Re: message bundle from listener
::SammyRulez:: [EMAIL PROTECTED] wrote on 08/25/2005 06:11:54 PM:
Hi all
I would like to store in an application scope variable a message
bundle via listener when the application stat up... something like:
public void contextInitialized(ServletContextEvent event) {
ResourceBundle strings = ResourceBundle.getBundle(strings);
event.getServletContext().setAttribute(labels,strings);
}
and retrive the string messages via jstl whit fmt:message
fmt:message bundle=${labels} key=appWelcome/
but I bump in an exception:
org.apache.jasper.JasperException: jsp.error.beans.property.conversion
org.apache.jasper.runtime.JspRuntimeLibrary.
getValueFromPropertyEditorManager(JspRuntimeLibrary.java:885)
any ideas on what obj to use to store string in the application scope
via servlet context?
snip/
javax.servlet.jsp.jstl.fmt.LocalizationContext
i.e.
event.getServletContext().setAttribute(labels, new
LocalizationContext(strings));
And if anyone asks, I insisted on the two argument constructor ;-)
[
http://java.sun.com/products/jsp/jstl/1.1/docs/api/javax/servlet/jsp/jstl/fmt/LocalizationContext.html
]
-Rahul
thanks
--
::SammyRulez::
http://sammyprojectz.blogspot.com
Re: Dynamic vars in c:set
Dylan MacDonald [EMAIL PROTECTED] wrote on 08/25/2005 06:42:03 PM:
Hi -
I was wondering whether it was possible to set the var in the c:set tag
dynamically? For example, within my forEach tags I want to build a
list of page variables dynamically based on the index (or the varStatus
count) of the loop:
c:forEach var=parameter items=${paramValues} varStatus=status
c:set var=variable${status.count} value=${parameter.key} /
/c:forEach
which I can then access in the page as variable1, variable2, etc. I
assumed this was alright since I am using EL within the var attribute,
but alas it doesn't seem to work. Any ideas or workarounds?
snip/
You can't gensym name-from-attribute's in JSP. The more important question
is why you think it is necessary. In other words, even if you created
variable1, ... ,variableN in pageContext, you'd still want to run a loop
through them to do something else. You should be able to do without
gensym, whatever it is you are doing.
-Rahul
Thanks,
Dylan MacDonald
Re: Uri confusion
Murray Steele [EMAIL PROTECTED] wrote on 08/10/2005 04:15:26 AM: Hi, The _rt bit means that the taglibs understand runtime expressions as the values of their attributes. snip/ On 9 Aug 2005, at 08:41, Karianne Berg wrote: Hi, I'm quite new to JSTL, and crawling through tutorials and code examples on the net got me confused. In the documentation, it says that in order to use JSLT 1.1, you have to use the following uri: http://java.sun.com/jsp/jstl/core . However, I have seen that most code examples use this uri with their JSTL 1.1 examples: http://java.sun.com/jstl/core_rt. I managed to find out that the mystical rt meant runtime, but that was about it. What is the difference between those two? Which should I use? In addition to the earlier reply, with reference to the Standard implementation, the former is part of JSTL 1.1 (JSP 2.0 taglib), the latter is part of JSTL 1.0 (JSP 1.2 taglib). This page doesn't answer your question directly, but you might find it useful [ http://wiki.apache.org/jakarta-taglibs/FrequentlyAskedQuestions ] -Rahul
Re: Struts Validator - missing longRange and doubleRange
Givler, Eric [EMAIL PROTECTED] wrote on 08/10/2005 08:16:25 AM: Our development tool at the office includes an older version of Struts (v1.16). The tool mentions some incompatibilities with newer versions of struts, so we have not changed this. Anyway, there is a Validator in struts that includes an XML validation-rules file as well as a validator file (for validations specific to your application). The struts Validator site (online) mentions these two validators, but there is NO entry for either in the validator-rules file. If I open the class (org.apache.struts.validator.FieldChecks), I can see validateDoubleRange, but I don't see the other. Can anyone shed a little light on this one? snip/ Will be a more appropriate post for: user@struts.apache.org [ http://struts.apache.org/mail.html ] -Rahul
[ANNOUNCEMENT] Reusable Dialog Components (RDC) 1.0.0 released
The Apache Jakarta Taglibs team is pleased to announce the release of version 1.0.0 of the Reusable Dialog Components (RDC) tag library. JSP 2.0 based Reusable Dialog Components (RDC) is a framework for creating JSP taglibs that aid in rapid development of voice and multimodal applications. Server-side generation of HTML has proven an effective way of generating the user interface for visual Web applications that are implemented using server-side application frameworks such as Struts. Over time, the effort involved in such HTML generation has been reduced by the availability of various JSP tag libraries that abstract away the minutiae of HTML markup. The goal of the RDC project is to achieve for voice applications what JSP tag libraries have already achieved in the world of visual Web applications. * RDC webpage is here [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/intro.html ] * RDC wiki is here [ http://wiki.apache.org/jakarta-taglibs/ReusableDialogComponents ] * RDC 1.0.0 release is available for download here [ http://jakarta.apache.org/site/downloads/downloads_taglibs-rdc.cgi ]. Source and binaries are distributed separately. * RDC revision history is available here [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/changes.html ] - The Apache Jakarta Taglibs Team
Re: c:forEach index in looop
Marco Mistroni [EMAIL PROTECTED] wrote on 07/27/2005 05:30:17 PM:
hello all,
i am trying to replace struts logic:iterate with JSTL c:forEach...
however, i need what in logic:iterate is the index in the loop
is it possible to get it from anywhere?
c:forEach ... varStatus=status ...
...
I am in iteration number ${status.count}
...
/c:forEach
-Rahul
Re: TagSupport vs BodyTagSupport
Luca Passani [EMAIL PROTECTED] wrote on 07/23/2005 02:28:14 PM: ooops, I felt this was a general taglib question, more than a Struts question. It only takes struts as an example, but the question is is there a reason why it makes sense to always subclass BodyTagSupport in a tag-library? snip/ Outside of frameworks, I am usually guarded about any statement that matches always do *, especially those related to patterns and subclassing. I often extend SimpleTagSupport (JSP 2.0), and have on occassions, preferred it over BodyTagSupport, due to its simpler life cycle. For a far more detailed commentary, please read the jsp.tagext Javadocs here [ http://java.sun.com/j2ee/1.4/docs/api/javax/servlet/jsp/tagext/package-summary.html ] -Rahul
Re: URL for String TagLib source code
Henri Yandell [EMAIL PROTECTED] wrote on 07/15/2005 01:29:09 PM: On 7/15/05, Robert Taylor [EMAIL PROTECTED] wrote: Thanks Martin, but at the risk of sounding ignorant. The last source link is labeled archives... (not very intuitive) I clicked on it and then downloaded the file: jakarta-taglibs-string-current.zip This appears to be a binary distribution. Yeah, I can't find a source distribution for any of the taglibs I randomly clicked on, either in archives or on the main mirrors. Seems bad. In the mean time, current sources have always been available and linked via [ http://jakarta.apache.org/taglibs/#Downloads ], by clicking on the appropriate Source links, for supported or sandbox taglibs distributions. The sources, AFAIK, were always available as supported or sandbox bundles (rather than individual taglib sources), primarily because the top level build script fires the individual taglib builds (i.e. you'd be at a loss to build without having the bits one level up). Once you have those, however, you can use your discretion to build the taglibs that interest you. Links from the download section mentioned above lead to (Robert, you'll find the String taglib sources by following the first link below): Supported [ http://cvs.apache.org/builds/jakarta-taglibs/nightly/src/ ] Sandbox [ http://cvs.apache.org/builds/jakarta-taglibs-sandbox/nightly/src/ ] -Rahul Hen
Re: URL for String TagLib source code
Henri Yandell [EMAIL PROTECTED] wrote on 07/15/2005 01:53:07 PM: snip/ While I'm on a slightly different topic: * we really don't release source when we release code? AFAICT. Which is why I had to prepend my earlier post with In the mean time, ... Do we have tags for all past releases? I think this is becoming more appropriate for the dev list. -Rahul P.S.- Robert, thanks for bringing this up! Hen
Re: c:import and context attribute
Dylan MacDonald [EMAIL PROTECTED] wrote on 07/13/2005 05:41:46 PM: Can someone explain the context attribute in the c:import tag? snip/ It is used in cross-context imports and is the name of the context you're importing from. Most containers need to be configured to enable cross-context access. I'm working in a development environment that differs from the production environment in that the root of the server isn't /, it's /www (the production root is obviously /). snap/ I'm not sure I got this (I thought for a bit you were implying the difference between deploying, say, ROOT.war and mywebapp.war on Tomcat). In any case, relative urls like the one in the first c:import below should work if you are migrating all the contents of the context from the production to the dev environment. Portable code will prepend the context path obtained from the request at JSP invocation to any urls that need it. I don't think there is a cross-context angle to your problem, in reference to your earlier question about the context attribute. -Rahul Needless to say this makes it difficult to code anything with root-relative (e.g. /images/logo.gif) links. I can probably work around some of this but I thought I might be able to solve some of my issues with my jsp includes using the context attribute. But I cannot get it to work. I assumed I would change my c:import tag from this c:import url=includes/footer.jsp / to this: c:import url=/includes/footer.jsp context=/www / but that doesn't work. Any ideas? Thanks, Dylan MacDonald
Re: [SOLVED] Re: how to debug jsp compilation error in Tomcat 5.5 ?
Woodchuck [EMAIL PROTECTED] wrote on 06/17/2005 11:18:17 AM: The lists have been slow. Please do not send multiple copies :-) i also noticed there seems to be multiple copies of my original post. i do not know why, i only sent once. OK, sorry about the banter ;-) It seems we had quite a few incidences like that yesterday. aha! just as you stated, i found HTML tags sitting within c:choose but outside of c:when and c:otherwise well, i guess this was ok in Struts 1.1 but not anymore in Struts 1.2.7 (or is this an unintentional change?) The message is produced by the standard's TLVs (JSTL). The behavior you are seeing is expected (the semantics of c:choose means the intent of having template text outside a legal child is unclear). Not sure why it worked for you before. thanks, Rahul, and everyone that tried to help! :) You're welcome. -Rahul
Re: [OT] Using JSTL 1.1.2 with Tomcat 5.5 J2SE 5.0
Dino Klein [EMAIL PROTECTED] wrote on 06/06/2005 09:16:07 AM: I have two jspx files (pasted below), where one includes the other. When I remove the xmlns:c declaration/importation - it works; similarly, when I do not include the other file, it also works fine. snip/ (1) main.jspx == ?xml version=1.0 encoding=UTF-8? jsp:root xmlns:jsp=http://java.sun.com/JSP/Page; version=2.0 maintag xmlns:c=http://java.sun.com/jsp/jstl/core; jsp:directive.include file=one.jspx/ /maintag /jsp:root = (2) one.jspx = ?xml version=1.0 encoding=UTF-8? jsp:root xmlns:jsp=http://java.sun.com/JSP/Page; version=2.0 tagahello/taga /jsp:root = This is not a JSTL issue, marked OT. In XML views of JSPs, the included file is expanded in place. And jasper2 isn't happy because there are more than one jsp:root elements in the XML view of main.jspx. In your case, I don't see any need for either of the two jsp:root elements (unless you simplified before posting, and you have a legitimate need). Remove either, or both, and you should be fine. I think you've raised an interesting question. It seems jasper2 will not allow includes in the scenarios that result in more than one jsp:root elements in the expanded view. Please check if the spec says anything about this. If not, I would: 1) Ask for clarification on jsp spec public (unless Mark or Pierre listen to this list, and answer) 2) Post on tomcat-user, wait for feedback (follow with a bug report, depending on feedback) -Rahul
Re: Using JSTL 1.1.2 with Tomcat 5.5 J2SE 5.0
Dino Klein [EMAIL PROTECTED] wrote on 06/06/2005 03:44:38 PM: snip/ I did some primitive debugging, and all I could find out is that if I use xmlns:fn=http://java.sun.com/jsp/jstl/functions;, then I have no problems; I checked the TLDs, and I noticed that it does not include a validator like the rest (c, sql, fmt, x - all do, and cause problems). I tried this, you are right. It turns out the bit in my earlier email that talked about multiple jsp:root's in the expanded view was FUD, my sincere apologies. I've removed the OT marker. Unless someone else has an explanation, you can: 1) Dig into the JSTL TLV impls 2) File a bug report (if you're sufficiently convinced) The two options are not mutually exclusive ;-) -Rahul
Re: Upgrade from Tomcat 4 to 5
Nic - I recommend downloading the JSTL 1.1 Spec Maintenance Release from here [ http://java.sun.com/products/jsp/jstl/ ] and reading Appendix A (Compatibility Migration), specifically A.2 deals which JSTL 1.0 - 1.1 migration. Should have everything you need to know. -Rahul
Re: c:out problem
See Question 1 on Taglibs Wiki FAQ [
http://wiki.apache.org/jakarta-taglibs/FrequentlyAskedQuestions ]
-Rahul
Andy [EMAIL PROTECTED] wrote on 05/31/2005 09:29:37 AM:
I have another problem ! (what is going on today?) Why can't I print out
the
value of a variable using c:out ? The below example was copied from
http://javaalmanac.com/egs/javax.servlet.jsp.jstl.core/attr.html
c:set var=name1 value=value1 scope=page /
c:set var=com_mycompany_name2 value=value2 scope=request /
c:set var=com_mycompany_name3 value=value3 scope=session /
c:set var=com_mycompany_name4 value=value4 scope=application
/
%-- Show the saved values --%
c:out value='${pageScope.name1}' / br
c:out value='${requestScope.com_mycompany_name2}' / br
c:out value='${sessionScope.com_mycompany_name3}' / br
c:out value='${applicationScope.com_mycompany_name4}' / br
All I get is -
${pageScope.name1}
${requestScope.com_mycompany_name2}
${sessionScope.com_mycompany_name3}
${applicationScope.com_mycompany_name4}
Thanks,
Andy.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
RE: c:choose not evaluating correctly
A jsp:useBean declaration, that changes things ;-)
On 5/31/05, Andy [EMAIL PROTECTED] wrote:
snip/
I can see lots of horrible conditional glue logic in my future.
No, really, atleast not because of whats discussed in this thread.
I would:
1) Set the session attribute in the Struts layer only if the
username/password were valid (thats what you're doing)
2) Not have a jsp:useBean declaration in the JSP (so if the
username/password is not valid, a bean isn't instantiated)
3) Use the empty operator in EL expressions (on the 'userSession' bean
itself, not any property)
4) Change this bit:
Username:
jsp:useBean id=userSession scope=session
class=template.filter.UserSession/
jsp:getProperty name=userSession property=username/jsp:getProperty
(I suspect the userSession constructor initializes username to a non-null
value? Otherwise, you'd see a NPE here?)
to this:
c:if test=${not empty sessionScope.userSession}
Username: ${sessionScope.userSession.username}
/c:if
-Rahul
Re: join and arrays
Lookup the varStatus attribute of x:forEach, it will allow you to query
for boundary cases (first/last) amongst other things and you can use that
to decide when the trailing delimiter is not needed (for the last
iteration of the forEach).
-Rahul
On 5/26/05, Kralidis,Tom [Burlington] [EMAIL PROTECTED] wrote:
Hi,
I have a JSP which parses an XML document and inserts records into a DB.
For multiple XML elements, I do the following:
sql:transaction dataSource=${ds_devgeodb}
sql:update
insert into service_endpoints (keywords) values
('x:forEach
select=$doc/WMT_MS_Capabilities/Service/KeywordList/Keyword
var=kwx:out select=text()/,/x:forEach')
/sql:update
/sql:transaction
Which works okay, except that the values in the DB end up as:
keyword1,keyword2,keyword3,
..whereas I would prefer
keyword1,keyword2,keyword3
(i.e. without the trailing comma)
I checked the fn:join() function, which looks like it would do the
trick, but how does one set an array in JSTL?
Any advice would be appreciated.
Thanks
..Tom
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: counter in forEach tag?
varStatus is good here too, it can give you the count of the current
iteration. With that, for example, you can do a modulo 2 operation for
alternating bgcolors or style classes.
-Rahul
Scott Purcell [EMAIL PROTECTED] wrote on 05/26/2005 12:58:20 PM:
Hello,
I have this scenario come up occasionally, and have to opt to JSP
expressions on the page for a work-around.
c:choose
c:when test=${assetCollection == null}
nothing
/c:when
c:otherwise
c:forEach var=asset items=${assetCollection}
BODY
c:out value=${asset.assetId} /
/c:forEach
/c:otherwise
/c:choose
In the forEach section, I am going to create a rows in a table which
will be different colors for each row. How can I put in a counter
that I can then change the row color for? Basically, I would like to
know what asset I am on in the body.
Thanks,
Scott
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: Getting a substring
Jakarta Taglibs has a JSTL implementation here: http://jakarta.apache.org/taglibs/doc/standard-doc/intro.html Please look at the EL functions library, specifically the substring function. -Rahul On 5/10/05, Darryl Pierce [EMAIL PROTECTED] wrote: How can one get the substring from a string value in JSTL? -- Darryl L. Pierce [EMAIL PROTECTED] Visit the Infobahn Offramp: http://mcpierce.mypage.org Bury me next to my wife. Nothing too fancy... - Ulysses S. Grant - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: JSTL Recursion possible??
I, for one, did not understand the question (or what the nesting you show has to do with JSTL recursion). Can you please try again? Thanks, -Rahul On 5/2/05, Gros, Martin [EMAIL PROTECTED] wrote: JSTL and recursion, is this possible? e.g. walk an xml tree with same element names on different levels? a b/ b/ c b/ /c c b b/ /b /c b /a i want an output like b b --b --b b b is this possible? if yes, could you provide a short example? thanks Martin
[ANN] Recent changes to the RDC sandbox taglib (was Re: [ANN] First beta release of the RDC sandbox taglib)
A clarification is due, the recent changes do not constitute a release of the RDC sandbox taglib, please view Martin's note here for details: http://marc.theaimsgroup.com/?l=taglibs-devm=111440395623056w=2 -Rahul On 4/24/05, Rahul P Akolkar [EMAIL PROTECTED] wrote: The RDC sandbox taglib [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/intro.html ] has a new features beta 1.0 incorrectrelease/incorrect correctionversion/correction. Major Changes: 1) Defining Custom Event tuples - It is possible to leverage VoiceXML support for user defined event (name, grammar, handler) tuples. These tuples can be conveniently added to the RDC configuration files. 2) Dialog Context Switching - A recipe for switching dialog contexts using the RDC framework i.e. having the ability to leave a task (dialog) unfinished, initiate and complete another task and then come back to the earlier task, has been demonstrated. 3) Multi-channel amazon application - This sample application uses the Amazon Web Services to provide users with a Small Device GUI -or- Voice User Interface for browsing and buying music albums [ sample applications are available in rdc-examples.war ] 4) More documentation in rdc-doc.war [also available online at: http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/index.html ] New RDC Tag: 1) rdc:pause - Pauses until the user resumes the application. Modified RDC Tag: 1) rdc:struts-submit - Added the ability to selectively clear some or all of the RDC session state. Sandbox source is here [ http://cvs.apache.org/builds/jakarta-taglibs-sandbox/nightly/src/ ] RDC binaries are here [ http://cvs.apache.org/builds/jakarta-taglibs-sandbox/nightly/projects/rdc/ ] RDC revision history is available here [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/changes.html ] -Rahul Akolkar
[ANN] First beta release of the RDC sandbox taglib
The RDC sandbox taglib [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/intro.html ] has a new features beta 1.0 release. Major Changes: 1) Defining Custom Event tuples - It is possible to leverage VoiceXML support for user defined event (name, grammar, handler) tuples. These tuples can be conveniently added to the RDC configuration files. 2) Dialog Context Switching - A recipe for switching dialog contexts using the RDC framework i.e. having the ability to leave a task (dialog) unfinished, initiate and complete another task and then come back to the earlier task, has been demonstrated. 3) Multi-channel amazon application - This sample application uses the Amazon Web Services to provide users with a Small Device GUI -or- Voice User Interface for browsing and buying music albums [ sample applications are available in rdc-examples.war ] 4) More documentation in rdc-doc.war [also available online at: http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/index.html ] New RDC Tag: 1) rdc:pause - Pauses until the user resumes the application. Modified RDC Tag: 1) rdc:struts-submit - Added the ability to selectively clear some or all of the RDC session state. Sandbox source is here [ http://cvs.apache.org/builds/jakarta-taglibs-sandbox/nightly/src/ ] RDC binaries are here [ http://cvs.apache.org/builds/jakarta-taglibs-sandbox/nightly/projects/rdc/ ] RDC revision history is available here [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/changes.html ] -Rahul Akolkar
Re: c:when problem
c:when test=${nc.Response=='Exchange Failure' !empty nc.readyDate}
On 4/21/05, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
Hi There,
i have a strange problem when i am using a when clause with multiple
conditions, ie
c:choose
c:when test=${nc.Response=='Exchange Failure'} ${!empty
nc.readyDate}
Planned
/c:when
c:when test=${nc.Response!='SUCCESS'}
${nc.Response}
/c:when
c:otherwise
Enabled
/c:otherwise
/c:choose
for my example I know that Response = 'Exchange Failure' and that there
is a
readyDate. However the condition never evaluates to true. ie i would
expect the word planned to be printed, but it isn't. Is there something
i've over
looked here?
thx in advance pg
Re: packing images and style sheets with a taglib
I have a taglib that I am packaging up as a jar and I'm trying to figure out how I can package it with images and style sheets that could be referenced in a relative path like img src=/mytaglibimages/car.gif in the taglib code. Not sure if you can get relative paths to work with no overhead to the taglib user. One possibility is to serve the resources using a servlet that maps to the url pattern /mytaglibimages/* (or some such). This allows you to pack the resources and the servlet in your distro, but the user needs to add the servlet definition/mapping to the deployment descriptor (which, arguably, might be less error-prone, less of an overhead and easier to document). -Rahul P.S.- If you choose to put the servlet in, package java.util.jar provides the necessary classes.
Re: packing images and style sheets with a taglib
It also means you have to deal with cache headers, so that you're not serving up the same images to the same browser again and again. Very true, cache, response-type, in short, you have to behave like a good web server ;-) However, its definitely manageable, and by pulling some of the overhead upon the taglib author, it aims for taglib user convenience. OTOH, the vanilla solution seems to be an auxillary resource bundle (not in the i18n sense) users need to download and plop into each app/war. So. -Rahul
Re: Retrieve a session attribute
I thought a JSP would share the same session within the servlet if a session already exists in the container. JSP is eventually translated into servlet code, doesn't it? I insert the log message code in the servlet before getting to the JSP and find that a session already exists at the point. If a session in JSP is unrelated with a session in servlet, how these two share session attributes? Again, the page directive specifies whether the JSP joins a session (if one is already associated with the current request -- as in your case). By specifying the attribute value false, you're saying you do not want to join the session, and hence, waiving your ability to access anything within the session. Please read the related bit here: http://java.sun.com/products/jsp/tags/syntaxref.fm7.html -Rahul
Re: c:forEach problem
c:forEach var=directory items=CONTEXT_STRUCTURE.directories EL expression for items? String argument will be treated as a CSV. -Rahul
Re: Odd behavior of Number/String conversion??
Rick wrote: Thanks Rahul for the info. I'm wondering which would have less overhead behind the scenes... using the replace like you have it above or storingwhat I want to replace in a temp var? The reason I ask is this operation will have to actually be repeated several times within an outer loop. For your example, it doesn't matter. First off, performance discussions will be container specific. In the reference implementation (tomcat/jasper), when o.a.j.compiler.Generator gets you the Servlet class to compile, you'll find a bunch of PageContextImpl.proprietaryEvaluate() calls in the code, which are in turn using commons.el to do the actual EL evaluation. An rtexpr true tag attribute will be sent to commons.el expression evaluator impl as-is, a standalone EL expression is broken down in an ordered list of ELNode.Text, ELNode.Root, ELNode.Function nodes (broken down once per JSP mod) so only the true expression snippets make it to EL evaluation. There are a bunch of StringBuffer manipulations along most of these steps, fn:replace() has its own StringBuffer manipulations, and in the larger scheme of things, for the example we're discussing, the difference in overheads is definitely nothing to lose sleep over ;-) Take your pick. If you want to pursue this, try a simple (t2-t1) timing study on the two options. I'd be very surprised if either option comes out a clear winner. -Rahul
Re: Odd behavior of Number/String conversion??
trying to convert String field. to type java.lang.Doublesnip
I even tried using the fn:trim tag which is supposed to return a String
In short, the spec says so :-)
Remember, + is not string concatenation in JSPs, it is an arthmetic
operator. The rogue argument here is not the index (so trimming the index
won't help), its the String field., and the dot triggers the String to
Double coercion (again, as per the spec).
there must be a way to display this in one expression?
c:set var='theField' value=${metaProps[fn:replace('field.*.type', '*',
fieldNum)]}/
Conceptually, a fn:join is a better fit here, but I'm not too keen on
creating arrays in JSPs ;-) Finally, if temporary variables is what you're
after, 'theField' would be the next in line, although readability has its
own benefits.
-Rahul
Re: Taglib / context-param question
If I put the hex values of the colors in the context-param area of the web.xml file, how do you call them? pageContext. getServletContext().getInitParameter(insert-context-param-name-here); -Rahul
Changes to the RDC sandbox taglib
The RDC sandbox taglib [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/intro.html ] has a new features pre beta release. There have been a multitude of code enhancements. Major Changes: 1) Rule based directed dialog strategy for rule based dialog management across child RDCs using the rdc:group container. Illustrates the benefits of pluggable dialog management strategies at the group level. [ Example: rulebased-conditions.jsp in rdc-examples.war ] 2) RDC template, a mechanism for rapid prototyping of components in the RDC framework. [ Example: simple-template-test.jsp in rdc-examples.war ] 3) Better formalism for authoring components 4) More documentation in rdc-doc.war [also available online at: http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/index.html ] New RDC Tags: 1) rdc:struts-errors - Plays out any errors messages from the preceding struts action as VoiceXML prompts 2) rdc:template - As mentioned above 3) rdc:number - RDC used to collect a number Sandbox source is here [ http://cvs.apache.org/builds/jakarta-taglibs-sandbox/nightly/src/ ] RDC binaries are here [ http://cvs.apache.org/builds/jakarta-taglibs-sandbox/nightly/projects/rdc/ ] RDC revision history is available here [ http://jakarta.apache.org/taglibs/sandbox/doc/rdc-doc/changes.html ] -Rahul Akolkar
Re: trying to create a Html table from a Taglib
The the tag lib will receive two parameters: the list off skills and the number of columns. snip Can I pass a collection, a list, to a tag lib? how? The c:forEach tag [JSTL core] will allow you to iterate over items in a collection, you may add the tr and td elements appropriately before and after each item [the varStatus attribute of the c:forEach tag will allow you to take care of tracking the item count for the number of columns bit and the special case for the last item]. You can also do the same in a tag impl and create your own JSP tag. There are some HTML table generating tags already out there, so you might want to look at what those offer first. -Rahul
Re: Struts integration
well, how else would I know if that's just a string I need to send verbatim to the client, or if I need to retieve the actual URL another way? of course, the resul will be an href attribute sent to the client anyway. You're tempting me to reply, so let me blame you for this email ;-) We can continue this offline (w.r.t this list) if needed. Its always going to be a string sent verbatim to the client, irrespective of struts/no struts. The struts machinery comes into play once the client follows the href on the anchor (think of it as a servlet mapping that you as the taglib author don't have to worry about). are you saying that I don't need to import any Struts specific Jars? Probably not, if you can rely on HTTP name-value pairs to submit the form data to go along with. The part that is most unclear to me in this discussion is what your anchor tag will submit to struts. It makes sense when the tag submits some data to support operations on the application model that the struts controller will use to determine the next view. -Rahul
Re: Mixing JSTL with scriptlet and problematic evaluation order
c:url var=encodedUrl
value=Blah.do?sortId=${sortId}...title=${titleText}.../
onclick=javascript:window.open('${encodedUrl}', ...)
Because I couldn't find a nice JSTL solution, I tried it with a mixture
of JSTL and the scriptlet:
...title=' + '%= URLEncoder.encode(c:out value='${titleText}'/,
UTF-8) %'
Guess what this results in?
Well, the scriptlet part must be getting interpreted first, so the
URLEncoder.encode gets that c:out value... string literal as input,
instead of getting the output of c:out. So that doesn't work either.
Here is a bit more of the JavaScript call in question:
onclick=javascript:window.open('Blah.do?sortId=c:out
value=${sortId}/..title=' + '%= URLEncoder.encode(c:out
value='${titleText}'/, UTF-8) %', 'Foobar'
jsp:include and internet explorer
Return Receipt Your jsp:include and internet explorer document : was Rahul P Akolkar/Watson/IBM received by: at: 11/30/2004 10:51:25 - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: JSTL converts ' to #039;
Don't escape XML in the JavaScript bit.
%@ taglib prefix=c uri=http://java.sun.com/jsp/jstl/core; %
c:set var=foobar value=Foo's Moos /
html
headscript language=JavaScriptalert(c:out value='${foobar}'
escapeXml='false' /);/script/head
body bgcolor=white
c:out value=${foobar} /
/body
/html
-Rahul
Re: [INPUT] Is input:text .../ producing the correct HTML?
Is this a non-impacting bug? This will be helpful (W3C HTML and XHTML FAQ): http://www.w3.org/MarkUp/2004/xhtml-faq -Rahul
RE: Error: org.xml.sax.SAXParseException: Content is not allowed in prolog.
c:import var=nieuwsXML url=nieuws.xml
x:parse doc=${nieuwsXML} var=nieuwsDOC /
/c:import
-Rahul
Re: How to avoid the line breaks when a tag file used within javascript?
This thread has gotten too long to ignore :-)
trimSpaces probably won't do it for you.
As my bit, I tried this trivial JSP:
%@ taglib prefix=c uri=http://java.sun.com/jsp/jstl/core; %%--
--%tag attr=foo('jsp:include page=test.jsp /')%--
--%anotherTag /
where test.jsp is another trivial JSP as follows:
%@ taglib prefix=c uri=http://java.sun.com/jsp/jstl/core; %%--
--%A, c:if test=${true}%--
--%B, %--
--%/c:ifC
I don't have a newline after /c:ifC, and I get this output:
tag attr=foo('A, B, C')anotherTag /
Ugly code? Yes. But I believe the output is similar to what you want from
your example.
Now why it doesn't work for you is another issue. I'm using Tomcat and
Emacs.
-Rahul
Re: VoiceXML RDC with return element
Looks like you are all set. The submit URI defaults to the JSP page hosting the RDC tag. Sidebar: Think of each RDC as a conversation with the user, with potentially multiple round trips. As an example, rdc:date id=myDate minDate=01012004 maxDate=01012005 confirm=true echo=true / might lead to this use case (prompts might differ, you can customize the prompts for your application using a custom XML config ... customization examples are available in the distribution): S: Please specify a date. U: *silence--timeout* S: I did not hear anything. Please specify a date. // client-side no-input event U: january second two thousand five S: Please specify an earlier date. // Round Trip 1: input did not meet contraints U: january second two thousand four S: I think you said january second two thousand four. Is that right? // Round Trip 2: input was good ... JSP author said confirm=true U: Yes S: OK, january second two thousand four. Got it. // Round Trip 3: confirmed value ... JSP author said echo=true Note that the same date RDC tag produces different VXML when it executes server-side depending on where it is in this conversation and what the user just said. Submitting to the same JSP page hosting the RDC is done underneath the hood (by defaulting submit). -Rahul Thank you, Rahul. You answered my question. :) Anyway, I think I'm having setup problem. The content returns from date-test.jsp has .. script src=/rdc-examples/.grammar/nbest.js/ var name=dateResultNBest expr=serializeNBest()/ submit next=/rdc-examples/date-test.jsp;jsessionid=4C96B24FB06E4408691A9BB 08E4E84F7 namelist=dateResultNBest/ /filled /field It shouldn't have it since I didn't use submit attribute in rdc:date, correct? Besides installing rdc-examples.war from jakarta-taglibs-sandbox-rdc-20040921.tar.gz, do I need to do anything else? Thanks again,
Re: VoiceXML RDC with return element
I don't understand what you mean by generate RDC with return element.
A RDC can be used in any VoiceXML form, and whether that form is
referenced as a subdialog elsewhere does not change the RDC behavior (and
hence, you shouldn't have to create a RDC that can be used in
subdialog).
The underlying principle that might help here is the fact that whenever a
RDC completes execution (and hence, has a value that is legal within the
constraints set via the tag attributes) its value appears as a PageContext
variable with the same name as the ID of the RDC. As an example, here is a
parent VXML page (which might as well be a JSP page):
?xml version=1.0 encoding=UTF-8 ?
vxml version=2.0 xml:lang=en-US xmlns=http://www.w3.org/2001/vxml;
form
subdialog name=subdia src=test.jsp#return_example
filled
promptValue is value
expr=subdia.dateReturned //prompt
/filled
/subdialog
/form
/vxml
Where test.jsp is like so (Grab date-test.jsp from the examples in the
distribution and replace the form element with the following):
form id=return_example
rdc:date id=myDate minDate=01012004 maxDate=01012005
confirm=true echo=true /
c:if test=${!(empty myDate)}
block
var name=dateReturned expr='${myDate}' /
return namelist=dateReturned /
/block
/c:if
/form
Note that you are doing nothing special because the RDC is within a form
referenced as a subdialog.
-Rahul
Soonthorn A wrote:
Hi,
Does anyone has an example that generate RDC with return element?
I just learn about jsp taglib today and looked at RDC's tag lib files
and config files. I couldn't find anything relate to return. Is
VoiceXML's return element in your current model?
Basically, what I tried to do is to create a RDC that can be used in
subdialog.
Thanks,
