usual way: HttpSession session = request.getSession(); As you know this will create a session if one doesn't exist or it will get an existing session. Because this is a Filter that catches everything, all the sessions will be created at this point. This is also the point where the sessionCreated will be executed if the session.isNew(). In the sessionCreated a session is always created, but if maxSessions was exceeded then we flag the session. Back to the filter code, if the session is flagged then we invalidate it and we redirect. It is always benefic to try to write down in words what your code does. I think I found a a little problem in my solution :-) : Ex.: MAX_SESSIONS = 1 and there is one active session (User A) User B tries to connect, filter fires, flagged session is created sessionCount is 2 now. We are back in the filter before session.invalidate(). In another thread User A's session is destroyed and the sessionDestroyed() is executed which decremenets sessionCount. sessionCount = 1 User C comes in, in another thread the filter fires for his request, but he will get a flagged session too, 'cause sessionCount is still 1. However this is a minor issue and it probably can be easily changed by not incrementing/decrementing sessionCount in sessionCreated()/sessionDestroyed() if the session is flagged. This is what you suggested in the first place? :-) Tamas - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Problem with Dyna Form
Hi All, My requirement is to generate a set of dynamic rows(with checkbox and combo box) at client side using JSP and Struts 1.1. As of now, i am generating the above stuff using java script. How can i use Dyna Form to meet the above requirement (or) suggest me other way if u have any solution for this. -Thanks in Advance Jeevan __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Re: [OT] Create/Edit Form
You need an identifier to update an object, which usually corresponds to a PK in your database. When the identifier is not present or does not correspond to a persisted entry, you probably want to create. hth, Manos Andrew Tomaka wrote: Hey all, The following is pretty hard to understand. I'm no good at describing my problems and wasn't sure how to explain it. Hopefully, someone will get what I'm attempting to say I'm working on a screen that allows a user to dynamically add rows via Javascript. I then have have a form that contains a list of beans that stores the data (but that's really beside the point). My issue comes with the storage of this information. The form can either be a blank form or it can be preloaded with other information depending on whether the record exists or not. If it is preloaded, the user can edit the existing rows or add new rows. The problem comes when I need to update the database. When the user hits the save button, it submits all the information, but I have no way of telling if a specific row was edited or created. Because of this, I don't know whether to make an UPDATE query or an INSERT query. Can anyone think of a creative way to do this without adding an extra query for each row? Any insight would be greatly appreciated! Thanks, ~ Andrew Tomaka - 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]
Re: JSF is the beginning of the end of Struts !!!
news to some: http://bugs.sakaiproject.org/confluence/pages/viewpage.action?pageId=4981 I think it was said us Java dudes should talk less and produce apps more in this thread, which is what I said +1 too. It's hard to debate PHP vs Java on the server, which is better, which is bigger... bla, bla. I know you can't do http://roomity.com/launch.jsp in PHP, that would be JDNC. And as of tmrw night it will run Linux-Mozilla and OSX-Saffari, not just Windoze!. So if somone says why Java, you say RiA! won't you. Lets see somone do that. (and ah... there is allways the business benefit of not having to post your email to mail groups... but we use apps for tech benfit :-) ) hth, .V ps: Like Ted said... sometimes we just use MVC we learned in Struts in other projects, and my JDNC impl is all MVC, developed by Struts dudes. John Public wrote: Make no mistake about it. JSF rocks. JSF is based on true MVC architecture, not the MODEL 2 version that Struts uses. J.J. JSF Developer (former Struts Lead Developer) -- .V People are conversing... without posting their email or filling up their mail box. roomity.com http://roomity.com/launch.jsp No sign up to read or search this Rich Internet App. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: mutli-checkbox
Hi Sorry for bothering u guys , I could figure out the mistake But now the problem is , I dont get the values vback in the Action (when submit is hit). I am using form oy type DynaValidatorForm and my Form Bean has all the properties of type String Can u pl help. Thanks Floowing is the JSP Page :- %@ page language=java import=java.util.*% %@ page import=org.questforum.fwork.struts.DisplayLists% %@ page import=org.questforum.fwork.WP.UserWP% %@ page import=org.questforum.fwork.struts.StringConstants% %@ page import=org.questforum.fwork.core.perm.UserPermission% %@ page import=org.questforum.fwork.struts.GeneralService% %@ page import=org.questforum.fwork.struts.WrapperClass % %@ taglib uri=http://struts.apache.org/tags-bean; prefix=bean% %@ taglib uri=http://struts.apache.org/tags-html; prefix=html% %@ taglib uri=http://jakarta.apache.org/struts/tags-logic; prefix=logic% %@ taglib uri=http://java.sun.com/jstl/core; prefix=c% % String strContactId = request.getParameter(contactId); UserWP objUserWp = (UserWP) session.getAttribute(StringConstants.strUserObject); String []listOfAllPerms = DisplayLists.getAllUserPerms (); WrapperClass userPerms = DisplayLists.getPermissionForUser(Integer.parseInt (strContactId)); pageContext.setAttribute(userPerms, userPerms); pageContext.setAttribute(listOfAllPerms, listOfAllPerms); % html head titlebean:message key=editUserPerm.page.title //title /head body bean:message key=editUserPerm.page.title / br html:form action=editUserPermAction logic:iterate id = itrListOfAllPerms name=listOfAllPerms html:multibox name=userPerms property=collectionObj bean:write name=itrListOfAllPerms / /html:multibox bean:write name=itrListOfAllPerms/ br /logic:iterate br html:submit property=PermActions bean:message key=editUserPerm.back / /html:submit /html:form html:errors / /body /html __ How much free photo storage do you get? Store your friends 'n family snaps for FREE with Yahoo! Photos http://in.photos.yahoo.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: How to have logical 'OR' between two validator rules ?
Thanks for the reply,
This only allows you to compare fields, not to enter a regular
expression mask in the test var. So uit's not enought, except if I'm
totally wrong.
I don't need to have an 'OR' between fields, but between validator
results. That's the problem.
In fact, OR between validwhen and mask should be ok.
The only solution I'fe found until now is to write a new validator for me.
On 27/07/05, Glen Mazza [EMAIL PROTECTED] wrote:
Yes, much clearer now. If you're using Struts 1.2.x, There is a
validwhen clause that may help you--look at the test example just
below it:
http://struts.apache.org/userGuide/dev_validator.html#validwhen
Glen
Marc Demlenne escribió:
Yes, sorry I made a mistake while explaining. I'll try to do it better now.
Of course I need always to validate my form, you're right, sorry for
the confusion. But I need the form to be valid if either it is
validated threw the mask validator, OR if it is unchanged, even if in
this case it does not look like the mask wants any more.
So I want a my field to be something like ^[0-9]{4}$. Ok, I use the
mask validator, and everything is fine.
Now, if the value already stored in the DB isn't like this mask (there
are some good reasons to allow this), I want the user to be able to
modify everything in his form. If he doesn't change the value of this
field, everything should be correct. BUT if he decide to modify this
field, then he MUST enter a mask-valid field.
To cope with this, I can use another field, the same, but hidden,
containing the original value, for example. Then in my validation, I
would like,
- First to check if the given field is the same as the hidden one. In
this case it has not been modified, and I say validation returns true.
- Then, if those 2 fields are not the same, I check the real field
(the one enterered by user) against the mask threw the mask validator,
and I only accept all of this if this last validation returns true.
Hope it's better explained ?
Thanks Joe for this wonderful link. I have not yet verified
everything, but I'm affraid it won't be satisfying in my case ... I'm
still investigating ...
On 27/07/05, Glen Mazza [EMAIL PROTECTED] wrote:
Marc Demlenne wrote:
Hi and thanks for the answer,
In fact my problem could be explained the following way :
Having two fields A and B, I need to validate A if
(A == mask) OR (A == B)
Your way of stating the problem doesn't seem right and may be adding to
the difficulties here--I would again think that you *always* need to
validate A, it's just that validation passes 100% if the above two
conditions fit, but passes/fails (depending on the results of more
complex logic) if the above isn't correct.
I.e., instead of saying, If a ZIP code is provided, I need to validate
that it is five digits long, why not: I need to validate that the ZIP
code is either empty, or if not 5 digits?
Again, I'm unsure here (sorry, still a newbie), but using the p. 38
example below, my inclination is that you can always validate but just
change the validation logic if the scenario above occurs, something like:
(Pseudocode):
validate() {
if (A.equals(mask) || a.equals(B)) {
return null; // validate always true
} else {
// do actual validation,
// return ActionErrors or null as appropriate
}
}
I hope someone else can provide a better answer for you though--as I'm
also quite curious here.
Glen
I don't know if it is possible to include a field in a mask
expression. But in the other case, I need to have an alternative
between the two validators. Is thos possible with struts validator ?
If this isn't possible, I'll need to create my own validator plugin
But if someone has eny advice, it's welcomed !
On 26/07/05, Glen Mazza [EMAIL PROTECTED] wrote:
I'm not exactly certain of your needs, but I think you can always
activate validation but just change the validation code based on the
value/state of other fields (such as the original one you mention).
Perhaps the Java code example on page 38 of
http://www.objectsource.com/Struts_Survival_Guide.pdf can be of help for
you--I am unsure.
Glen
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Marc Demlenne
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Struts-EL problem with struts-el-1.3.0-dev
Hello all,
I have an app which is using struts 1.3.. I wanted to 'replace'
existing struts tag with struts-el tag.
So I downloaded struts-el-1.3.0-dev and put it into my jar, (along with
struts-taglib-1.3.0-dev)
I have declared taglibs in web.xml, and I wrote a simple page which uses
Struts-el inside a logic:tag
And, to my surprise, it did't work
Basically I have an attribute of type collection in my HttpSession, and
I was trying to access it from the page using
Logic:present name=${sessionScope.results}
for doublechecking, I wrote a scriptlet code that was trying to get the
'results' attribute from the session, to see that if it was actually
there..
and it was there
Even the c:forEAch loop that uses JSTL to get that attribute works
fine..
for providing as much info as I can, here is my web.xml
** WEB.XML (TAGLIBS ONLY) *
!-- Struts Tag Library Descriptors --
taglib
taglib-uri/WEB-INF/struts-bean.tld/taglib-uri
taglib-location/WEB-INF/struts-bean.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-html.tld/taglib-uri
taglib-location/WEB-INF/struts-html.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-logic.tld/taglib-uri
taglib-location/WEB-INF/struts-logic.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-nested.tld/taglib-uri
taglib-location/WEB-INF/struts-nested.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-tiles.tld/taglib-uri
taglib-location/WEB-INF/struts-tiles.tld/taglib-location
/taglib
!-- Struts-EL Tag Library Descriptors --
taglib
taglib-uri/WEB-INF/struts-bean-el.tld/taglib-uri
taglib-location/WEB-INF/struts-bean-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-html-el.tld/taglib-uri
taglib-location/WEB-INF/struts-html-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-logic-el.tld/taglib-uri
taglib-location/WEB-INF/struts-logic-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-tiles-el.tld/taglib-uri
taglib-location/WEB-INF/struts-tiles-el.tld/taglib-location
/taglib
***
and here is my JSP (only relevant part)***
%@ taglib prefix='c' uri='http://java.sun.com/jstl/core' %
%@ taglib prefix=fmt uri=http://java.sun.com/jstl/fmt; %
%@ taglib uri=/WEB-INF/struts-html-el.tld prefix=html %
%@ taglib uri=/WEB-INF/struts-bean-el.tld prefix=bean %
%@ taglib uri=/WEB-INF/struts-logic-el.tld prefix=logic%
% if(session.getAttribute(results) != null) {
java.util.Collection coll =
(java.util.Collection)session.getAttribute(RESULTS);
System.out.println( in session there are + coll.size() +
elements!!!); // this print successfully
}
%
TR
TD colspan=4centerlogic:present
name=${sessionScope.results}hello/logic:presentc:out
value=${statusMsg}//center
/TD
/TR
c:forEach var=entry items=${sessionScope.results}
// this displays the loop data
/c:forEach
***
other than that, in my \lib directory I have both
struts-taglib-1.3.0-dev and struts-el-1.3.0-dev,
and I am running my code on Jboss 3.2.5
anyone could give me some hints on what's wrong?
Do you think the fact that I have both struts-taglib and struts-el in
my
Lib directory could cause problems?
Thanks in advance and regards
marco
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Struts-EL problem with struts-el-1.3.0-dev
At 11:08 AM +0100 7/28/05, Marco Mistroni wrote:
anyone could give me some hints on what's wrong?
Do you think the fact that I have both struts-taglib and struts-el in
my
Lib directory could cause problems?
No, because el actually depends on taglib -- you have to have taglib to use el.
If you're also using JSTL, I would always advise using c:if
whereever it works -- not that this shouldn't be fixed if it turns
out to be a bug.
Joe
Hello all,
I have an app which is using struts 1.3.. I wanted to 'replace'
existing struts tag with struts-el tag.
So I downloaded struts-el-1.3.0-dev and put it into my jar, (along with
struts-taglib-1.3.0-dev)
I have declared taglibs in web.xml, and I wrote a simple page which uses
Struts-el inside a logic:tag
And, to my surprise, it did't work
Basically I have an attribute of type collection in my HttpSession, and
I was trying to access it from the page using
Logic:present name=${sessionScope.results}
for doublechecking, I wrote a scriptlet code that was trying to get the
'results' attribute from the session, to see that if it was actually
there..
and it was there
Even the c:forEAch loop that uses JSTL to get that attribute works
fine..
for providing as much info as I can, here is my web.xml
** WEB.XML (TAGLIBS ONLY) *
!-- Struts Tag Library Descriptors --
taglib
taglib-uri/WEB-INF/struts-bean.tld/taglib-uri
taglib-location/WEB-INF/struts-bean.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-html.tld/taglib-uri
taglib-location/WEB-INF/struts-html.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-logic.tld/taglib-uri
taglib-location/WEB-INF/struts-logic.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-nested.tld/taglib-uri
taglib-location/WEB-INF/struts-nested.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-tiles.tld/taglib-uri
taglib-location/WEB-INF/struts-tiles.tld/taglib-location
/taglib
!-- Struts-EL Tag Library Descriptors --
taglib
taglib-uri/WEB-INF/struts-bean-el.tld/taglib-uri
taglib-location/WEB-INF/struts-bean-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-html-el.tld/taglib-uri
taglib-location/WEB-INF/struts-html-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-logic-el.tld/taglib-uri
taglib-location/WEB-INF/struts-logic-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-tiles-el.tld/taglib-uri
taglib-location/WEB-INF/struts-tiles-el.tld/taglib-location
/taglib
***
and here is my JSP (only relevant part)***
%@ taglib prefix='c' uri='http://java.sun.com/jstl/core' %
%@ taglib prefix=fmt uri=http://java.sun.com/jstl/fmt; %
%@ taglib uri=/WEB-INF/struts-html-el.tld prefix=html %
%@ taglib uri=/WEB-INF/struts-bean-el.tld prefix=bean %
%@ taglib uri=/WEB-INF/struts-logic-el.tld prefix=logic%
% if(session.getAttribute(results) != null) {
java.util.Collection coll =
(java.util.Collection)session.getAttribute(RESULTS);
System.out.println( in session there are + coll.size() +
elements!!!); // this print successfully
}
%
TR
TD colspan=4centerlogic:present
name=${sessionScope.results}hello/logic:presentc:out
value=${statusMsg}//center
/TD
/TR
c:forEach var=entry items=${sessionScope.results}
// this displays the loop data
/c:forEach
***
other than that, in my \lib directory I have both
struts-taglib-1.3.0-dev and struts-el-1.3.0-dev,
and I am running my code on Jboss 3.2.5
Thanks in advance and regards
marco
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Joe Germuska
[EMAIL PROTECTED]
http://blog.germuska.com
Narrow minds are weapons made for mass destruction -The Ex
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
RE: Struts-EL problem with struts-el-1.3.0-dev
Hello Joe,
Yes I actually reverted to using c:if, and it worked
Just fine
I wanted just to report the situation to see if I was
Using the taglib incorrectly...
Thanx again and regards
marco
-Original Message-
From: Joe Germuska [mailto:[EMAIL PROTECTED]
Sent: 28 July 2005 12:15
To: Marco Mistroni; 'Struts Users Mailing List'
Subject: Re: Struts-EL problem with struts-el-1.3.0-dev
At 11:08 AM +0100 7/28/05, Marco Mistroni wrote:
anyone could give me some hints on what's wrong?
Do you think the fact that I have both struts-taglib and struts-el in
my
Lib directory could cause problems?
No, because el actually depends on taglib -- you have to have taglib to
use el.
If you're also using JSTL, I would always advise using c:if
whereever it works -- not that this shouldn't be fixed if it turns
out to be a bug.
Joe
Hello all,
I have an app which is using struts 1.3.. I wanted to 'replace'
existing struts tag with struts-el tag.
So I downloaded struts-el-1.3.0-dev and put it into my jar, (along with
struts-taglib-1.3.0-dev)
I have declared taglibs in web.xml, and I wrote a simple page which
uses
Struts-el inside a logic:tag
And, to my surprise, it did't work
Basically I have an attribute of type collection in my HttpSession, and
I was trying to access it from the page using
Logic:present name=${sessionScope.results}
for doublechecking, I wrote a scriptlet code that was trying to get the
'results' attribute from the session, to see that if it was actually
there..
and it was there
Even the c:forEAch loop that uses JSTL to get that attribute works
fine..
for providing as much info as I can, here is my web.xml
** WEB.XML (TAGLIBS ONLY) *
!-- Struts Tag Library Descriptors --
taglib
taglib-uri/WEB-INF/struts-bean.tld/taglib-uri
taglib-location/WEB-INF/struts-bean.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-html.tld/taglib-uri
taglib-location/WEB-INF/struts-html.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-logic.tld/taglib-uri
taglib-location/WEB-INF/struts-logic.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-nested.tld/taglib-uri
taglib-location/WEB-INF/struts-nested.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-tiles.tld/taglib-uri
taglib-location/WEB-INF/struts-tiles.tld/taglib-location
/taglib
!-- Struts-EL Tag Library Descriptors --
taglib
taglib-uri/WEB-INF/struts-bean-el.tld/taglib-uri
taglib-location/WEB-INF/struts-bean-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-html-el.tld/taglib-uri
taglib-location/WEB-INF/struts-html-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-logic-el.tld/taglib-uri
taglib-location/WEB-INF/struts-logic-el.tld/taglib-location
/taglib
taglib
taglib-uri/WEB-INF/struts-tiles-el.tld/taglib-uri
taglib-location/WEB-INF/struts-tiles-el.tld/taglib-location
/taglib
***
and here is my JSP (only relevant part)***
%@ taglib prefix='c' uri='http://java.sun.com/jstl/core' %
%@ taglib prefix=fmt uri=http://java.sun.com/jstl/fmt; %
%@ taglib uri=/WEB-INF/struts-html-el.tld prefix=html %
%@ taglib uri=/WEB-INF/struts-bean-el.tld prefix=bean %
%@ taglib uri=/WEB-INF/struts-logic-el.tld prefix=logic%
% if(session.getAttribute(results) != null) {
java.util.Collection coll =
(java.util.Collection)session.getAttribute(RESULTS);
System.out.println( in session there are + coll.size() +
elements!!!); // this print successfully
}
%
TR
TD colspan=4centerlogic:present
name=${sessionScope.results}hello/logic:presentc:out
value=${statusMsg}//center
/TD
/TR
c:forEach var=entry items=${sessionScope.results}
// this displays the loop data
/c:forEach
***
other than that, in my \lib directory I have both
struts-taglib-1.3.0-dev and struts-el-1.3.0-dev,
and I am running my code on Jboss 3.2.5
Thanks in advance and regards
marco
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Joe Germuska
[EMAIL PROTECTED]
http://blog.germuska.com
Narrow minds are weapons made for mass destruction -The Ex
-
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]
UrlValidator
Hi, can it be that UrlValidator from Commons-Validator doesnt validate URLs with localhost in it? The following URL - http://localhost:8081/context/jsp/versand_1.jsp breaks in isValidAuthority() within UrlValidator. It seems he tries to check for toplevel domain length and sees my localhost as a toplevel domain. I am hoping i ve done something wrong because i would suspect that a UrlValidator validates this pretty straightforward URL. -- regards Marc Logemann [blog] http://www.logemann.org [busn] http://www.logentis.de - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Error Message Problem (Again)
Hi All,
Seems this is a constant area of irritation. I am attempting to put an
error message into the ActionMessage in an Action Class and redisplay this
on a JSP page. This application is using tiles and is located in a Struts
Module. Seems simple enough, but I can't get the darn thing to display on
the JSP.
Here is extracts from my code.
Action Class:
public ActionForward execute(final ActionMapping mapping,
final
ActionForm form,
final
HttpServletRequest request,
final
HttpServletResponse response) {
ActionForward forward = mapping.findForward(fail);
ActionMessages errors = new ActionMessages();
...
errors.add(displayerror, new ActionMessage(myerror, false));
...
saveErrors(request, errors);
}
JSP Page:
html:form action=/myAction.do method=post
html:messages id=error property=displayerror
bean:write name=error /
/html:messages
...
/html:form
I have also tried:
html:form action=/myAction.do method=post
html:errors/
...
/html:form
Neither work
I inspect errors on the save error line and can see the error message in
the ActionMessage list, so I know it is being added to the request. I also
used the false attribute to simply display myerror on the page to
eliminate the possibilty of a properties file typo problem. Any
recommendations on where to look?
RE: Error Message Problem (Again)
Hi,
Try the following:
Use saveMessages(request, messages) instead of saveErrors(request,
errors)
And use something like
logic:messagesPresent message=true
tr
tdnbsp;/td
tdnbsp;/td
tdnbsp;/td
/tr
html:messages id=message message=true
tr
tdnbsp;/td
td colspan=8 class=style1font color=redlibean:write
name=message//li/font/td
tdnbsp;/td
/tr
tr
tdnbsp;/td
td colspan=8 class=style1nbsp;/td
tdnbsp;/td
/tr
/html:messages
/logic:messagesPresent
In your web page.
Regards
David.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
Sent: 28 July 2005 13:35
To: user@struts.apache.org
Subject: Error Message Problem (Again)
Hi All,
Seems this is a constant area of irritation. I am attempting to put an
error message into the ActionMessage in an Action Class and redisplay
this
on a JSP page. This application is using tiles and is located in a
Struts
Module. Seems simple enough, but I can't get the darn thing to display
on
the JSP.
Here is extracts from my code.
Action Class:
public ActionForward execute(final ActionMapping mapping,
final
ActionForm form,
final
HttpServletRequest request,
final
HttpServletResponse response) {
ActionForward forward = mapping.findForward(fail);
ActionMessages errors = new ActionMessages();
...
errors.add(displayerror, new ActionMessage(myerror, false));
...
saveErrors(request, errors);
}
JSP Page:
html:form action=/myAction.do method=post
html:messages id=error property=displayerror
bean:write name=error /
/html:messages
...
/html:form
I have also tried:
html:form action=/myAction.do method=post
html:errors/
...
/html:form
Neither work
I inspect errors on the save error line and can see the error message in
the ActionMessage list, so I know it is being added to the request. I
also
used the false attribute to simply display myerror on the page to
eliminate the possibilty of a properties file typo problem. Any
recommendations on where to look?
This e mail is from DLA Piper Rudnick Gray Cary UK LLP.
The contents of this email and any attachments are confidential to the intended
recipient. They may not be disclosed to or used by or copied in any way by
anyone other than the intended recipient. If this email is received in error,
please contact DLA Piper Rudnick Gray Cary UK LLP on +44 (0) 8700 11
quoting the name of the sender and the email address to which it has been sent
and then delete it.
Please note that neither DLA Piper Rudnick Gray Cary UK LLP nor the sender
accept any responsibility for viruses and it is your responsibility to scan or
otherwise check this email and any attachments.
DLA Piper Rudnick Gray Cary UK LLP is a limited liability partnership
registered in England and Wales (registered number OC307847) which provides
services from offices in England, Belgium, Germany and the People's Republic of
China. A list of members is open for inspection at its registered office and
principal place of business 3 Noble Street, London EC2V 7EE. Partner denotes
member of a limited liability partnership.
DLA Piper Rudnick Gray Cary UK LLP is regulated by the Law Society and is a
member of DLA Piper Rudnick Gray Cary, a global legal services organisation,
the members of which are separate and distinct legal entities. For further
information, please refer to www.dlapiper.com.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Struts-EL problem with struts-el-1.3.0-dev
From: Marco Mistroni [EMAIL PROTECTED]
Yes I actually reverted to using c:if, and it worked
Just fine
I wanted just to report the situation to see if I was
Using the taglib incorrectly...
logic:present name=${sessionScope.results}
Even if you're going to use an expression, the 'name' attribute should still
evaluate to the _name_ of the bean, not the bean itself.
http://struts.apache.org/userGuide/struts-logic.html#present
If this works:
c:forEach var=entry items=${sessionScope.results}
// this displays the loop data
/c:forEach
Then the logic:present tag would be:
%@ taglib uri=http://struts.apache.org/tags-logic-el; prefix=logic %
logic:present name=results scope=session
No EL required for that one. But do use the JSTL tags instead whenever
possible. :)
[BTW, in a Servlet 2.3 container, you do not need all those taglib tags in
web.xml. If you use the correct URI, the tlds will be located in the .jar
files in WEB-INF/lib.]
--
Wendy Smoak
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
RE: Error Message Problem (Again)
No luck.
I also added test inside the logic:messagesPresent ... tag. The text was not displayed which indicates the save did not put any messages into the request, Yet, the ActionMessages does have a message in it and the saveMessage(request, errors) line is called.
Harland, David [EMAIL PROTECTED]
Harland, David [EMAIL PROTECTED]
07/28/2005 08:52 AM
Please respond to
Struts Users Mailing List user@struts.apache.org
To
Struts Users Mailing List user@struts.apache.org
cc
Subject
RE: Error Message Problem (Again)
Hi,
Try the following:
Use saveMessages(request, messages) instead of saveErrors(request,
errors)
And use something like
logic:messagesPresent message=true
tr
tdnbsp;/td
tdnbsp;/td
tdnbsp;/td
/tr
html:messages id=message message=true
tr
tdnbsp;/td
td colspan=8 class=style1font color=redlibean:write
name=message//li/font/td
tdnbsp;/td
/tr
tr
tdnbsp;/td
td colspan=8 class=style1nbsp;/td
tdnbsp;/td
/tr
/html:messages
/logic:messagesPresent
In your web page.
Regards
David.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]
Sent: 28 July 2005 13:35
To: user@struts.apache.org
Subject: Error Message Problem (Again)
Hi All,
Seems this is a constant area of irritation. I am attempting to put an
error message into the ActionMessage in an Action Class and redisplay
this
on a JSP page. This application is using tiles and is located in a
Struts
Module. Seems simple enough, but I can't get the darn thing to display
on
the JSP.
Here is extracts from my code.
Action Class:
public ActionForward execute(final ActionMapping mapping,
final
ActionForm form,
final
HttpServletRequest request,
final
HttpServletResponse response) {
ActionForward forward = mapping.findForward(fail);
ActionMessages errors = new ActionMessages();
...
errors.add(displayerror, new ActionMessage(myerror, false));
...
saveErrors(request, errors);
}
JSP Page:
html:form action="" method=post
html:messages id=error property=displayerror
bean:write name=error /
/html:messages
...
/html:form
I have also tried:
html:form action="" method=post
html:errors/
...
/html:form
Neither work
I inspect errors on the save error line and can see the error message in
the ActionMessage list, so I know it is being added to the request. I
also
used the false attribute to simply display myerror on the page to
eliminate the possibilty of a properties file typo problem. Any
recommendations on where to look?
This e mail is from DLA Piper Rudnick Gray Cary UK LLP.
The contents of this email and any attachments are confidential to the intended recipient. They may not be disclosed to or used by or copied in any way by anyone other than the intended recipient. If this email is received in error, please contact DLA Piper Rudnick Gray Cary UK LLP on +44 (0) 8700 11 quoting the name of the sender and the email address to which it has been sent and then delete it.
Please note that neither DLA Piper Rudnick Gray Cary UK LLP nor the sender accept any responsibility for viruses and it is your responsibility to scan or otherwise check this email and any attachments.
DLA Piper Rudnick Gray Cary UK LLP is a limited liability partnership registered in England and Wales (registered number OC307847) which provides services from offices in England, Belgium, Germany and the People's Republic of China. A list of members is open for inspection at its registered office and principal place of business 3 Noble Street, London EC2V 7EE. Partner denotes member of a limited liability partnership.
DLA Piper Rudnick Gray Cary UK LLP is regulated by the Law Society and is a member of DLA Piper Rudnick Gray Cary, a global legal services organisation, the members of which are separate and distinct legal entities. For further information, please refer to www.dlapiper.com.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
RE: Error Message Problem (Again)
Sorry, it does work, I forgot to change the messages name from errors.
Thank you for the help...
Harland, David [EMAIL PROTECTED]
Harland, David [EMAIL PROTECTED]
07/28/2005 08:52 AM
Please respond to
Struts Users Mailing List user@struts.apache.org
To
Struts Users Mailing List user@struts.apache.org
cc
Subject
RE: Error Message Problem (Again)
Hi,
Try the following:
Use saveMessages(request, messages) instead of saveErrors(request,
errors)
And use something like
logic:messagesPresent message=true
tr
tdnbsp;/td
tdnbsp;/td
tdnbsp;/td
/tr
html:messages id=message message=true
tr
tdnbsp;/td
td colspan=8 class=style1font color=redlibean:write
name=message//li/font/td
tdnbsp;/td
/tr
tr
tdnbsp;/td
td colspan=8 class=style1nbsp;/td
tdnbsp;/td
/tr
/html:messages
/logic:messagesPresent
In your web page.
Regards
David.
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]
Sent: 28 July 2005 13:35
To: user@struts.apache.org
Subject: Error Message Problem (Again)
Hi All,
Seems this is a constant area of irritation. I am attempting to put an
error message into the ActionMessage in an Action Class and redisplay
this
on a JSP page. This application is using tiles and is located in a
Struts
Module. Seems simple enough, but I can't get the darn thing to display
on
the JSP.
Here is extracts from my code.
Action Class:
public ActionForward execute(final ActionMapping mapping,
final
ActionForm form,
final
HttpServletRequest request,
final
HttpServletResponse response) {
ActionForward forward = mapping.findForward(fail);
ActionMessages errors = new ActionMessages();
...
errors.add(displayerror, new ActionMessage(myerror, false));
...
saveErrors(request, errors);
}
JSP Page:
html:form action="" method=post
html:messages id=error property=displayerror
bean:write name=error /
/html:messages
...
/html:form
I have also tried:
html:form action="" method=post
html:errors/
...
/html:form
Neither work
I inspect errors on the save error line and can see the error message in
the ActionMessage list, so I know it is being added to the request. I
also
used the false attribute to simply display myerror on the page to
eliminate the possibilty of a properties file typo problem. Any
recommendations on where to look?
This e mail is from DLA Piper Rudnick Gray Cary UK LLP.
The contents of this email and any attachments are confidential to the intended recipient. They may not be disclosed to or used by or copied in any way by anyone other than the intended recipient. If this email is received in error, please contact DLA Piper Rudnick Gray Cary UK LLP on +44 (0) 8700 11 quoting the name of the sender and the email address to which it has been sent and then delete it.
Please note that neither DLA Piper Rudnick Gray Cary UK LLP nor the sender accept any responsibility for viruses and it is your responsibility to scan or otherwise check this email and any attachments.
DLA Piper Rudnick Gray Cary UK LLP is a limited liability partnership registered in England and Wales (registered number OC307847) which provides services from offices in England, Belgium, Germany and the People's Republic of China. A list of members is open for inspection at its registered office and principal place of business 3 Noble Street, London EC2V 7EE. Partner denotes member of a limited liability partnership.
DLA Piper Rudnick Gray Cary UK LLP is regulated by the Law Society and is a member of DLA Piper Rudnick Gray Cary, a global legal services organisation, the members of which are separate and distinct legal entities. For further information, please refer to www.dlapiper.com.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
String concatenation
Hi, I would like to use Struts tags in JSP file in order to dynamically merge two beans. First is constant String, second is an Integer. The current, very dirty : -(, version is as follows (I present it in order to give you a better view of what I'm trying to do): bean:define id=dictionaryId name=dictionaryId scope=request / ... % String action = createDictionaryEntity.do?dictionary= + (Integer) dictionaryId; % html:form action=%= action % Probably it's quite easy and straightforward but I have been searching and haven't found an answer. Best regards, Konrad Billewicz - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: UrlValidator
I encountered this awhile back but didn't have time to investigate. As a workaround I use the localhost IP address 127.0.0.1 but it would be nice to get this fixed. -Original Message- From: Marc Logemann [mailto:[EMAIL PROTECTED] Sent: Thursday, July 28, 2005 8:33 AM To: user@struts.apache.org Subject: UrlValidator Hi, can it be that UrlValidator from Commons-Validator doesnt validate URLs with localhost in it? The following URL - http://localhost:8081/context/jsp/versand_1.jsp breaks in isValidAuthority() within UrlValidator. It seems he tries to check for toplevel domain length and sees my localhost as a toplevel domain. I am hoping i ve done something wrong because i would suspect that a UrlValidator validates this pretty straightforward URL. -- regards Marc Logemann [blog] http://www.logemann.org [busn] http://www.logentis.de - 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]
Re: String concatenation
Hello Konrad, I'm not sure I've understood correctly what you want to do, but here's what I would do : bean:define id=myBean value=%= SomeConstantsHolderClass.CONSTANT_STRING + SomeConstantsHolderClass.CONSTANT_STRING % / html:form action=%= action % ... Don't forget that you can still use an action _before_ you go to your JSP to prepopulate it. And by the way, in the JSP, everything is toString()ed, if I'm not mistaken, so you can always do something like bean:define id=aValue value=%= someString + someInteger %/ -- Stéphane Zuckerman - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Struts-EL problem with struts-el-1.3.0-dev
Hello Wendy,
I missed one point of your statement..
logic:present name=${sessionScope.results}
Even if you're going to use an expression, the 'name' attribute should
still
evaluate to the _name_ of the bean, not the bean itself.
http://struts.apache.org/userGuide/struts-logic.html#present
what do you mean? If I store the bean under the attribute name 'result',
why
the above EL is wrong?
Then the logic:present tag would be:
%@ taglib uri=http://struts.apache.org/tags-logic-el;
prefix=logic%
logic:present name=results scope=session
No EL required for that one. But do use the JSTL tags instead whenever
possible. :)
I use EL to replace logic:present name=%= Constants.RESULT%/
where Constants.RESULT has value 'results'
Where does the above declaration differ from the EL?
What does ${sessionScope.result} represent in this case?
can you show me a quick sample on a simple use of EL ?
if I store objects in session/request/applicationScope, why can't I use
struts-el with EL to retrieve them?
Thanx in advance for your time and regards
Marco mistroni
[BTW, in a Servlet 2.3 container, you do not need all those taglib
tags in
web.xml. If you use the correct URI, the tlds will be located in the
.jar
files in WEB-INF/lib.]
--
Wendy Smoak
-
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]
Re: String concatenation
hi, Konrad. do you need to put dictionaryId to querystring? i come up with html:form action=createDictionaryEntity input type=hidden name=dictionaryId value=bean:write name=dictionaryId/ as an alternative regards ichy
Re: Re: JSF is the beginning of the end of Struts !!!
I think I need clarify my statements before. Craig, I assure you that I always hope Java success and I do think it popular for web app. The only thing I think I said Java was not doing well as PHP is that PHP becomes more popular in public sites. That is definitely what you claimed. I claim that such a standard is irrelevant. I just try to find where I said Java, as a whole, is not popular in web application from my previous email. Unfortunately I can not find that statement. Can someone post my original saying so I can find it? I really like to know where I said that statement. [snip] If it uses Java in Apache, I may find mod_jk, or jsp, .do?... I found .shtml mostly. Consider a Struts based web site that maps *.html to the Struts action servlet (and there are more such sites on the public Internet than you would possibly believe :-). Your approach is going to come to the mistaken conclusion that this is a static HTML based site instead of a Struts based site. Actually, my favorite sites are the ones that map *.asp to the Struts servlet. Gotta love people that bend minds like that! But the serious point is that, if you think you can conclude anything based on the textual content of a URL, you are naively mistaken. Only the server you are calling knows for sure what technologies are used to produce a particular response. Did I say I can conclude based on textual content of a URL? Someone please find where I said that also. I really want see how I concluded that. What I say was that there is No Evidence that www.javaworld.com uses java in their site. If someone have evidence, please let me know. I really like to see how and why that site tried to hide java implementation. Am I so naive about thinking html is static site? Just go to our sites, see all the html there. If someone think those html are real html, they are wrong. Actually, my favorite sites are the ones that map *.asp to the Struts servlet. Gotta love people that bend minds like that! I cannot imagine what motivates people tried to map there servlets to .asp. Even I could easily map my site to .aspx, I would not do that. Although I respect MS, I am not really a huge fun of Microsoft. If I implement sites with Java, I want java take credits. If I implement sites in PHP, I want PHP take credits, I would not map that to .asp. If I use .net, I don't map it to jsp also. If I saw a site ends with .php or .aspx, I assume that was a php and .net site, and I belieave I am right on that with 99% confidence. John H. Xu http://www.usanalyst.com http://www.GetusJobs.com (The largest free job portal in North America) -- ___ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm
Re: probably a cleaner way... testing for just one user
Yes, I think what you caught yourself is what I was driving towards :)
Just for comparisons' sake, here's the filter and listener I came up with
to deal with this (this is condensed for space, just showing the main bits
of the puzzle)... this is what's going in JWP (unless someone points out a
major flaw of course!)...
The filter:
import java.io.IOException;
import javax.servlet.Filter;
import javax.servlet.FilterChain;
import javax.servlet.FilterConfig;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpSession;
import javax.servlet.ServletException;
import javax.servlet.ServletRequest;
import javax.servlet.ServletResponse;
public class SessionLimiterFilter implements Filter {
private static String forwardTo;
public void destroy() {
}
public void init(FilterConfig filterConfig) throws ServletException {
int maxSessions = Integer.parseInt(
filterConfig.getInitParameter(maxSessions));
SessionLimiterListener.setMaxSessions(maxSessions);
forwardTo = filterConfig.getInitParameter(forwardTo);
}
public void doFilter(ServletRequest request, ServletResponse response,
FilterChain filterChain)
throws ServletException, IOException {
filterChain.doFilter(request, response);
HttpSession session = ((HttpServletRequest)request).getSession();
String ds =
(String)session.getAttribute(SessionLimiter_DisallowSession);
if (ds != null) {
session.setAttribute(SessionLimiter_NoDecreaseCount, true);
session.invalidate();
request.getRequestDispatcher(forwardTo).forward(request, response);
}
}
}
The listener:
import javax.servlet.http.HttpSession;
import javax.servlet.http.HttpSessionEvent;
import javax.servlet.http.HttpSessionListener;
public class SessionLimiterListener implements HttpSessionListener {
private static Log log = LogFactory.getLog(SessionLimiterListener.class);
private static int maxSessions;
private static volatile int sessionCount;
public static void setMaxSessions(int inMaxSessions) {
maxSessions = inMaxSessions;
}
public synchronized void sessionCreated(HttpSessionEvent se) {
if (sessionCount == maxSessions) {
HttpSession session = se.getSession();
session.setAttribute(SessionLimiter_DisallowSession, true);
} else {
sessionCount++;
}
}
public synchronized void sessionDestroyed(HttpSessionEvent se) {
HttpSession session = se.getSession();
String ndc =
(String)session.getAttribute(SessionLimiter_NoDecreaseCount);
if (ndc == null) {
sessionCount--;
}
}
}
Our code looks pretty similar... that's to be expected based on the
solution path we took :) This code allows you to not really care how or
where the session got created, and it of course allows you to set any
number of sessions as the maximum.
Basically, the flow is this:
1. Filter fires, nothing happens initially, chain continues.
2. Listener fires when session is created. If max sessions is already in
use, set a flag in session and return. If max sessions NOT already in
use, bump up session count and return.
3. We wind up back in the filter, which checks for the flag that may have
been set in sessionCreated(). If it's found, we set a flag in session
indicating NOT to decrease the session count (because the session that was
just created doesn't really count towards reaching max sessions),
invalidate the session (which of course executes sessionDestroyed() in the
listener) and forward to the configured landing page. If the flag WAS NOT
found, filter is done.
4. Now, when a session is destroyed, we of course reach sessionDestroyed()
in the listener. In this case we check for that flag mentioned above that
could have been set by the filter. If it's NOT present we decrease the
session count because we know this was the result of logoff or timeout.
If the flag IS found, count is not touched.
In the end we came up with an almost identical solution really :) Great
minds and all that jazz right?!?
--
Frank W. Zammetti
Founder and Chief Software Architect
Omnytex Technologies
http://www.omnytex.com
On Thu, July 28, 2005 2:24 am, Tamas Szabo said:
Hi Frank,
Doesn't the sessionCount get decremented regardless of whether the
session is being destroyed as the result of a regular logoff or
timeout as well as if it was a rejected logon (i.e., max sessions
already in use)? I dealt with this problem in the code I'm adding to
Java Web Parts tomorrow, but maybe you found a better way than I did :)
Yes, the sessionCount is decremented everytime a session is destroyed
regardless of regular logoff, 'forced' logoff (max sessions in use) :-))
or timeout.
But that's what I wanted. Do you see a problem with it?
sessionCount is the number of existing sessions, so I decrement the
sessionCount everytime a session is destroyed(no matter why it was
destroyed).
Also, I was a bit surprised to learn that filters
Re: UrlValidator
Hi, if i have the time in the next days i will work on that, but where should i send a possible fix? struts-dev or to the commons-validator guys? Lindholm, Greg wrote: I encountered this awhile back but didn't have time to investigate. As a workaround I use the localhost IP address 127.0.0.1 but it would be nice to get this fixed. -Original Message- From: Marc Logemann [mailto:[EMAIL PROTECTED] Sent: Thursday, July 28, 2005 8:33 AM To: user@struts.apache.org Subject: UrlValidator Hi, can it be that UrlValidator from Commons-Validator doesnt validate URLs with localhost in it? The following URL - http://localhost:8081/context/jsp/versand_1.jsp -- regards Marc Logemann [blog] http://www.logemann.org [busn] http://www.logentis.de - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
multipart request issues
I am having issues to get the request parameters from a multipart request, le
me explain my problem first:
Our flow of process is:
(1) a multipart request is submitted to a struts action (let's say
updateTableAction.do) using POST
(2) the multipart request is processed by ActionServlet.process() method
(3) in our processPreprocess method, which overrides processPreprocess in
RequestProcessor.java from struts,
we need the parameters from this multipart request thus I am doing this:
CommonsMultipartRequestHandler commonMultiPartHandler = new
CommonsMultipartRequestHandler();
try {
commonMultiPartHandler.handleRequest(request);
}catch(Exception e){
// some handling
}
Hashtable ht = commonMultiPartHandler.getTextElements(); // thus I get the
parameters from the request
(4) however, all the fields/parameters in multipart request are no longer there
in the updateTableAction.do
Can you help me find out why this is happening? One theory is that a multipart
request can't be parsed twice. Is this correct, if it is correct, do you have
any alternative?
Thanks.
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Restricting Get requests
Part of the OWASP recommendations is that we do not allow authentication and session data to be submitted via GET request, this includes the session id. This comes from the OWASP top ten (http://www.owasp.org/documentation/topten/a3.html) under A3.5 under Browser Caching and Session ID Protection. I can tell jsp's to pass parameters in using a POST request, but if the user passes the parameters in through the url, I need to be able to restrict that. Does anyone know of any way to do this in struts? Also, does anyone know how to force struts to not pass along the session id if cookies are disabled? -- Josh Cronemeyer Information Network of Kansas I don't understand, said the scientist, why you lemmings all rush down to the sea and drown yourselves. How curious, said the lemming. The one thing I don't understand is why you human beings don't. From Interview With a Lemming, by James Thurber CONFIDENTIALITY NOTICE: This E-mail and any attachments are confidential. If you are not the intended recipient, you do not have permission to disclose, copy, distribute, or open any attachments. If you have received this E-mail in error, please notify us immediately by returning it to the sender and delete this copy from your system. Thank you. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Restricting Get requests
Are you asking if Struts can control what URL your user types into his browser? I could be wrong, but somehow, I don't think Struts is _that_ powerful. ;) -Dennis Josh Cronemeyer [EMAIL PROTECTED] 07/28/2005 10:48 AM Please respond to Struts Users Mailing List user@struts.apache.org To Struts Users Mailing List user@struts.apache.org cc Subject Restricting Get requests Part of the OWASP recommendations is that we do not allow authentication and session data to be submitted via GET request, this includes the session id. This comes from the OWASP top ten (http://www.owasp.org/documentation/topten/a3.html) under A3.5 under Browser Caching and Session ID Protection. I can tell jsp's to pass parameters in using a POST request, but if the user passes the parameters in through the url, I need to be able to restrict that. Does anyone know of any way to do this in struts? Also, does anyone know how to force struts to not pass along the session id if cookies are disabled? -- Josh Cronemeyer Information Network of Kansas I don't understand, said the scientist, why you lemmings all rush down to the sea and drown yourselves. How curious, said the lemming. The one thing I don't understand is why you human beings don't. From Interview With a Lemming, by James Thurber CONFIDENTIALITY NOTICE: This E-mail and any attachments are confidential. If you are not the intended recipient, you do not have permission to disclose, copy, distribute, or open any attachments. If you have received this E-mail in error, please notify us immediately by returning it to the sender and delete this copy from your system. Thank you. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Merging form properties
Hi, Is there a way to merge properties of two formbeans? I have two instances of the same form bean class, both containing some properties filled, the rest being null. I would like to copy the non-null properties from one form bean to the other. I tried using the PropertyUtils.copyProperties or BeanUtils.copyProperties but they copy all properties, whether they are null or not. Is there a way to tweak the copying, so that only non-null values are overwritten? TIA, Mihai -- The contents of this e-mail are intended for the named addressee only. It contains information that may be confidential. Unless you are the named addressee or an authorized designee, you may not copy or use it, or disclose it to anyone else. If you received it in error please notify us immediately and then destroy it. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: probably a cleaner way... testing for just one user
Tamas Szabo wrote the following on 7/28/2005 2:24 AM:
Yes, the sessionCount is decremented everytime a session is destroyed
regardless of regular logoff, 'forced' logoff (max sessions in use) :-))
or timeout.
But that's what I wanted. Do you see a problem with it?
sessionCount is the number of existing sessions, so I decrement the
sessionCount everytime a session is destroyed(no matter why it was
destroyed).
Thanks so much Tamas (and Frank, et al) for help with this. Tamas your
solution is what I ended up going with (started it before I saw Frank's
which looked similar). Here were some problems though...
1) For one, the business requirement created a bit more complication in
the filter. For example a check for the session timing out has to take
place on all pages except for (index.jsp/login/logout/appinuse). I'm
doing this check in the filter by checking for the session being null,
but it created complications of when to force a new one, which is needed
for the 'invalidate' check.
I'd love some help with the ugly logic I have (which I'll post at end of
this doc). I know it can be cleaned up, but you know how it is when your
trying a million different if/else things and your brain is frazzled:)
2) I noticed a lot of odd behavior when the server was shut down and the
pages were still up. For example, if the user was left on a page that
displayed a log off link and the server was restarted, when they then
clicked on log off (which calls session invalidate) it would decrement
the sessionCounter to -1. I'm not sure how sessionDestroyed could get
called before sessionCreated, but I guess it can? My hack here was to do
a check for sessionCount being less than 0 and if so reset it to 0.
3) Using [EMAIL PROTECTED] session=false % didn't seem to call the
sessionDestroyed method, which I really need to have happen when they
hit the logoff and appinuse.jsp. I went back to using %
session.invalidate(); % (I haven't looked at the docs on
session=false for JSPs does that just make sure a new Session isn't
created?)
OK here is the filter that I'd love some clean up on. There must be a
way to make this less ugly...you can try this link to see the full version:
http://rafb.net/paste/results/6APQEc67.html
or the relevant code is below but I'm sure it'll wrap ugly...
if ( pathNeedsCheckForSession(path) ) {
log.debug(pathNeedsCheckForSession = true);
if (session == null) {
log.debug(The session no longer exists, we are
creating new one and redirecting to index.jsp);
//create a new Session which will allow it to be tested
for too many
session = request.getSession(true);
response.sendRedirect(contextPath+/index.jsp);
return;
}
} else if (session == null ) {
//we are going to login/logout/or index page and no sesion
exists so we should create new one
session = request.getSession(true);
}
if ( path.indexOf(/appinuse.jsp) == -1
path.indexOf(/logout.jsp) == -1 ) {
log.debug(path not appinuser or logout.jsp so testing for
if too many users);
//if it's not the appinuse page or logout page, we need to
see if the app is in use
if ( session.getAttribute(Constants.INVALIDATE) != null ) {
log.debug(The application is in use);
response.sendRedirect(contextPath+/appinuse.jsp);
//make sure on appinuse page to also invalidate the
session!
return;
}
}
Thanks!
--
Rick
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Restricting Get requests
I think JSF and Shale are! LOL :) Sorry Craig, couldn't resist :) -- Frank W. Zammetti Founder and Chief Software Architect Omnytex Technologies http://www.omnytex.com On Thu, July 28, 2005 10:50 am, [EMAIL PROTECTED] said: Are you asking if Struts can control what URL your user types into his browser? I could be wrong, but somehow, I don't think Struts is _that_ powerful. ;) -Dennis Josh Cronemeyer [EMAIL PROTECTED] 07/28/2005 10:48 AM Please respond to Struts Users Mailing List user@struts.apache.org To Struts Users Mailing List user@struts.apache.org cc Subject Restricting Get requests Part of the OWASP recommendations is that we do not allow authentication and session data to be submitted via GET request, this includes the session id. This comes from the OWASP top ten (http://www.owasp.org/documentation/topten/a3.html) under A3.5 under Browser Caching and Session ID Protection. I can tell jsp's to pass parameters in using a POST request, but if the user passes the parameters in through the url, I need to be able to restrict that. Does anyone know of any way to do this in struts? Also, does anyone know how to force struts to not pass along the session id if cookies are disabled? -- Josh Cronemeyer Information Network of Kansas I don't understand, said the scientist, why you lemmings all rush down to the sea and drown yourselves. How curious, said the lemming. The one thing I don't understand is why you human beings don't. From Interview With a Lemming, by James Thurber CONFIDENTIALITY NOTICE: This E-mail and any attachments are confidential. If you are not the intended recipient, you do not have permission to disclose, copy, distribute, or open any attachments. If you have received this E-mail in error, please notify us immediately by returning it to the sender and delete this copy from your system. Thank you. - 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]
Re: Merging form properties
From: Cadariu, Mihai [EMAIL PROTECTED] I have two instances of the same form bean class, both containing some properties filled, the rest being null. I would like to copy the non-null properties from one form bean to the other. I tried using the PropertyUtils.copyProperties or BeanUtils.copyProperties but they copy all properties, whether they are null or not. Is there a way to tweak the copying, so that only non-null values are overwritten? I haven't done it, but you can use a BeanInfo class to control what properties are exposed: http://java.sun.com/docs/books/tutorial/javabeans/beaninfo/ -- Wendy Smoak - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Problem with Dyna Form
Kade Jeevan Kumar wrote: Hi All, My requirement is to generate a set of dynamic rows(with checkbox and combo box) at client side using JSP and Struts 1.1. As of now, i am generating the above stuff using java script. How can i use Dyna Form to meet the above requirement (or) suggest me other way if u have any solution for this. You can you use a dyna action form with a property like this: form-property name=blah type=java.lang.String[]/ - Dave - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Using XHTML and no form name tag
i'm havig problem with struts. i'm using 1.2.7 and i'm using the validator plug-in the problem is when i'm using html:javascript forname= the name attribiute for form form name= is not rendered when html:html xhtml=true however when only html:html is specified, without turning on the xhtml feature, the form name= tag is rendered. the question is, how to use the html:javascript feature for client side validation when using html:html xhtml=true - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: probably a cleaner way... testing for just one user
Hi again, Frank W. Zammetti wrote: Yes, I think what you caught yourself is what I was driving towards :) Just for comparisons' sake, here's the filter and listener I came up with to deal with this (this is condensed for space, just showing the main bits of the puzzle)... this is what's going in JWP (unless someone points out a major flaw of course!)... I wouldn't call itt a major flaw but there is a part which could cause some problems in your solution. You execute the filterChain.doFilter(request, response); in every case. This way you allow users to make requests which could have side effects on the server. I mean that I could execute a /updateUser.jsp?id=1firstName=TomlastName=Sawyer I could also execute POST requests. A second problem is that you can access all servlets that don't create HttpSessions in their doGet, doPost methods and jsp's with [EMAIL PROTECTED] session=false %. If you make it like I did it you won't have these problems. The two solutions will look more similar if you make this modification, but no worries, I won't sue you. ;-) (I would if I had the money for it, but I'm just a poor unemployed Java programmer hehe) Tamas - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
I after Wendy gave me the advice to rephrase the question I will try to rephrase it. Being new to struts, I want to know how people deal with data objects which need a scope bigger than request but smaller than session. I will give an example, you have some kind of wizard interface with myriads of input fields distributed over multiple forms. There are some ways to handle this a) You can dump anything into the session, which means you run into memory problems in the long run. b) you can put everything into hidden fields, which means you have myriads of hidden fields, which is slow and error prone to implement. c) You can try to add the stuff manually to the session and remove it later, which also is problematic in case of an error because then you have a mem leak caused, by a tangeling object in the session. Other frameworks have interesting approaches to solve this. MyFaces has x:saveState which basically serializes the data objects away and loads then once you reach the next page (and before the request data is assigned) Rails according to a friend has some kind of data structure which you can tell to extend to the next request, where you can dump data. And there is Shale with the dialog system, which allows you to specify scopes smaller than session but bigger than request. How is this problem resolved in struts in a proper manner? Werner - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: JSF is the beginning of the end of Struts !!!
I presume preserving URLs is the reason to map one technology's default
extention to another. For example, if you have a static web site with
files than end in .html, and switch to Java/Struts, then you can map
*.html to the Struts' action servlet.
The end users' bookmarks don't have to change, but you're really running
a Struts action under the covers.
Stéphane Zuckerman wrote:
John Henry Xu a écrit :
Actually, my favorite sites are the
ones that map *.asp to the Struts servlet. Gotta love people that
bend minds like that!
I cannot imagine what motivates people tried to map there servlets to
.asp. Even I could easily map my site to .aspx, I would not do that.
Although I respect MS, I am not really a huge fun of Microsoft. If I
implement sites with Java, I want java take credits. If I implement sites
in PHP, I want PHP take credits, I would not map that to .asp. If I use
.net, I don't map it to jsp also. If I saw a site ends with .php or
.aspx, I assume that was a php and .net site, and I belieave I am right
on that with 99% confidence.
I guess this is against wannabe crackers. Yes, yes, I know, security
through obscurity, blah blah blah... Well, once you've done everything
that isn't obfuscation-related, there is only obscurity that remains,
and frankly, it does help gain a few {minutes,hours} before the nasty
black hats break the server. :-)
--
Neil Erdwien, [EMAIL PROTECTED], Web Technologies Manager
Computing and Network Services, Kansas State University
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Restricting Get requests
At 9:48 AM -0500 7/28/05, Josh Cronemeyer wrote: Part of the OWASP recommendations is that we do not allow authentication and session data to be submitted via GET request, this includes the session id. This comes from the OWASP top ten (http://www.owasp.org/documentation/topten/a3.html) under A3.5 under Browser Caching and Session ID Protection. I can tell jsp's to pass parameters in using a POST request, but if the user passes the parameters in through the url, I need to be able to restrict that. Does anyone know of any way to do this in struts? Struts doesn't have programmatic support for this right now; I believe a Bugzilla enhancement was entered (or maybe just a discussion on the lists) to add HTTP method to the action mappings so that you could have different mappings for path + protocol. I haven't heard a lot of use cases for that, so I'd be hesitant to forge off on an implementation without a little more clarity about how people want to use it (it's not something I've ever needed.) In the meantime, I guess the best thing would be to use HttpServletRequest.getMethod() to test and evaluate whether to proceed or not. Also, does anyone know how to force struts to not pass along the session id if cookies are disabled? This is a function of the servlet container, not Struts; if you use JSP tags which rewrite URLs, they will append the session ID when the session ID is not being sent as a cookie-- the solution is to not use them! Of course, that leads to links which have the servlet context hard-coded in them, which is generally annoying. Why do you not want to pass along the session ID? Do you have an app which really doesn't use sessions? Joe -- Joe Germuska [EMAIL PROTECTED] http://blog.germuska.com Narrow minds are weapons made for mass destruction -The Ex - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Struts-EL problem with struts-el-1.3.0-dev
On 7/28/05, Marco Mistroni [EMAIL PROTECTED] wrote:
Hello Wendy,
I missed one point of your statement..
logic:present name=${sessionScope.results}
Even if you're going to use an expression, the 'name' attribute should
still
evaluate to the _name_ of the bean, not the bean itself.
http://struts.apache.org/userGuide/struts-logic.html#present
what do you mean? If I store the bean under the attribute name 'result',
why
the above EL is wrong?
What you want is:
logic:present name=results scope=session
This will check for a bean stored under the name 'results' in session scope.
${sessionScope.results} evaluates to the bean itself. The logic tag
receives *that bean* where before it received a String. I believe
that the logic tag would then convert that bean to a String and look
for an object stored under that name, instead.
Instead of dealing with all of this, just try the following JSTL:
c:if test=${not empty sessionScope.results}
-- Jeff
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: JSF is the beginning of the end of Struts !!!
[EMAIL PROTECTED] is the lead? Cool! On 7/25/05, John Public [EMAIL PROTECTED] wrote: After just finishing my JSF class, I can confidently confirm that JSF will eventually lead to Struts becoming OBSOLETE. Make no mistake about it. JSF rocks. JSF is based on true MVC architecture, not the MODEL 2 version that Struts uses. Struts had a good run, now it is time to build on the JSF foundation. JSF is the ONLY foundation that can compete witn ASP.NET 2.0 at this time. Let's all get behind JSF before MS takes over the web. Hopefully Tool Vendors will harness JSF potential and make Front-end web dev as easy as VS 2005 does. Time is not on our side. Let's get busy !!! J.J. JSF Developer (former Struts Lead Developer) Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- You can lead a horse to water but you cannot make it float on its back. ~Dakota Jack~ - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: UrlValidator (why not use java.net.URL?)
Hi, after looking into the code of UrlValidator and moreover looking into a similar class in Tapestry which is no longer active as it seems, i wonder why commons-validator UrlValidator is using Regex so much. Wouldnt it be enough to just use java.net.URL and let this class do the verification? Of course i am not quite sure how it deals non-schema URLs but for complete URLs to verifiy, its perhaps a better choice. Perhaps this one is for the developer list but i try it here first ;-) Lindholm, Greg wrote: I encountered this awhile back but didn't have time to investigate. As a workaround I use the localhost IP address 127.0.0.1 but it would be nice to get this fixed. -Original Message- From: Marc Logemann [mailto:[EMAIL PROTECTED] Sent: Thursday, July 28, 2005 8:33 AM To: user@struts.apache.org Subject: UrlValidator Hi, can it be that UrlValidator from Commons-Validator doesnt validate URLs with localhost in it? The following URL - http://localhost:8081/context/jsp/versand_1.jsp -- regards Marc Logemann [blog] http://www.logemann.org [busn] http://www.logentis.de - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: JSF is the beginning of the end of Struts !!!
Struts could live by improving in ways unlikely with the present management. The people in charge of Struts these days are really interested in something else. Unless the baton passes to those actively building something, Struts likely will either die from be non-competitive with like products or from being turned into something it is not, like JSF, Shale, Clay, Dirt. On 7/26/05, Greg Reddin [EMAIL PROTECTED] wrote: Interesting this discussion would come up now. I've been in and out of Struts for about 5 years. I'm now coming back in and taking interest in JSF/Shale, etc. A few of points to think about: 1. Struts will someday die. If it doesn't, then we've seen the end of technology advancement. If those of us who love Struts and what it provides want to continue to have influence going forward, we have to be ready to embrace the next thing that's better than Struts and add value there. Staying with something because of familiarity is typically the road to becoming a dinosaur in technology. 2. I only started studying JSF seriously about 2 weeks ago. From what I've seen it takes the best things Struts offers and brings them into a somewhat cleaner package. Those things are: * Nice HTML tag library. * Auto-populating of form beans. * Simple controller interface. It also expands on these things by making the view easier to develop, less restriction on how form beans and controllers are developed, and some other really interesting services that I've only begun to play with. To sum it up, using JSF out of the box I can do most of what I can with Struts, only quicker and with less code. I see JSF as the next step of Struts. But it's still not complete. I'm only beginning to see what Shale adds to JSF and am following Shale with great interest. 3. PHP. I've done some PHP over the last couple years. I'm by no means an expert, but I do know my way around somewhat. The whole time I've caught myself thinking if I was in Java I'd have a better way to do just about everything. But most of the better way is because of the J2EE APIs, Struts, and Tiles. The language is a bit more cumbersome than Java and it's much more difficult to modularize. I find PHP is easier (to me) if I don't try to make things elegant, but just spew code to some extent. I finally found a CMS to use as a Struts-like framework, but I still find Java/Struts and now JSF to be better, cleaner, and more modular. I am now trying to figure out a way to do my future on-the-side consulting in Java instead of PHP. I think another of PHPs success factors has to do with its acceptance from the open source community. Somehow Java has had a somewhat tainted reputation due to the lack of a truly open complete implementation -- and probably some are just bitter that Sun hasn't opened the whole thing up. I don't personally have anything to contribute to that debate, but I suspect it's a factor in why PHP has been so widely embraced on the web compared to Java. So all in all, it doesn't much matter to me if JSF supercedes Struts. If it's better, then it should. If we get behind Shale it think we can have our cake and eat it too. BTW, for JSF info, try MyFaces. Their Getting Started section points to some pretty good tutorials. In short testing I've found the myfaces implementation to be pretty stable. I'm actually surprised Craig is not listed as a contributor to that project. That's my 50 cents. Greg Daniel Perry wrote: Sorry for the OT postings. My point was that you cant compare usage of PHP with Struts. The number of Struts sites (or even java sites) will never overtake the number of PHP sites for the reasons i pointed out (although... zend are doing their best to kill off php by trying to move it into the enterprise arena and tying it with java). But for those same reasons, comparing the two is useless. As for JSF... It seems like a nice idea, but i havnt fully got my head round it. I cant see it killing off struts in the short term (just look at daily rate of postings on this list!!!) I do like the idea of replacing struts html tags with JSF. I certainly think JSTL/EL are nicer than the logic tags. I now have a dilema... I'm just about to start on another project. What technologies do i use? I guess i'll probably stick with struts. Though i may dabble with JSF a bit. Can anyone recommend any good resources? Sure a google search provides tons of information... but which is any good? Daniel. -Original Message- From: Mark Benussi [mailto:[EMAIL PROTECTED] Sent: 26 July 2005 10:38 To: 'Struts Users Mailing List' Subject: RE: Re: JSF is the beginning of the end of Struts !!! No offence but this is a daft conversation that I have seen too many times. Say Struts is dead to a certain Blue vendor who has just releases their latest Portal server which is built on Struts and they may
Re: Path like urls with objects' OIDs: /objects/object/17/198/4 possible?
Tomasz Nazar wrote:
But: I'd like to have something more prettier:
Say: /context/emplyee/17
Or even more: /context/archive/1998/employee/tomasz/tasks/7
Can you point me where to look for this? Is there any standard
option for that; some URL pattern rewriting or God knows what.
Do I have to write my own implementation of some method/class and
then parse the url, set proper parameters and pass over to Struts
(dispatcher?) ?
I suspect there are (substantially :) better ways to do what I did, but
here's one option:
- Use wildcard mappings to point to an action
- That action splits the URL based on whatever you need to accomplish (I
had a basic /admin/${table}[/${command}[/${identifier}]] structure)
- The action retrieves a previously-initialized hash of ${command} =
command-handler-classes
- The command-handler-classes extended a base class that Did the Right
Thing for simple database cases (generic CRUD) and povided hooks for
specialized form initialization stuff (like populating a drop-down list
for foreign keys, whatever).
- Somewhere in there I had a way to instantiate appropriate classes to
represent the form and DTOs (the only form bean I have in my struts
config looks like this:
form-beans
!-- Mysterious, huh? --
form-bean name=adminForm
type=org.apache.struts.validator.DynaValidatorActionForm/
/form-beans
Hrm... uh. Ah, here we go: I have a request processor that automagically
instantiates ActionForm subclasses based on the ${table} wildcard value.
I vaguely remember digging through source code and needing to have the
dummy entry (the action discussed above has the dummy form as its 'name'
attribute) but I don't recall why.
I was playing with Spring for IoC so ended up using that for a lot of
configuration; I could make fairly substantial changes in functionality
without having to touch the actual webapp code--neato.
Then they yelled at me for not doing billable work and I had to stop;
another couple of weeks and it would have been a poor duplicate of code
that already exists in the wild.
Sorry for the ramble--I'm motivated to work on that code again now, though!
Dave
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
RE: multipart request issues
You're theory is correct. The input stream is consumed as it is parsed.
The process flow doesn't seem to match what I'm seeing (Struts 1.2.4). In
1.2.4 all off this is handled in the request processor - not the servlet.
The very first thing the request processor does (in processMultipart) is to
parse the input stream and put all of the request paramaters in the wrapped
request object.
You can access these with a simple request.getParamater() in
processPreprocess in your extended request processor.
There is still a problem with multipart requests in that the parameters are
never available in JSPs, as they get the original (rather than the wrapped)
request. I overcome this by coppying the paramaters into a request scoped
map.
Hope this is of some use.
Paul
-Original Message-
From: Carl Smith [mailto:[EMAIL PROTECTED]
Sent: 28 July 2005 15:38
To: user@struts.apache.org
Subject: multipart request issues
I am having issues to get the request parameters from a
multipart request, le me explain my problem first:
Our flow of process is:
(1) a multipart request is submitted to a struts action
(let's say updateTableAction.do) using POST
(2) the multipart request is processed by
ActionServlet.process() method
(3) in our processPreprocess method, which overrides
processPreprocess in RequestProcessor.java from struts,
we need the parameters from this multipart request thus I am
doing this:
CommonsMultipartRequestHandler commonMultiPartHandler = new
CommonsMultipartRequestHandler();
try {
commonMultiPartHandler.handleRequest(request);
}catch(Exception e){
// some handling
}
Hashtable ht = commonMultiPartHandler.getTextElements(); //
thus I get the parameters from the request
(4) however, all the fields/parameters in multipart request
are no longer there in the updateTableAction.do
Can you help me find out why this is happening? One theory is
that a multipart request can't be parsed twice. Is this
correct, if it is correct, do you have any alternative?
Thanks.
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Axios Email Confidentiality Footer
Privileged/Confidential Information may be contained in this message. If you
are not the addressee indicated in this message (or responsible for delivery of
the message to such person), you may not copy or deliver this message to
anyone. In such case, you should destroy this message, and notify us
immediately. If you or your employer does not consent to Internet email
messages of this kind, please advise us immediately. Opinions, conclusions and
other information expressed in this message are not given or endorsed by my
Company or employer unless otherwise indicated by an authorised representative
independent of this message.
WARNING:
While Axios Systems Ltd takes steps to prevent computer viruses from being
transmitted via electronic mail attachments we cannot guarantee that
attachments do not contain computer virus code. You are therefore strongly
advised to undertake anti virus checks prior to accessing the attachment to
this electronic mail. Axios Systems Ltd grants no warranties regarding
performance use or quality of any attachment and undertakes no liability for
loss or damage howsoever caused.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: probably a cleaner way... testing for just one user
Hi Rick, It's past midnight here so I'll check the code tomorrow. A few questions though... 1) For one, the business requirement created a bit more complication in the filter. For example a check for the session timing out has to take place on all pages except for (index.jsp/login/logout/appinuse). I'm doing this check in the filter by checking for the session being null, but it created complications of when to force a new one, which is needed for the 'invalidate' check. I'd love some help with the ugly logic I have (which I'll post at end of this doc). I know it can be cleaned up, but you know how it is when your trying a million different if/else things and your brain is frazzled:) Could you clear this situation for me: User A is using the app, then his session expires. User B starts to use the app. User A comes back and want to access a page in the app. Remember that his session is expired and User B is already using the app. You redirect him to index.jsp? Or to appInUse.jsp? At first I would say that you could use a second filter which checks for expired session. This way the situation will be clear. One filter limits the sessions the other filter checks for timeouted sessions. 2) I noticed a lot of odd behavior when the server was shut down and the pages were still up. For example, if the user was left on a page that displayed a log off link and the server was restarted, when they then clicked on log off (which calls session invalidate) it would decrement the sessionCounter to -1. I'm not sure how sessionDestroyed could get called before sessionCreated, but I guess it can? My hack here was to do a check for sessionCount being less than 0 and if so reset it to 0. This is strange. What kind of webcontainer do you use? At first glance it seems like your web container has a 'feature' to serialize sessions on stop and to recreate them when you start the server again. In this case the sessionCounter would be set to 0 on restart but the session would be recreated = no sessionCreated() executed, just sessionDestroyed() on logout... 3) Using [EMAIL PROTECTED] session=false % didn't seem to call the sessionDestroyed method, which I really need to have happen when they hit the logoff and appinuse.jsp. I went back to using % session.invalidate(); % (I haven't looked at the docs on session=false for JSPs does that just make sure a new Session isn't created?) If you use [EMAIL PROTECTED] session=false % your session implicit object will not be initialized. If you use [EMAIL PROTECTED] session=true % in the translated servlet you will have a line: session = request.getSession(); So if you want to use session.invalidate() you cannot use [EMAIL PROTECTED] session=false %. I used [EMAIL PROTECTED] session=false % only in appInUse, not in logoff. I used it because otherwise a new session will be created when you redirect to the appinuse.jsp from the filter. The session will be flagged for invalidation, but because we just let through all the request to appInUse.jsp in the filter the sesiion will not be invalidated. OK here is the filter that I'd love some clean up on. There must be a way to make this less ugly...you can try this link to see the full version: I will check it tomorrow...sorry Tamas - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: probably a cleaner way... testing for just one user
Tamas Szabo wrote the following on 7/28/2005 12:51 PM: Could you clear this situation for me: User A is using the app, then his session expires. User B starts to use the app. User A comes back and want to access a page in the app. Remember that his session is expired and User B is already using the app. You redirect him to index.jsp? Or to appInUse.jsp? He should go to appInUse.jsp BUT if A's session has expired and no one else is using the app (B didn't log in), user A should go to index.jsp - this is a bit confusing since in order to determine if he can go to index.jsp we need to make him have a session - which is why path checking needs to take place to dictate what should/shouldn't be done. At first I would say that you could use a second filter which checks for expired session. This way the situation will be clear. One filter limits the sessions the other filter checks for timeouted sessions. That would probably be a good idea to use two filters vs doing it all in one. Although it could also be nice to see all that happens for each request by looking in just one filter vs two. 2) I noticed a lot of odd behavior when the server was shut down and the pages were still up. For example, if the user was left on a page that displayed a log off link and the server was restarted, when they then clicked on log off (which calls session invalidate) it would decrement the sessionCounter to -1. I'm not sure how sessionDestroyed could get called before sessionCreated, but I guess it can? My hack here was to do a check for sessionCount being less than 0 and if so reset it to 0. This is strange. What kind of webcontainer do you use? At first glance it seems like your web container has a 'feature' to serialize sessions on stop and to recreate them when you start the server again. In this case the sessionCounter would be set to 0 on restart but the session would be recreated = no sessionCreated() executed, just sessionDestroyed() on logout... Possibly that is what is happening. The scenario where I noticed this was using JBoss (Tomcat as the app server in JBoss). OK here is the filter that I'd love some clean up on. There must be a way to make this less ugly...you can try this link to see the full version: I will check it tomorrow...sorry No problem. I appreciate all your help so far. -- Rick - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
Michael Jouravlev wrote: There are no standard ways to do this. Creating wizards myself, I was thinking about some kind of garbage collector for session object. This project: http://wfnm.sourceforge.net/ seems to have this feature. It has some other features so I don't know can you pull out the session garbage collection feature only or not. I guess not, because it detects garbage in relation to the flow. Thus seems that you have to define the flow, which is unfortunate. Anyway, I did not really use this project, so you may try and report results ;-) just checked it out, looks exactly like what I need... I will give it a try, once I reach the point where I need it (which will come in a few days) thanks for pointing me towards it. Werner - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: probably a cleaner way... testing for just one user
Hello,
I cannon run an application
The error that I get when i asm going to run the
apllication is
type Status report
message Servlet action is currently unavailable
description The requested service (Servlet action is
currently unavailable) is not currently available.
It doesn{to procees my transaciton.
The web.xml file is:
?xml version=1.0 encoding=ISO-8859-1?
!DOCTYPE web-app
PUBLIC -//Sun Microsystems, Inc.//DTD Web
Application 2.2//EN
http://java.sun.com/j2ee/dtds/web-app_2_2.dtd;
web-app
!-- Action Servlet Configuration --
servlet
servlet-nameaction/servlet-name
servlet-classorg.apache.struts.action.ActionServlet/servlet-class
init-param
param-nameconfig/param-name
param-value/WEB-INF/struts-config.xml/param-value
/init-param
load-on-startup1/load-on-startup
/servlet
!-- Action Servlet Mapping --
servlet-mapping
servlet-nameaction/servlet-name
url-pattern*.do/url-pattern
/servlet-mapping
!-- The Usual Welcome File List --
welcome-file-list
welcome-fileindex.jsp/welcome-file
/welcome-file-list
!-- Struts Tag Library Descriptors --
taglib
taglib-uri/tags/struts-bean/taglib-uri
taglib-location/WEB-INF/struts-bean.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-html/taglib-uri
taglib-location/WEB-INF/struts-html.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-logic/taglib-uri
taglib-location/WEB-INF/struts-logic.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-nested/taglib-uri
taglib-location/WEB-INF/struts-nested.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-template/taglib-uri
taglib-location/WEB-INF/struts-template.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-tiles/taglib-uri
taglib-location/WEB-INF/struts-tiles.tld/taglib-location
/taglib
/web-app
and the struts-config.xml is:
?xml version=1.0 encoding=ISO-8859-1 ?
!DOCTYPE struts-config PUBLIC
-//Apache Software Foundation//DTD Struts
Configuration 1.1//EN
http://jakarta.apache.org/struts/dtds/struts-config_1_1.dtd;
struts-config
!-- Data Source
Configuration --
data-sources
data-source
set-property
property=autoCommit
value=true/
set-property
property=description
value=Configuración de la Base de Datos del
Sistema/
set-property
property=driverClassName
value=org.postgresql.Driver/
set-property
property=maxCount
value=100/
set-property
property=minCount
value=2/
set-property
property=url
value=jdbc:postgresql://159.90.16.8/oest4/
set-property
property=username
value=postgres/
set-property
property=password
value=postgres/
/data-source
/data-sources
!-- Form
Bean Definitions --
form-beans
form-bean
name=validacionForm
type=dide.bienestar.usuario.form.ValidacionForm /
form-bean
name=usuarioForm
type=dide.bienestar.usuario.form.UsuarioForm /
form-bean
name=presolicitudForm
type=dide.bienestar.solicitud.form.PresolicitudForm
/
form-bean
name=solicitudForm
type=dide.bienestar.solicitud.form.SolicitudForm /
form-bean
name=familiarForm
type=dide.bienestar.solicitud.form.FamiliarForm /
form-bean
name=exoneracionForm
type=dide.bienestar.solicitud.form.ExoneracionForm
/
form-bean
name=exoneracion1Form
type=dide.bienestar.solicitud.form.Exoneracion1Form
/
form-bean
name=estudianteForm
type=dide.bienestar.estudiante.form.EstudianteForm
/
form-bean
name=observacionesForm
type=dide.bienestar.solicitud.form.ObservacionesForm
/
form-bean
name=analisisForm
type=dide.bienestar.solicitud.form.AnalisisForm /
form-bean
name=exoneracionesForm
type=dide.bienestar.beneficios.form.ExoneracionesForm
/
form-bean
name=anularForm
type=dide.bienestar.beneficios.form.AnularForm /
form-bean
name=cestaForm
type=dide.bienestar.metodologia.form.CestaForm /
form-bean
name=ingresosForm
type=dide.bienestar.metodologia.form.IngresosForm /
form-bean
name=claseForm
type=dide.bienestar.metodologia.form.ClaseForm /
/form-beans
!-- === Global
Forward Definitions --
global-forwards
forward
Please i caanot run this
--- Tamas Szabo [EMAIL PROTECTED] wrote: Hi Rick, It's past midnight here so I'll check the code tomorrow. A few questions though... 1) For one, the business requirement created a bit more complication in the filter. For example a check for the session timing out has to take place on all pages except for (index.jsp/login/logout/appinuse). I'm doing this check in the filter by checking for the session being null, but it created complications of when to force a new one, which is needed for the 'invalidate' check. I'd love some help with the ugly logic I have (which I'll post at end of this doc). I know it can be cleaned up, but you know how it is when your trying a million different if/else things and your brain is frazzled:) Could you clear this situation for me: User A is using the app, then his session expires. User B starts to use the app. User A comes back and want to access a page in the app. Remember that his session is expired and User B is already using the app. You redirect him to index.jsp? Or to appInUse.jsp? At first I would say that you could use a second filter which checks for expired session. This way the situation will be clear. One filter limits the sessions the other filter checks for timeouted sessions. 2) I noticed a lot of odd behavior when the server was shut down and the pages were still up. For example, if the user was left on a page that displayed a log off link and the server was restarted, when they then clicked on log off (which calls session invalidate) it would decrement the sessionCounter to -1. I'm not sure how sessionDestroyed could get called before sessionCreated, but I guess it can? My hack here was to do a check for sessionCount being less than 0 and if so reset it to 0. This is strange. What kind of webcontainer do you use? At first glance it seems like your web container has a 'feature' to serialize sessions on stop and to recreate them when you start the server again. In this case the sessionCounter would be set to 0 on restart but the session would be recreated = no sessionCreated() executed, just sessionDestroyed() on logout... 3) Using [EMAIL PROTECTED] session=false % didn't seem to call the sessionDestroyed method, which I really need to have happen when they hit the logoff and appinuse.jsp. I went back to using % session.invalidate(); % (I haven't looked at the docs on session=false for JSPs does that just make sure a new Session isn't created?) If you use [EMAIL PROTECTED] session=false % your session implicit object will not be initialized. If you use [EMAIL PROTECTED] session=true % in the translated servlet you will have a line: session = request.getSession(); So if you want to use session.invalidate() you cannot use [EMAIL PROTECTED] session=false %. I used [EMAIL PROTECTED] session=false % only in appInUse, not in logoff. I used it because otherwise a new session will be created when you redirect to the appinuse.jsp from the filter. The session will be flagged for invalidation, but because we just let through all the request to appInUse.jsp in the filter the sesiion will not be invalidated. OK here is the filter that I'd love some clean up on. There must be a way to make this less ugly...you can try this link to see the full version: I will check it tomorrow...sorry Tamas - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: String concatenation
I'm not sure I've understood correctly what you want to do, but here's what I would do : bean:define id=myBean value=%= SomeConstantsHolderClass.CONSTANT_STRING + SomeConstantsHolderClass.CONSTANT_STRING % / html:form action=%= action % ... It's not exactly what I expected, but it's better solution than mine. Ideally I thought about complete aviodance of % and % tags. I wish to use only Struts tags in order to concatenate two strings and put them as an action property for html:form tag. Is it possible? I thought that using % and % in Struts-driven applications is rather bad idea for common things. Such form action with parameter is probably common thing... Was I right? Best regards, Konrad Billewicz - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: JSF is the beginning of the end of Struts !!!
I mapped to .html but not .asp, and I doubt many people would map java
apps to .php, .asp or .aspx in hosting companies (Of course u can
technically but I can not see the benefit). That's why I assume if I see
.php, I know it is php site with 99% confidence.
The reason for our mapping to html was that this page is rather stable
and I don't change their contents over time (in general). And for
security reason as Stéphane said. For PHP and perl it is very simple to
do that. For java it is a little more difficult but manageable. It can
give you more URL addresses as Neil said.
As for jsf, is it .jspx? or maybe mapping to others? It is possible jsf
maps to html. jsf using .aspx? It is unthinkable even technically
possible.
John H. Xu
- Original Message -
From: Neil Erdwien
To: Struts Users Mailing List
Subject: Re: JSF is the beginning of the end of Struts !!!
Date: Thu, 28 Jul 2005 11:01:29 -0500
I presume preserving URLs is the reason to map one technology's
default extention to another. For example, if you have a static
web site with files than end in .html, and switch to Java/Struts,
then you can map *.html to the Struts' action servlet.
The end users' bookmarks don't have to change, but you're really
running a Struts action under the covers.
Stéphane Zuckerman wrote:
John Henry Xu a écrit :
Actually, my favorite sites are the
ones that map *.asp to the Struts servlet. Gotta love people
that
bend minds like that!
I cannot imagine what motivates people tried to map there
servlets to
.asp. Even I could easily map my site to .aspx, I would not do
that.
Although I respect MS, I am not really a huge fun of Microsoft.
If I
implement sites with Java, I want java take credits. If I
implement sites
in PHP, I want PHP take credits, I would not map that to .asp.
If I use
.net, I don't map it to jsp also. If I saw a site ends with .php
or
.aspx, I assume that was a php and .net site, and I belieave I
am right
on that with 99% confidence.
I guess this is against wannabe crackers. Yes, yes, I know,
security
through obscurity, blah blah blah... Well, once you've done
everything
that isn't obfuscation-related, there is only obscurity that
remains,
and frankly, it does help gain a few {minutes,hours} before the
nasty
black hats break the server. :-)
-- Neil Erdwien, [EMAIL PROTECTED], Web Technologies Manager
Computing and Network Services, Kansas State University
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
John H. Xu
http://www.usanalyst.com
http://www.GetusJobs.com (The largest free job portal in North America)
--
___
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm
Re: String concatenation
do you need to put dictionaryId to querystring? i come up with html:form action=createDictionaryEntity input type=hidden name=dictionaryId value=bean:write name=dictionaryId/ as an alternative It is a good idea. But... shouldn't I avoid using non-Struts HTML form tags in Struts-driven applications? I am quite new in Struts and I don't want to get bad habits. :-) Best regards, Konrad Billewicz - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
Michael Jouravlev wrote: On 7/28/05, Werner Punz [EMAIL PROTECTED] wrote: I after Wendy gave me the advice to rephrase the question I will try to rephrase it. Being new to struts, I want to know how people deal with data objects which need a scope bigger than request but smaller than session. There are no standard ways to do this. Creating wizards myself, I was thinking about some kind of garbage collector for session object. This project: http://wfnm.sourceforge.net/ seems to have this feature. It has some other features so I don't know can you pull out the session garbage collection feature only or not. I guess not, because it detects garbage in relation to the flow. Thus seems that you have to define the flow, which is unfortunate. Anyway, I did not really use this project, so you may try and report results ;-) Actually I was thinking also along the lines of having some kind of garbage collecting mechanism. I did something like that in JSF very easy (reprogrammed parts of the x:saveState) the trick was to push the object into the session with some kind of key which stores the current and the old session and at the end of each rendering cycle (which is the problem because struts behaves totally different) I basically dumped the objects from the old url and pushed the newly registrated ones into the central object storage. That trick worked very well, because you have clear distinctive phases in JSF which mark the before assigning the data part and the after rendering part of the request and you already have a save and restore mechanism on component level which utilizes those two phases. I tried to move over a similar mechanism in Struts, but basically failed due to time constraints, which I cannot resolve by technical means ;-) - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Gonzalez's question posted as new thread:
Aleksander Gonzalez,
It's a very bad idea to hi-jack a thread by replying to it and changing
the content. (ref: probably a cleaner way..testing for one user). Here
is the question that you asked in the hijacked thread that I'm posting
new. But as far as an answer goes to your question, you need to look in
your server logs for a hint as to what the problem is.
Aleksnader's post:
Hello,
I cannon run an application
The error that I get when i asm going to run the
apllication is
type Status report
message Servlet action is currently unavailable
description The requested service (Servlet action is
currently unavailable) is not currently available.
It doesn{to procees my transaciton.
The web.xml file is:
?xml version=1.0 encoding=ISO-8859-1?
!DOCTYPE web-app
PUBLIC -//Sun Microsystems, Inc.//DTD Web
Application 2.2//EN
http://java.sun.com/j2ee/dtds/web-app_2_2.dtd;
web-app
!-- Action Servlet Configuration --
servlet
servlet-nameaction/servlet-name
servlet-classorg.apache.struts.action.ActionServlet/servlet-class
init-param
param-nameconfig/param-name
param-value/WEB-INF/struts-config.xml/param-value
/init-param
load-on-startup1/load-on-startup
/servlet
!-- Action Servlet Mapping --
servlet-mapping
servlet-nameaction/servlet-name
url-pattern*.do/url-pattern
/servlet-mapping
!-- The Usual Welcome File List --
welcome-file-list
welcome-fileindex.jsp/welcome-file
/welcome-file-list
!-- Struts Tag Library Descriptors --
taglib
taglib-uri/tags/struts-bean/taglib-uri
taglib-location/WEB-INF/struts-bean.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-html/taglib-uri
taglib-location/WEB-INF/struts-html.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-logic/taglib-uri
taglib-location/WEB-INF/struts-logic.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-nested/taglib-uri
taglib-location/WEB-INF/struts-nested.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-template/taglib-uri
taglib-location/WEB-INF/struts-template.tld/taglib-location
/taglib
taglib
taglib-uri/tags/struts-tiles/taglib-uri
taglib-location/WEB-INF/struts-tiles.tld/taglib-location
/taglib
/web-app
and the struts-config.xml is:
?xml version=1.0 encoding=ISO-8859-1 ?
!DOCTYPE struts-config PUBLIC
-//Apache Software Foundation//DTD Struts
Configuration 1.1//EN
http://jakarta.apache.org/struts/dtds/struts-config_1_1.dtd;
struts-config
!-- Data Source
Configuration --
data-sources
data-source
set-property
property=autoCommit
value=true/
set-property
property=description
value=Configuración de la Base de Datos del
Sistema/
set-property
property=driverClassName
value=org.postgresql.Driver/
set-property
property=maxCount
value=100/
set-property
property=minCount
value=2/
set-property
property=url
value=jdbc:postgresql://159.90.16.8/oest4/
set-property
property=username
value=postgres/
set-property
property=password
value=postgres/
/data-source
/data-sources
!-- Form
Bean Definitions --
form-beans
form-bean
name=validacionForm
type=dide.bienestar.usuario.form.ValidacionForm /
form-bean
name=usuarioForm
type=dide.bienestar.usuario.form.UsuarioForm /
form-bean
name=presolicitudForm
type=dide.bienestar.solicitud.form.PresolicitudForm
/
form-bean
name=solicitudForm
type=dide.bienestar.solicitud.form.SolicitudForm /
form-bean
name=familiarForm
type=dide.bienestar.solicitud.form.FamiliarForm /
form-bean
name=exoneracionForm
type=dide.bienestar.solicitud.form.ExoneracionForm
/
form-bean
name=exoneracion1Form
type=dide.bienestar.solicitud.form.Exoneracion1Form
/
form-bean
name=estudianteForm
type=dide.bienestar.estudiante.form.EstudianteForm
/
form-bean
name=observacionesForm
type=dide.bienestar.solicitud.form.ObservacionesForm
/
form-bean
name=analisisForm
type=dide.bienestar.solicitud.form.AnalisisForm /
form-bean
name=exoneracionesForm
type=dide.bienestar.beneficios.form.ExoneracionesForm
/
form-bean
name=anularForm
type=dide.bienestar.beneficios.form.AnularForm /
form-bean
name=cestaForm
type=dide.bienestar.metodologia.form.CestaForm /
form-bean
name=ingresosForm
RE: [OT] Hibernate vs. iBatis vs. POJO
I've work for the USPS for 3 years and now work for the DOD. At the PO we moved from pojos to iBatis, although there was another team trying to get Hibernate going. Now with the DOD, we use EJBs, although I and a colleague have tried to persuade our architech to use ANYTHING other than EJBs. I don't think this fully answers your question but it's a start. Mariano Hernandez -Original Message- From: Access Denied [mailto:[EMAIL PROTECTED] Sent: Friday, July 22, 2005 8:50 AM To: Struts Users Mailing List; Ed Griebel Subject: Re: [OT] Hibernate vs. iBatis vs. POJO I bought James Elliott's Hibernate (O'Reilly Developer Notebook Series 2004), but a search for iBATIS on amazon returns books like Pocket Guide to the Birds of Britain and North-West Europe and An Introduction To The Mystical Use of Classical Persian Poems. Is there any timely definitive literature available for iBATIS? In this realm, at least, it looks like Hibernate rules. I have also recently noticed that any job opportunities specifying a java persistence technology requires Hibernate; I have yet to run across any requirements for iBATIS. Does Hibernate dominate in business and government, too? It has also been my general experience that clients demand impossible deadlines, mainly, I think, because most do not have a clue when it comes to SDLC and due diligence. I've worked at many companies and agencies over the years and changing jobs has not changed the lack of appreciation for the SDLC outside of IT. The comment that contractors are assumed to already know the technology is true; the facts of tech life is anathema to this assumption. For a personal example, I was using java 1.2 and struts 1.0 on my last java job, but now with java 5.0, struts 1.2 and spring, the landscape has changed dramatically. We have no choice but to learn new stuff by the seat of our pants. ~buddy On 7/22/05, Ed Griebel [EMAIL PROTECTED] wrote: All too often that's how it works. If you say no, there are people that will say yes. Even if they end up taking as much time as you said it would, they've got the project, not you. Sometimes the short time is understandable, possibly due to uncontrollable situations, but most of the time it's due to poor (or too much!) planning. Many times the situation described below is a start to negotiations on budget, scope, resources, etc., but not always :-) -ed On 7/22/05, netsql [EMAIL PROTECTED] wrote: Frank W. Zammetti wrote: It has not been unusual to estimate three months for something, and that's fairly realistic to do it right, and the business says nope, 1.5 months is when we need it. I need you to paint the house, but I only have budget to wash my hair? That is no respect and a bit like slavery! .V - 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] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Filtering bean properties
I have a report with a huge number of columns (about 150). In the prior form user can choose some of them. My current task is to show only choosen columns in the report. I'm unexpectedly uncertain how to do it elegantly (using, of course, Struts). The current version of this huge result page (showing all colums) is similar to this: table tr tdbAgreement type/b/td tdbAgreement symbol/b/td !-- many, many (about 150) other columns) -- /tr logic:iterate name=treaties id=treaty scope=request tr td bean:write name=treaty property=agreementType/ /td td bean:write name=treaty property=agreementSymbol/ /td !-- many, many (150 too) values, mostly String and Date objects -- /tr In my opinion reports are quite common, so it should be a quite common problem. How would you handle it? Best regards, Konrad Billewicz - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: String concatenation
Konrad Billewicz wrote: bean:define id=myBean value=%= SomeConstantsHolderClass.CONSTANT_STRING + SomeConstantsHolderClass.CONSTANT_STRING % / html:form action=%= action % It's not exactly what I expected, but it's better solution than mine. Ideally I thought about complete aviodance of % and % tags. I wish to use only Struts tags in order to concatenate two strings and put them as an action property for html:form tag. Is it possible? I thought that using % and % in Struts-driven applications is rather bad idea for common things. Such form action with parameter is probably common thing... I've never put a parameter in a form's action attribute--why would you want to do that? Wouldn't it be much cleaner to use a hidden form field with the appropriate parameter name? So your original example would be (minus whatever silly syntax mistakes because I'm groggy): html:form action=createDictionaryEntity html:hidden property=dictionary/ !-- ... -- /html:form and whatever loads the form puts the appropriate dictionaryId value into the ActionForm? Dave
[OT] Re: bean traversal question
Werner Punz wrote: I tried to move over a similar mechanism in Struts, but basically failed due to time constraints, which I cannot resolve by technical means ;-) If you were Really Good you could. Probably not a pure software solution, though, especially if any space-time distortion is involved. Dave - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Filtering bean properties
You could use JSTL if you'd like to check for whether you should print
the column out - something like:
c:if test=${myParameter==true}tdprint my stuff here/td
/c:if
Im not sure how to do it with just struts, someone else might know that.
-Original Message-
From: news [mailto:[EMAIL PROTECTED] On Behalf Of Konrad Billewicz
Sent: Thursday, July 28, 2005 2:47 PM
To: user@struts.apache.org
Subject: Filtering bean properties
I have a report with a huge number of columns (about 150). In the prior
form
user can choose some of them. My current task is to show only choosen
columns in
the report. I'm unexpectedly uncertain how to do it elegantly (using, of
course,
Struts). The current version of this huge result page (showing all
colums) is
similar to this:
table
tr
tdbAgreement type/b/td
tdbAgreement symbol/b/td
!-- many, many (about 150) other columns) --
/tr
logic:iterate name=treaties id=treaty scope=request
tr
td
bean:write name=treaty property=agreementType/
/td
td
bean:write name=treaty property=agreementSymbol/
/td
!-- many, many (150 too) values, mostly String and Date objects --
/tr
In my opinion reports are quite common, so it should be a quite common
problem.
How would you handle it?
Best regards,
Konrad Billewicz
-
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]
Re: bean traversal question
On 7/28/05, Werner Punz [EMAIL PROTECTED] wrote: Actually I was thinking also along the lines of having some kind of garbage collecting mechanism. I did something like that in JSF very easy (reprogrammed parts of the x:saveState) the trick was to push the object into the session with some kind of key which stores the current and the old session and at the end of each rendering cycle (which is the problem because struts behaves totally different) I basically dumped the objects from the old url and pushed the newly registrated ones into the central object storage. That trick worked very well, because you have clear distinctive phases in JSF which mark the before assigning the data part and the after rendering part of the request and you already have a save and restore mechanism on component level which utilizes those two phases. You can do the same in Struts. Well, in Struts Dialogs ;-) which like JSF also has distinctive phases and stores state on server. Actually, I am going to implement this feature. The idea is that an action is a web resource. An action can have multiple states and can reder different pages. The fact that action mapping can have only one corresponding action form, works pretty well. I stick all data that I use in a resource, into action form and set form scope to session. Then I can have a wizard (controlled by a single action), or a login/logout component. Once I leave that action, I can simply remove the action form from the session. I will have to define subsession groups, and each action mapping would be able to participate in one or more subsessions. Once I leave a subsession, I can remove an action form with all its corresponding data. On a side note, I cannot get why storing data in an action form is considered wrong, while directly accessing session is OK. I think quite opposite. I prefer to stick data related to an action, into action form. It is more controlled, more visible, and then I can simply remove one action form object from the session to clear the whole tree. Michael. -- Struts Dialogs http://struts.sourceforge.net/strutsdialogs - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: String concatenation
Dave Newton newton at pingsite.com writes: I've never put a parameter in a form's action attribute--why would you want to do that? Wouldn't it be much cleaner to use a hidden form field with the appropriate parameter name? So your original example would be (minus whatever silly syntax mistakes because I'm groggy): html:form action=createDictionaryEntity html:hidden property=dictionary/ !-- ... -- /html:form and whatever loads the form puts the appropriate dictionaryId value into the ActionForm? I wish to pass the dictionary variable using GET - make it visible in the URL. Secondly, this action (createDictionaryEntity) can be accessed both from form and from link (using a href=...). I wish to have one, clear method of passing parameters to it. Of course, I can at first check the request and secondly the form (or inversely) but it is not, in my opinion, elegant solution. I thing it is not compulsory. Best regards, Konrad Billewicz - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
Michael Jouravlev wrote: On a side note, I cannot get why storing data in an action form is considered wrong, while directly accessing session is OK. I think Well it is quite the opposite in JSF... using the session is sort of a secure source of getting heavy beating (which is half right but too extreme), while the so called backend beans are combined controller/nmodel objects which basically are kept over all phases of the rendering cycle. The no use of session if possible results in such things like x:saveState which serializes everything to pass data over the cycle boundaries without having to pass parameters. An interesting approach if you ask me, but puts too much emphasis on the serialisation mechanism, a mixed form would be better, where you can chose one way or the other. quite opposite. I prefer to stick data related to an action, into action form. It is more controlled, more visible, and then I can simply remove one action form object from the session to clear the whole tree. Well a combind form, controller approach is closer to classical UIs than the way struts does it. Although I can see the logic behind the approach of struts in a http context, it is sort of confusing. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Filtering bean properties
This is the easiest and cleanest way to do this.
If not you could do it with JavaScript, but then again, not the best way!
__
Senior Programmer Analyst, Tax Distributed Systems Development
Tax Compliance Development, ADP IT
Phone: (909) 592-6411 Ext. 3863
e-mail: [EMAIL PROTECTED]
-Original Message-
From: Johnson, Kaerstin [mailto:[EMAIL PROTECTED]
Sent: Thursday, July 28, 2005 12:03 PM
To: Struts Users Mailing List
Subject: RE: Filtering bean properties
You could use JSTL if you'd like to check for whether you should print
the column out - something like:
c:if test=${myParameter==true}tdprint my stuff here/td
/c:if
Im not sure how to do it with just struts, someone else might know that.
-Original Message-
From: news [mailto:[EMAIL PROTECTED] On Behalf Of Konrad Billewicz
Sent: Thursday, July 28, 2005 2:47 PM
To: user@struts.apache.org
Subject: Filtering bean properties
I have a report with a huge number of columns (about 150). In the prior
form
user can choose some of them. My current task is to show only choosen
columns in
the report. I'm unexpectedly uncertain how to do it elegantly (using, of
course,
Struts). The current version of this huge result page (showing all
colums) is
similar to this:
table
tr
tdbAgreement type/b/td
tdbAgreement symbol/b/td
!-- many, many (about 150) other columns) --
/tr
logic:iterate name=treaties id=treaty scope=request
tr
td
bean:write name=treaty property=agreementType/
/td
td
bean:write name=treaty property=agreementSymbol/
/td
!-- many, many (150 too) values, mostly String and Date objects --
/tr
In my opinion reports are quite common, so it should be a quite common
problem.
How would you handle it?
Best regards,
Konrad Billewicz
-
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]
-
This message and any attachments are intended only for the use of the
addressee and may contain information that is privileged and confidential.
If the reader of the message is not the intended recipient or an authorized
representative of the intended recipient, you are hereby notified that any
dissemination of this communication is strictly prohibited. If you have
received this communication in error, notify the sender immediately by
return email and delete the message and any attachments from your system.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Filtering bean properties
Konrad Billewicz wrote: How would you handle it? If the columns are all formatted the same (or even if they're not and I'm feeling saucy) I might consider writing a custom tag that does the entire report (maybe outputting XML for easy format changes? I dunno) or iterating over a collection of columns to show and using existing or custom tags to output the appropriate value or table cell. Or I might just have a million c:if... tags, but I really hate JSP pages like that. Dave - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
bean:write format
Has Anyone used the format attribute to format currency? I have a property in an ActionForm which will hold a dollar figure and want to format when displaying. ie. listPrice = 930 Need to Display $930.00 or listPrice = 1299.9 Needs to Display $1,299.90 I have attempted the following bean:write name=myForm property=listPrice format=\\$\\d+\\.\\d\\d/ but that prints \$\D+\930.\d\d not exactly what I was hoping for :)
Re: bean:write format
Never mind was using regular _expression_ formatting, Needed to use format=$###,000.00 [EMAIL PROTECTED] [EMAIL PROTECTED] 07/28/2005 03:58 PM Please respond to Struts Users Mailing List user@struts.apache.org To user@struts.apache.org cc Subject bean:write format Has Anyone used the format attribute to format currency? I have a property in an ActionForm which will hold a dollar figure and want to format when displaying. ie. listPrice = 930 Need to Display $930.00 or listPrice = 1299.9 Needs to Display $1,299.90 I have attempted the following bean:write name=myForm property=listPrice format=\\$\\d+\\.\\d\\d/ but that prints \$\D+\930.\d\d not exactly what I was hoping for :)
Re: Filtering bean properties
You could have a collection/array of fields that needed to be
displayed. You would iterate on each record in the result set. For
each row, you would then iterate thru the collection of field names,
using the field name as a key to the record bean. I'd do it
something like below, though there may be some syntax errors
(particularly the c:out line, it's from memory). Again, this only
works if your records are key/value pairs (e.g. a Map) that can be
evaluated by the JSTL and EL.
c:set var=fieldList value=${myFieldList}/
c:set var=records value=${myRecordList}/
...
!-- print out record header --
tr
c:forEach var=field value=${fieldList}
thc:out value=${field}//th
/c:forEach
/tr
c:forEach var=record items=${records}
tr
c:forEach var=field items=${fieldList}
tdc:out value=${record[field]}//td
/c:forEach
/tr
/c:forEach
...
-ed
On 7/28/05, Dave Newton [EMAIL PROTECTED] wrote:
Konrad Billewicz wrote:
How would you handle it?
If the columns are all formatted the same (or even if they're not and
I'm feeling saucy) I might consider writing a custom tag that does the
entire report (maybe outputting XML for easy format changes? I dunno) or
iterating over a collection of columns to show and using existing or
custom tags to output the appropriate value or table cell.
Or I might just have a million c:if... tags, but I really hate JSP
pages like that.
Dave
-
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]
Re: [OT] Hibernate vs. iBatis vs. POJO
Hernandez, Mariano wrote: we use EJBs, although I and a colleague have tried to persuade our architech to use ANYTHING other than EJBs. Consider having a email Dear Achitect, thanks for meeting w/ us, we brought you some good reasons not to use EJB; we understand you do want to use them. -- just in case it heads south. -- .V People are conversing... without posting their email or filling up their mail box. roomity.com http://roomity.com/launch.jsp No sign up to read or search this Rich Internet App. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
On 7/28/05, Werner Punz [EMAIL PROTECTED] wrote: Michael Jouravlev wrote: On a side note, I cannot get why storing data in an action form is considered wrong, while directly accessing session is OK. I think Well it is quite the opposite in JSF... using the session is sort of a secure source of getting heavy beating (which is half right but too extreme), while the so called backend beans are combined controller/nmodel objects which basically are kept over all phases of the rendering cycle. The no use of session if possible results in such things like x:saveState which serializes everything to pass data over the cycle boundaries without having to pass parameters. Ahem, actually I am all in for sessions. I just don't see why it storing stuff in an action form with session scope is universally frowned upon, while sticking same stuff directly into session is OK ;-) Michael - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
Michael Jouravlev wrote the following on 7/28/2005 4:30 PM: Ahem, actually I am all in for sessions. I just don't see why it storing stuff in an action form with session scope is universally frowned upon, while sticking same stuff directly into session is OK ;-) Most seem to frown on both scenarios. Personally, I like the Session:) The concept of the session cleanup tool has been talked about so many times on this list, but I've yet to see an easy implementation of it. For me, I'd rather use the Session until some manager starts complaining that the app has performance issues (and even then I'd probably add more RAM vs throw away my reliance on the Session:) For a wizard like Werner initially described, I'd use a Session based form in a heart-beat. I use session scoped forms and they work wonderfully. -- Rick - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
Werner Punz wrote the following on 7/28/2005 11:46 AM: There are some ways to handle this a) You can dump anything into the session, which means you run into memory problems in the long run. I use this solution and have yet to run into any sort of memory problems. Resources aren't 'that' expensive. Use the Session unless you find out you have performance issues. b) you can put everything into hidden fields, which means you have myriads of hidden fields, which is slow and error prone to implement. Maintenance nightmare. c) You can try to add the stuff manually to the session and remove it later, which also is problematic in case of an error because then you have a mem leak caused, by a tangeling object in the session. Above is talked about a lot on this list, but have yet to see a good implementation of it. I think it will be a difficult undertaking and could be a pain to configure. The control mechanism will have to figure out when and what to clear out of the Session at any given time. Say the user clicks on some other link.. that path(I'm assuming) will have to be checked to see if it's safe to remove x number of Session based objects. I've tossed around with the idea of associating certain 'paths' with a Session scoped action form and as each request passes through the main filter it would check the path and based on the path and certain objects in scope certain ones would be be removed. Seemed like such a pain though so I gave up before I even really started on it:) -- Rick - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
Rick- I'm faced with this same problem on a new project and have been thinking along these same lines. What kind of issues have you run into? The kind of things that I see being a problem is specifying which sets of actions are associated with what items in a session, and which sets of actions are resumable and which ones clear state if one navigates away from the path. Oh yeah, and the general mechanism used to store information, is it one big bean or a session, but that's almost trivial compared to the other things I mentioned. Note I've purposefully avoided mentioning the W [wizard] word :-) -ed On 7/28/05, Rick Reumann [EMAIL PROTECTED] wrote: Werner Punz wrote the following on 7/28/2005 11:46 AM: c) You can try to add the stuff manually to the session and remove it later, which also is problematic in case of an error because then you have a mem leak caused, by a tangeling object in the session. Above is talked about a lot on this list, but have yet to see a good implementation of it. I think it will be a difficult undertaking and could be a pain to configure. The control mechanism will have to figure out when and what to clear out of the Session at any given time. Say the user clicks on some other link.. that path(I'm assuming) will have to be checked to see if it's safe to remove x number of Session based objects. I've tossed around with the idea of associating certain 'paths' with a Session scoped action form and as each request passes through the main filter it would check the path and based on the path and certain objects in scope certain ones would be be removed. Seemed like such a pain though so I gave up before I even really started on it:) -- Rick - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
1) I use session-scoped ActionForms with extra fields plenty (in search scenarios for example, where the form is for search, where the search is related to a product, and where you want to hang on to the product details as the search results change) and couldn't care less who frowns upon it. Why do you care? 2) Implementing a session cleanup manager is much easier if you *do* associate keys with paths (path == scenario/set of related use cases) and if you nest the attributes in Maps. Remove the base Map, remove everything. Remove a scenario-level Map, remove all attributes associated with that scenario. Remove a higher-level Map, remove attributes associated only with one use case or view. I use a request processor extension, where I'm already tracking user navigation. Cleaning up the session is a related operation. I usually don't bother to clean up at the page/use case level, but rather at the scenario level. When a user is done with that scenario, I remove that scenario's Map. This is of course, implemented by removing everything that doesn't pertain to the current scenario (usually triggered by a return to a home/base page). Erik - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
On 7/28/05, Ed Griebel [EMAIL PROTECTED] wrote: Rick- I'm faced with this same problem on a new project and have been thinking along these same lines. What kind of issues have you run into? I am interested too ;) The kind of things that I see being a problem is specifying which sets of actions are associated with what items in a session, and which sets of actions are resumable and which ones clear state if one navigates away from the path. I think it should be pretty easy. An action mapping can belong to one or several subsessions. This can be defined in action element of config file. In my case ;) I have pretty coarse actions, I base them on DispatchAction (actually SelectAction, but whatever). One action contains a component, and serves all requests to this component, and displays all pages of this component. It can be a Login component, or CRUD component, or a wizard. So, correlating a component to action to data that is used by it is easy. In some cases, say in one version of CRUD sample, I may have two actions which work with the same set of data. In this case both actions would correspond to one subsession. Once I load any other action which does not correspond to this subsession, I can safely remove the CRUD action form with all stuff sitting in it from the session. One line of code. Ok, two lines ;) On next request RequestProcessor will call RequestUtils.createActionForm which will create fresh instance if needed. Voila. Note that this works even with Back button, because I use no-store pages, which are reloaded on Back. Reload calls actio outside of subsession, and subsession is removed. This will fail only if a user will go to a completely different site, so at most application will keep *one* object hanging in the session. I don't care ;) Note I've purposefully avoided mentioning the F [flow] word. One of my design ideas (which probably are stolen from someone whom I've purposefully avoided mentioning too) is that flow is not needed. Each component like login, or menu or subscriptions or language selection (these are from MailReader) should be totally independent, but stateful. Flow only has to be supported for small sections, like a wizard, which fits fully into one action. Why application flow should not be needed? Because flow ties your hands and does not allow to easily develop reusable components. Say, you have flow: welcome - choose language - login - main menu - subscriptions. What if you get to main menu, then go back, log out, and then navigate to main menu again? Should flow control engine throw an exception, notifying you that you broke flow? Nah. Instead, main menu simply checks can it be shown or not (logged in or logged out). If it can, the it shows itself, if not, it goes to login page. Generic application flow is a bad thing, small action-scoped wizards rule! ;-)) Michael. -- Struts dialogs, components, controls, wizards and other good stuff http://struts.sourceforge.net/strutsdialogs - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
Ed Griebel wrote the following on 7/28/2005 5:11 PM: Rick- I'm faced with this same problem on a new project and have been thinking along these same lines. What kind of issues have you run into? I really haven't run into many problems when I use a Session based form for a wizard approach. Since you can only start the wizard typically at a particular entry point, you just make sure the session scoped form is a fresh new one at that point. I'm lazy though and don't implement a clean-up mechanism. -- Rick - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
htnm:base rewrites anchor tag
I have a jsp page which in which I am using the html:base at the top of the page. However I find that then all the links to anchor tags get rewrriten with the jsp page. Is there any way to prevent this from happening Shekhar Jain - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
On 7/28/05, Rick Reumann [EMAIL PROTECTED] wrote: Ed Griebel wrote the following on 7/28/2005 5:11 PM: Rick- I'm faced with this same problem on a new project and have been thinking along these same lines. What kind of issues have you run into? I really haven't run into many problems when I use a Session based form for a wizard approach. Since you can only start the wizard typically at a particular entry point, you just make sure the session scoped form is a fresh new one at that point. I'm lazy though and don't implement a clean-up mechanism. Wizard should be handled by *one* action, then cleanup would be simple. Have you found time to look at my stuff? ;-) Michael. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: htnm:base rewrites anchor tag
Shekhar Jain wrote: I have a jsp page which in which I am using the html:base at the top of the page. However I find that then all the links to anchor tags get rewrriten with the jsp page. Is there any way to prevent this from happening Are the links-to-anchor-tags struts tags? If so, perhaps consider using straight html tag if you don't need the struts tags' functionality; one point of html:base is to allow html:link tags to rewrite appropriately. (Or don't use html:base?) Dave - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: htnm:base rewrites anchor tag
No I'm just using a plain a href=#top ... to go to the top of the page Shekhar On 7/28/05, Dave Newton [EMAIL PROTECTED] wrote: Shekhar Jain wrote: I have a jsp page which in which I am using the html:base at the top of the page. However I find that then all the links to anchor tags get rewrriten with the jsp page. Is there any way to prevent this from happening Are the links-to-anchor-tags struts tags? If so, perhaps consider using straight html tag if you don't need the struts tags' functionality; one point of html:base is to allow html:link tags to rewrite appropriately. (Or don't use html:base?) Dave - 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]
Re: bean traversal question
Michael Jouravlev wrote the following on 7/28/2005 6:48 PM: Wizard should be handled by *one* action, then cleanup would be simple. Have you found time to look at my stuff? ;-) Sorry Michael I have not:( Send me the link again. But, that being said, where clean up becomes a problem is when the user goes to some other action? I agree one Action helps but what if they click on some other link from your menu that goes to some other Action? In other words they don't 'officially' quit the wizard flow. That's when it becomes a bit of a pain (assuming you care to clean up:) -- Rick - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Problem with Dyna Form
Kade Jeevan Kumar wrote: Hi All, My requirement is to generate a set of dynamic rows(with checkbox and combo box) at client side using JSP and Struts 1.1. As of now, i am generating the above stuff using java script. How can i use Dyna Form to meet the above requirement (or) suggest me other way if u have any solution for this. JSP / Struts are server-side, not client-side. If what you need is to be able to add rows to a page without making another request to the server each time, you'll need to stick with a Javascript solution. If what you want to do is generate a page with a number of rows generated dynamically at request time, you have a range of options. If that's what you need you have a wide range of options ;-) Can you explain your requirements in more detail? L. -- Laurie, Open Source advocate, Java geek and novice blogger: http://www.holoweb.net/laurie - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: htnm:base rewrites anchor tag
Shekhar Jain wrote: No I'm just using a plain a href=#top ... to go to the top of the page I think you need to look at what the html base... tag does. http://www.faqs.org/docs/htmltut/linking/_BASE.html Dave - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: multipart request issues
Carl Smith wrote:
I am having issues to get the request parameters from a multipart request, le
me explain my problem first:
Our flow of process is:
(1) a multipart request is submitted to a struts action (let's say
updateTableAction.do) using POST
(2) the multipart request is processed by ActionServlet.process() method
(3) in our processPreprocess method, which overrides processPreprocess in
RequestProcessor.java from struts,
we need the parameters from this multipart request thus I am doing this:
CommonsMultipartRequestHandler commonMultiPartHandler = new CommonsMultipartRequestHandler();
try {
commonMultiPartHandler.handleRequest(request);
}catch(Exception e){
// some handling
}
Hashtable ht = commonMultiPartHandler.getTextElements(); // thus I get the
parameters from the request
(4) however, all the fields/parameters in multipart request are no longer there
in the updateTableAction.do
Can you help me find out why this is happening? One theory is that a multipart
request can't be parsed twice. Is this correct, if it is correct, do you have
any alternative?
That's certainly true; once the request has been consumed (parsed) it can't
be parsed again since the data is no longer there. I don't know the normal
flow of RequestProcessor but I assume you need to store the result of
handleRequest() in the same way as Struts normally does -- probably using
an HttpRequestWrapper. If that's not normally handled in
processPreprocess(), you'll need to also override whichever method in
RequestProcessor does handle it.
L.
--
Laurie, Open Source advocate, Java geek and novice blogger:
http://www.holoweb.net/laurie
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
What we have done is kept all forms in session as form. That way if the user jumps to another set of pages and discards halfway tru the current process the form session variable just gets reused and we don't have to worry too much about cleanup Shekhar On 7/28/05, Rick Reumann [EMAIL PROTECTED] wrote: Michael Jouravlev wrote the following on 7/28/2005 6:48 PM: Wizard should be handled by *one* action, then cleanup would be simple. Have you found time to look at my stuff? ;-) Sorry Michael I have not:( Send me the link again. But, that being said, where clean up becomes a problem is when the user goes to some other action? I agree one Action helps but what if they click on some other link from your menu that goes to some other Action? In other words they don't 'officially' quit the wizard flow. That's when it becomes a bit of a pain (assuming you care to clean up:) -- Rick - 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]
Re: UrlValidator (why not use java.net.URL?)
I'd guess the reason is to make it easier to keep the client- and server-side validation in sync. java.net.URL is, obviously, not available to the client-side validation. The Commons Validator list is probably the right place to follow this up. L. Marc Logemann wrote: Hi, after looking into the code of UrlValidator and moreover looking into a similar class in Tapestry which is no longer active as it seems, i wonder why commons-validator UrlValidator is using Regex so much. Wouldnt it be enough to just use java.net.URL and let this class do the verification? Of course i am not quite sure how it deals non-schema URLs but for complete URLs to verifiy, its perhaps a better choice. Perhaps this one is for the developer list but i try it here first ;-) Lindholm, Greg wrote: I encountered this awhile back but didn't have time to investigate. As a workaround I use the localhost IP address 127.0.0.1 but it would be nice to get this fixed. -Original Message- From: Marc Logemann [mailto:[EMAIL PROTECTED] Sent: Thursday, July 28, 2005 8:33 AM To: user@struts.apache.org Subject: UrlValidator Hi, can it be that UrlValidator from Commons-Validator doesnt validate URLs with localhost in it? The following URL - http://localhost:8081/context/jsp/versand_1.jsp -- Laurie, Open Source advocate, Java geek and novice blogger: http://www.holoweb.net/laurie - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
On 7/28/05, Rick Reumann [EMAIL PROTECTED] wrote: Michael Jouravlev wrote the following on 7/28/2005 6:48 PM: Wizard should be handled by *one* action, then cleanup would be simple. Have you found time to look at my stuff? ;-) Sorry Michael I have not:( Send me the link again. No rush :) Though you might like it, because it uses session. On the other hand, you may hate it, because it uses DispatchAction flavor ;) Project link: http://struts.sourceforge.net/strutsdialogs Login/Logout/Signup combo component: http://www.superinterface.com/wizard/signupWizard.do Signup Wizard only: http://www.superinterface.com/strutsdialog/wizardaction.do All samples: http://www.superinterface.com/strutsdialog/index.htm But, that being said, where clean up becomes a problem is when the user goes to some other action? I agree one Action helps but what if they click on some other link from your menu that goes to some other Action? In other words they don't 'officially' quit the wizard flow. That's when it becomes a bit of a pain (assuming you care to clean up:) No pain. You just have to decide, do you want to allow people to leave wizard in the middle and then come back, you you do not. If you do, then you leave it as it is, so when they come back to the same component, they would see same page and same data. If you don't want to keep it, then trash the wizard. This should be set per component. There should be set of policies. For example, if you navigate to an action which is not part of any subsession (or scenario, I like this word), do you trash all existing scenarios or you keep them? The downside of my components, is that there can be only one instance of every component per session. Wicket guys, for example, use versions. But their version usage is not very smart, they simply create new version each time a component is accessed. I need to think of better way of creating of several versions to be used either on one page, or in different browser windows. But I am not sure that this is possible... Michael. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: bean traversal question
On 7/28/05, Michael Jouravlev [EMAIL PROTECTED] wrote: There should be set of policies. For example, if you navigate to an action which is not part of any subsession (or scenario, I like this word), do you trash all existing scenarios or you keep them? If you look at Spring Web Flow, you would see that they allow to define huge app-sized flows and subflows. This is madness, and I do not favor this. I think it is much easier not to have a flow at all, and instead to define single-level peer components (resources, actions, scenarios, whatever you call them). If you need a wizard, make it one action. The idea of subflows is do you allow certain set of actions in certain state or not. State is the keyword. If state is used uniformly in application, subflows would not be needed, just set or clear the state. Right... Essentially, flow with sublows is a hierarchical state. I guess I need to think of something generic app-wide state management. Michael. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: String concatenation
Konrad Billewicz wrote:
Dave Newton newton at pingsite.com writes:
I've never put a parameter in a form's action attribute--why would you
want to do that? Wouldn't it be much cleaner to use a hidden form field
with the appropriate parameter name?
So your original example would be (minus whatever silly syntax mistakes
because I'm groggy):
html:form action=createDictionaryEntity
html:hidden property=dictionary/
!-- ... --
/html:form
and whatever loads the form puts the appropriate dictionaryId value into
the ActionForm?
I wish to pass the dictionary variable using GET - make it visible in the URL.
Secondly, this action (createDictionaryEntity) can be accessed both from form
and from link (using a href=...). I wish to have one, clear method of
passing parameters to it. Of course, I can at first check the request and
secondly the form (or inversely) but it is not, in my opinion, elegant solution.
I thing it is not compulsory.
Firstly, you can combine get with forms; if you use method=GET the hidden
parameter Dave suggested will show up in the URL. Either way, you wouldn't
have to 'check the request and secondly the form' since GET and POST
parameters are both accessed the same way, though request.getParameter() or
via your form bean.
If you really still want to construct the URL with a query parameter,
though, you have a few options. Using JSTL:
html:form action=createDictionaryEntity.do?
dictionary=${requestScope['dictionaryId'}
Alternatively you might be able to do something along the lines of (off the
top of my head; check the docs for syntax):
c:url var=action value=createDictionaryEntity.do
c:param name=dictionary
bean:write name=dictionaryId scope=request/
%= dictionaryId %
/c:param
/c:url
html:form action=% action %
But that still uses JSTL tags; I don't know if you can get to the same
result with only Struts tags an on EL or runtime expressions.
L.
--
Laurie, Open Source advocate, Java geek and novice blogger:
http://www.holoweb.net/laurie
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Merging form properties
Wendy Smoak wrote: From: Cadariu, Mihai [EMAIL PROTECTED] I have two instances of the same form bean class, both containing some properties filled, the rest being null. I would like to copy the non-null properties from one form bean to the other. I tried using the PropertyUtils.copyProperties or BeanUtils.copyProperties but they copy all properties, whether they are null or not. Is there a way to tweak the copying, so that only non-null values are overwritten? I haven't done it, but you can use a BeanInfo class to control what properties are exposed: http://java.sun.com/docs/books/tutorial/javabeans/beaninfo/ That wont work, though, if you don't know at design time which properties you need to copy and which to ignore. My first thought if that's an issue would be to describe() the source bean into a map, filter the null valued properties, then populate() the target bean. Seems a bit long-winded though. It might be worth asking on the BeanUtils list. L. -- Laurie, Open Source advocate, Java geek and novice blogger: http://www.holoweb.net/laurie - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Merging form properties
Laurie Harper wrote: Wendy Smoak wrote: I haven't done it, but you can use a BeanInfo class to control what properties are exposed: http://java.sun.com/docs/books/tutorial/javabeans/beaninfo/ That wont work, though, if you don't know at design time which properties you need to copy and which to ignore. Is that true? I glanced briefly at the tutorial and it wasn't immediately clear to me exactly when the BeanInfo stuff was used: if in the getPropertyDescriptors method you checked the values for null and only returned those that weren't null would this work? Dave - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Merging form properties
From: Dave Newton [EMAIL PROTECTED] if in the getPropertyDescriptors method you checked the values for null and only returned those that weren't null would this work? That's what I was thinking... but looking at the BeanInfo interface, you don't have access to the bean itself, so as Laurie said you would have to know in advance which properties to expose. Never mind, then. :) You should probably ask on commons-user... I bet there's a simple way to do this. (The next thing that comes to mind is subclassing BeanUtilsBean and overriding the copyProperty method to make it ignore properties for which the incoming value is null. Would that work?) -- Wendy Smoak - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] saving stats from session when session destroyed
On 7/25/05, Frank W. Zammetti [EMAIL PROTECTED] wrote:
SessionListener is what you want.
Here's an example from one of my apps (condensed for space)... it is used
to delete a temporary PDF that may have been generated for the user when
they log off:
import company.app.User;
import javax.servlet.http.HttpSession;
import javax.servlet.http.HttpSessionEvent;
import javax.servlet.http.HttpSessionListener;
public class MySessionListener implements HttpSessionListener {
public synchronized void sessionCreated(HttpSessionEvent se) {
}
public synchronized void sessionDestroyed(HttpSessionEvent se) {
HttpSession sess = se.getSession();
User user = (User)sess.getAttribute(user);
String userID = user.getUserID();
AppHelpers.deletePDF(userID);
}
}
I could be losing my mind, but, I swear that I tried this technique
once and by the time the sessionDestroyed() method was called, I was
not able to get any of the session attributes back anymore. For
various reasons, the webapps I develop need to be deployable in a
Servlet 2.3 environment. Is this something where the behavior changed
for the better in Servlet 2.4 versus 2.3?
-Van
--
- Mike Van Riper
[EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Path like urls with objects' OIDs: /objects/object/17/198/4 possible?
This might be useful: http://article.gmane.org/gmane.comp.jakarta.struts.user/110666/match= Tomasz Nazar wrote: Hi! I like the idea (pattern) of having actions named in a path/menu-like way. Say: /context/employees/emplyee?id=17 But: I'd like to have something more prettier: Say: /context/emplyee/17 Or even more: /context/archive/1998/employee/tomasz/tasks/7 Can you point me where to look for this? Is there any standard option for that; some URL pattern rewriting or God knows what. Do I have to write my own implementation of some method/class and then parse the url, set proper parameters and pass over to Struts (dispatcher?) ? Thanks in advance - Tomasz -- Laurie, Open Source advocate, Java geek and novice blogger: http://www.holoweb.net/laurie - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] saving stats from session when session destroyed
Nope, not as far as I know... in fact, I just implemented a listener
today that does some maintenace-type things like the code below, and it
works fine, that's under Tomcat 5.5.28 I believe... spec 2.3 as I tecall?
sessionDestroyed() would be all but useless if you couldn't get at the
attributes, I would be utterly shocked if the spec changed in that
way... I'd bet you saw either a bad implementation, or you actually ARE
losing your mind :) (not to worry, wherever yours is, mine is right
behind it!)
Frank
Van wrote:
On 7/25/05, Frank W. Zammetti [EMAIL PROTECTED] wrote:
SessionListener is what you want.
Here's an example from one of my apps (condensed for space)... it is used
to delete a temporary PDF that may have been generated for the user when
they log off:
import company.app.User;
import javax.servlet.http.HttpSession;
import javax.servlet.http.HttpSessionEvent;
import javax.servlet.http.HttpSessionListener;
public class MySessionListener implements HttpSessionListener {
public synchronized void sessionCreated(HttpSessionEvent se) {
}
public synchronized void sessionDestroyed(HttpSessionEvent se) {
HttpSession sess = se.getSession();
User user = (User)sess.getAttribute(user);
String userID = user.getUserID();
AppHelpers.deletePDF(userID);
}
}
I could be losing my mind, but, I swear that I tried this technique
once and by the time the sessionDestroyed() method was called, I was
not able to get any of the session attributes back anymore. For
various reasons, the webapps I develop need to be deployable in a
Servlet 2.3 environment. Is this something where the behavior changed
for the better in Servlet 2.4 versus 2.3?
-Van
--
Frank W. Zammetti
Founder and Chief Software Architect
Omnytex Technologies
http://www.omnytex.com
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Merging form properties
Wendy Smoak wrote: From: Dave Newton [EMAIL PROTECTED] if in the getPropertyDescriptors method you checked the values for null and only returned those that weren't null would this work? That's what I was thinking... but looking at the BeanInfo interface, you don't have access to the bean itself, so as Laurie said you would have to know in advance which properties to expose. Never mind, then. :) You should probably ask on commons-user... I bet there's a simple way to do this. (The next thing that comes to mind is subclassing BeanUtilsBean and overriding the copyProperty method to make it ignore properties for which the incoming value is null. Would that work?) For what it's worth, I think I just realized I need to solve the same problem, so I'll try and remember to post here if I come up with a nice solution. If you get anything from commons-user drop me a note ;-) L. -- Laurie, Open Source advocate, Java geek and novice blogger: http://www.holoweb.net/laurie - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Where can I find taglibs
I want to write a web app with jsp/struts. I know that there should be a struts tag library for this. Where can I find it? Thanks, -Paul - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] saving stats from session when session destroyed
On 7/28/05, Frank W. Zammetti [EMAIL PROTECTED] wrote:
Nope, not as far as I know... in fact, I just implemented a listener
today that does some maintenace-type things like the code below, and it
works fine, that's under Tomcat 5.5.28 I believe... spec 2.3 as I recall?
I believe that Tomcat 4.x was at spec 2.3 level and Tomcat 5.x is at
spec 2.4 level. In any case, I'm using a Jetty 4.2.x container which
I'm absolutely certain only has spec 2.3 support.
sessionDestroyed() would be all but useless if you couldn't get at the
attributes, I would be utterly shocked if the spec changed in that
way... I'd bet you saw either a bad implementation, or you actually ARE
losing your mind :) (not to worry, wherever yours is, mine is right
behind it!)
I totally agree, but, all I was able to get was the jsessionid. It
turns out we had a database table for auditing purposes already. So, I
was able to stick the jsessionid into the login audit entry for the
user. This allowed me to determine the user associated with session
just destroyed so that I could add an audit table entry for the
matching logout for that user too.
Believe me, I wouldn't have gone to all that trouble if the session
data was available for me to get to the user attribute in the session.
This has stirred my curiosity. I'll give it another go for my current
project (we weren't using Jetty on the previous project) and report
back.
-Van
Frank
Van wrote:
On 7/25/05, Frank W. Zammetti [EMAIL PROTECTED] wrote:
SessionListener is what you want.
Here's an example from one of my apps (condensed for space)... it is used
to delete a temporary PDF that may have been generated for the user when
they log off:
import company.app.User;
import javax.servlet.http.HttpSession;
import javax.servlet.http.HttpSessionEvent;
import javax.servlet.http.HttpSessionListener;
public class MySessionListener implements HttpSessionListener {
public synchronized void sessionCreated(HttpSessionEvent se) {
}
public synchronized void sessionDestroyed(HttpSessionEvent se) {
HttpSession sess = se.getSession();
User user = (User)sess.getAttribute(user);
String userID = user.getUserID();
AppHelpers.deletePDF(userID);
}
}
I could be losing my mind, but, I swear that I tried this technique
once and by the time the sessionDestroyed() method was called, I was
not able to get any of the session attributes back anymore. For
various reasons, the webapps I develop need to be deployable in a
Servlet 2.3 environment. Is this something where the behavior changed
for the better in Servlet 2.4 versus 2.3?
-Van
--
Frank W. Zammetti
Founder and Chief Software Architect
Omnytex Technologies
http://www.omnytex.com
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
- Mike Van Riper
[EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Where can I find taglibs
The Struts taglibs currently ship with the main Struts distribution. Download Struts and you get the taglibs. FYI, the development branch has them separated out now as a separate sub-project. I'm not sure how they are intended to be distributed, but for now it doesn't really matter. Frank Paul Goepfert wrote: I want to write a web app with jsp/struts. I know that there should be a struts tag library for this. Where can I find it? Thanks, -Paul - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] . -- Frank W. Zammetti Founder and Chief Software Architect Omnytex Technologies http://www.omnytex.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [OT] saving stats from session when session destroyed
Van wrote: I believe that Tomcat 4.x was at spec 2.3 level and Tomcat 5.x is at spec 2.4 level. In any case, I'm using a Jetty 4.2.x container which I'm absolutely certain only has spec 2.3 support. I'd bet your right about Tomcat... of course, either of us could go look it up in 10 seconds :) Let's stick to the suppositions though, they're more fun :) Believe me, I wouldn't have gone to all that trouble if the session data was available for me to get to the user attribute in the session. Interesting... I'd bet it was a peculiarity of the container's spec implementation. Did you say it was Jetty at that time too? I'm not at all familiar with Jetty, so it's just a guess. This has stirred my curiosity. I'll give it another go for my current project (we weren't using Jetty on the previous project) and report back. Please do... I'd be intersted to know the answer as well :) -Van Frank - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: probably a cleaner way... testing for just one user
Hi Rick, That would probably be a good idea to use two filters vs doing it all in one. Although it could also be nice to see all that happens for each request by looking in just one filter vs two. I like this idea because in my opinion filters should do small tasks. This way they can be used separately or they can be combined in a chain to achieve some more complicated functionality. For example Frank's JWP could have a filter which limits the number of sessions for a configurable max value and another filter which checks for a session timeout. If the session timeout occurs it can redirect to a configurable page. This way you have two filters which provide functionality that can be used separately or they can be combined. And I don't think that it will be confusing for someone to look at two filters. They don't have to check the code. The filters are black-boxes and you have to look just in the web.xml to know what is going on. You will see that first all requests are filtered to limit the sessions and then that expired sessions are redirected by the second filter. 2) I noticed a lot of odd behavior when the server was shut down and the pages were still up. For example, if the user was left on a page that displayed a log off link and the server was restarted, when they then clicked on log off (which calls session invalidate) it would decrement the sessionCounter to -1. I'm not sure how sessionDestroyed could get called before sessionCreated, but I guess it can? My hack here was to do a check for sessionCount being less than 0 and if so reset it to 0. This is strange. What kind of webcontainer do you use? At first glance it seems like your web container has a 'feature' to serialize sessions on stop and to recreate them when you start the server again. In this case the sessionCounter would be set to 0 on restart but the session would be recreated = no sessionCreated() executed, just sessionDestroyed() on logout... Possibly that is what is happening. The scenario where I noticed this was using JBoss (Tomcat as the app server in JBoss). I've just checked it and Tomcat has this feature. However you must configure Tomcat to never recover your sessions! I'm afraid that your check if sessionCount is less than 0 is just not enough. Ex: User A uses the app Tomcat is restarted User B comes in and gets a new session (sessionCount becomes 1 which is the max allowed) User A tries to use the app and he has a recovered and unflagged session. sessionCreated will not be executed and nothing will prevent the user A to use the app while user B is using it. If you put a META-INF/context.xml in your WAR with the following content: Context Manager class=org.apache.catalina.session.StandardManager pathname= / /Context you should be ok. The Manager can be also configured to limit the sessions with the maxActiveSessions attribute :-D So if it was not a problem for you to tie your implementation to Tomcat then everything would be easier. You set the maxActiveSessions attribute to 1, and in the filter when you execute request.getSession you catch the IllegalStateException thrown by Tomcat. If it's thrown you redirect to appInUse.jsp You can forget about the SessionListener, you won't need it. However, I would like to have a general solution for this and since we took the time to think it over we could have this code somewhere so it will be reusable in the future. For example Frank's JWP would be a great place to include these filters, if he accepted to include them in his project. Regards, Tamas p.s. the open question: you want a Tomcat specific solution or a general one? - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: JSF is the beginning of the end of Struts !!!
Michael / Jack / Thingamabob, Struts is licensed in such a way that if you are unhappy with the way it is being managed, you are quite free to fork the code and continue to develop it in whatever way *you* see fit. I suspect you'd make a lot of people very happy if you went away and did that. Steve Dakota Jack wrote: Struts could live by improving in ways unlikely with the present management. The people in charge of Struts these days are really interested in something else. Unless the baton passes to those actively building something, Struts likely will either die from be non-competitive with like products or from being turned into something it is not, like JSF, Shale, Clay, Dirt. On 7/26/05, Greg Reddin [EMAIL PROTECTED] wrote: Interesting this discussion would come up now. I've been in and out of Struts for about 5 years. I'm now coming back in and taking interest in JSF/Shale, etc. A few of points to think about: 1. Struts will someday die. If it doesn't, then we've seen the end of technology advancement. If those of us who love Struts and what it provides want to continue to have influence going forward, we have to be ready to embrace the next thing that's better than Struts and add value there. Staying with something because of familiarity is typically the road to becoming a dinosaur in technology. 2. I only started studying JSF seriously about 2 weeks ago. From what I've seen it takes the best things Struts offers and brings them into a somewhat cleaner package. Those things are: * Nice HTML tag library. * Auto-populating of form beans. * Simple controller interface. It also expands on these things by making the view easier to develop, less restriction on how form beans and controllers are developed, and some other really interesting services that I've only begun to play with. To sum it up, using JSF out of the box I can do most of what I can with Struts, only quicker and with less code. I see JSF as the next step of Struts. But it's still not complete. I'm only beginning to see what Shale adds to JSF and am following Shale with great interest. 3. PHP. I've done some PHP over the last couple years. I'm by no means an expert, but I do know my way around somewhat. The whole time I've caught myself thinking if I was in Java I'd have a better way to do just about everything. But most of the better way is because of the J2EE APIs, Struts, and Tiles. The language is a bit more cumbersome than Java and it's much more difficult to modularize. I find PHP is easier (to me) if I don't try to make things elegant, but just spew code to some extent. I finally found a CMS to use as a Struts-like framework, but I still find Java/Struts and now JSF to be better, cleaner, and more modular. I am now trying to figure out a way to do my future on-the-side consulting in Java instead of PHP. I think another of PHPs success factors has to do with its acceptance from the open source community. Somehow Java has had a somewhat tainted reputation due to the lack of a truly open complete implementation -- and probably some are just bitter that Sun hasn't opened the whole thing up. I don't personally have anything to contribute to that debate, but I suspect it's a factor in why PHP has been so widely embraced on the web compared to Java. So all in all, it doesn't much matter to me if JSF supercedes Struts. If it's better, then it should. If we get behind Shale it think we can have our cake and eat it too. BTW, for JSF info, try MyFaces. Their Getting Started section points to some pretty good tutorials. In short testing I've found the myfaces implementation to be pretty stable. I'm actually surprised Craig is not listed as a contributor to that project. That's my 50 cents. Greg Daniel Perry wrote: Sorry for the OT postings. My point was that you cant compare usage of PHP with Struts. The number of Struts sites (or even java sites) will never overtake the number of PHP sites for the reasons i pointed out (although... zend are doing their best to kill off php by trying to move it into the enterprise arena and tying it with java). But for those same reasons, comparing the two is useless. As for JSF... It seems like a nice idea, but i havnt fully got my head round it. I cant see it killing off struts in the short term (just look at daily rate of postings on this list!!!) I do like the idea of replacing struts html tags with JSF. I certainly think JSTL/EL are nicer than the logic tags. I now have a dilema... I'm just about to start on another project. What technologies do i use? I guess i'll probably stick with struts. Though i may dabble with JSF a bit. Can anyone recommend any good resources? Sure a google search provides tons of information... but which is any good? Daniel. -Original Message- From: Mark Benussi [mailto:[EMAIL PROTECTED] Sent: 26 July 2005 10:38 To: 'Struts Users Mailing List' Subject: RE: Re: JSF is the beginning of
Re: probably a cleaner way... testing for just one user
Tamas Szabo wrote: For example Frank's JWP would be a great place to include these filters, if he accepted to include them in his project. I would absolutely accept them, so long as they work :) And frankly, right now I'm trying to figure out why my formerly working version of the session limiting stuff isn't working any longer, so I'd be even MORE inclined to accept them now :) The session timeout filter I'm not as sure about only because it seems redundant... isn't there a way to do that already built in to the spec? Weird... I'm looking at a web.xml reference right now and I don't see anything that allows you to set a page to redirect to when a session times out... can that really be the case? You CAN redirect to a specific page based on an HTTP error code, do you maybe get a specific code when a session times out? If there isn't a way to do this, then yeah, such a filter would be an excellent (and one would think easy!) addition. In any case, feel perfectly free to lob any additions to JWP you like my way. There's no way in the world I can do it all alone :) -- Frank W. Zammetti Founder and Chief Software Architect Omnytex Technologies http://www.omnytex.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
