HtmlUnit

HtmlUnit is not a maven 2 project. If there is a compatability problem with maven 1
then that's going to be a question for the maven team. If there's a change we can
make to HtmlUnit that will help then we can do that but since I don't even know what
jsr173 or JWSDP are I wouldn't even know where to start looking :)

Brad C

--- Ryan Lubke <Ryan.Lubke@...> wrote:

> Hello Folks,
> 
> I'm trying to use Antlib for Maven 2.0 [1] to obtain HtmlUnit 1.8
> for a project I'm working on.  However, during the dependency
> resolution process I receive the following:
> 
>     [artifact:dependencies] Unable to resolve artifact: required 
> artifacts missing:
>     [artifact:dependencies]   javax.xml:jsr173:jar:1.0
> 
> If I browse the pom for this dependency, I see a URL to JWSDP.  Is
> this right?  It seems overkill to download JWSDP just to use HtmlUnit
> (I certainly don't need to when I use the standalone distribute of 
> HtmlUnit).
> 
> Ideas?
> 
> Thanks,
> -rl
> 
> [1] http://maven.apache.org/ant-tasks.html
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live webcast
> and join the prime developer group breaking into this new coding territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>

Yes, I know what you mean.

The problem is the dependency on dom4j-1.5, which in turns declares to
depend on the artifact:
   javax.xml:jsr173:jar:1.0

as you can read from here:
http://www.ibiblio.org/maven2/dom4j/dom4j/1.5/dom4j-1.5.pom


The problem can be solved in two ways:
1) The htmlunit team changes its pom stating that htmlunit depends on
dom4j-1.6.1 which does *not* depend on the jsr173 thing and is indeed tha
latest stable version of dom4j (at least on ibiblio)

2) You "override" yourself the dom4j dependency in your project pom
declaring to depend explicitly on dom4j-1.6.1, in this way:

<dependency>
    <groupId>dom4j</groupId>
    <artifactId>dom4j</artifactId>
    <version>1.6.1</version>
</dependency>

Which is what I did in a project, which is maven2 and uses htmlunit.

Was it of any help?

Alessio Pace.


On 3/7/06, Ryan Lubke <Ryan.Lubke@...> wrote:
>
> Hello Folks,
>
> I'm trying to use Antlib for Maven 2.0 [1] to obtain HtmlUnit 1.8
> for a project I'm working on.  However, during the dependency
> resolution process I receive the following:
>
>     [artifact:dependencies] Unable to resolve artifact: required
> artifacts missing:
>     [artifact:dependencies]   javax.xml:jsr173:jar:1.0
>
> If I browse the pom for this dependency, I see a URL to JWSDP.  Is
> this right?  It seems overkill to download JWSDP just to use HtmlUnit
> (I certainly don't need to when I use the standalone distribute of
> HtmlUnit).
>
> Ideas?
>
> Thanks,
> -rl
>
> [1] http://maven.apache.org/ant-tasks.html
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting
> language
> that extends applications into web and mobile media. Attend the live
> webcast
> and join the prime developer group breaking into this new coding
> territory!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>



--
Alessio Pace.

I've upgraded dependency to dom4j 1.6.1 in CVS.

Thanks for the info Alessio.

Marc.

Alessio Pace wrote:
> Yes, I know what you mean.
> 
> The problem is the dependency on dom4j-1.5, which in turns declares to
> depend on the artifact:
>    javax.xml:jsr173:jar:1.0
> 
> as you can read from here:
> http://www.ibiblio.org/maven2/dom4j/dom4j/1.5/dom4j-1.5.pom
> 
> 
> The problem can be solved in two ways:
> 1) The htmlunit team changes its pom stating that htmlunit depends on
> dom4j-1.6.1 which does *not* depend on the jsr173 thing and is indeed
> tha latest stable version of dom4j (at least on ibiblio)
> 
> 2) You "override" yourself the dom4j dependency in your project pom
> declaring to depend explicitly on dom4j-1.6.1, in this way:
>  
> <dependency>
>     <groupId>dom4j</groupId>
>     <artifactId>dom4j</artifactId>
>     <version>1.6.1</version>
> </dependency>
> 
> Which is what I did in a project, which is maven2 and uses htmlunit.
> 
> Was it of any help?
> 
> Alessio Pace.
> 
> 
> On 3/7/06, *Ryan Lubke* <Ryan.Lubke@... <mailto:Ryan.Lubke@...>>
> wrote:
> 
>     Hello Folks,
> 
>     I'm trying to use Antlib for Maven 2.0 [1] to obtain HtmlUnit 1.8
>     for a project I'm working on.  However, during the dependency
>     resolution process I receive the following:
> 
>         [artifact:dependencies] Unable to resolve artifact: required
>     artifacts missing:
>         [artifact:dependencies]   javax.xml:jsr173:jar:1.0
> 
>     If I browse the pom for this dependency, I see a URL to JWSDP.  Is
>     this right?  It seems overkill to download JWSDP just to use HtmlUnit
>     (I certainly don't need to when I use the standalone distribute of
>     HtmlUnit).
> 
>     Ideas?
> 
>     Thanks,
>     -rl
> 
>     [1] http://maven.apache.org/ant-tasks.html
> 
> 
>     -------------------------------------------------------
>     This SF.Net email is sponsored by xPML, a groundbreaking scripting
>     language
>     that extends applications into web and mobile media. Attend the live
>     webcast
>     and join the prime developer group breaking into this new coding
>     territory!
>     http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
>     <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642>
>     _______________________________________________
>     Htmlunit-user mailing list
>     Htmlunit-user@...
>     <mailto:Htmlunit-user@...>
>     https://lists.sourceforge.net/lists/listinfo/htmlunit-user
> 
> 
> 
> 
> -- 
> Alessio Pace.

Thanks Marc and Alessio. 

When do you expect this to be pushed to ibilio?

Thanks,

-rl

Marc Guillemot wrote:
> I've upgraded dependency to dom4j 1.6.1 in CVS.
>
> Thanks for the info Alessio.
>
> Marc.
>
> Alessio Pace wrote:
>   
>> Yes, I know what you mean.
>>
>> The problem is the dependency on dom4j-1.5, which in turns declares to
>> depend on the artifact:
>>    javax.xml:jsr173:jar:1.0
>>
>> as you can read from here:
>> http://www.ibiblio.org/maven2/dom4j/dom4j/1.5/dom4j-1.5.pom
>>
>>
>> The problem can be solved in two ways:
>> 1) The htmlunit team changes its pom stating that htmlunit depends on
>> dom4j-1.6.1 which does *not* depend on the jsr173 thing and is indeed
>> tha latest stable version of dom4j (at least on ibiblio)
>>
>> 2) You "override" yourself the dom4j dependency in your project pom
>> declaring to depend explicitly on dom4j-1.6.1, in this way:
>>  
>> <dependency>
>>     <groupId>dom4j</groupId>
>>     <artifactId>dom4j</artifactId>
>>     <version>1.6.1</version>
>> </dependency>
>>
>> Which is what I did in a project, which is maven2 and uses htmlunit.
>>
>> Was it of any help?
>>
>> Alessio Pace.
>>
>>
>> On 3/7/06, *Ryan Lubke* <Ryan.Lubke@... <mailto:Ryan.Lubke@...>>
>> wrote:
>>
>>     Hello Folks,
>>
>>     I'm trying to use Antlib for Maven 2.0 [1] to obtain HtmlUnit 1.8
>>     for a project I'm working on.  However, during the dependency
>>     resolution process I receive the following:
>>
>>         [artifact:dependencies] Unable to resolve artifact: required
>>     artifacts missing:
>>         [artifact:dependencies]   javax.xml:jsr173:jar:1.0
>>
>>     If I browse the pom for this dependency, I see a URL to JWSDP.  Is
>>     this right?  It seems overkill to download JWSDP just to use HtmlUnit
>>     (I certainly don't need to when I use the standalone distribute of
>>     HtmlUnit).
>>
>>     Ideas?
>>
>>     Thanks,
>>     -rl
>>
>>     [1] http://maven.apache.org/ant-tasks.html
>>
>>
>>     -------------------------------------------------------
>>     This SF.Net email is sponsored by xPML, a groundbreaking scripting
>>     language
>>     that extends applications into web and mobile media. Attend the live
>>     webcast
>>     and join the prime developer group breaking into this new coding
>>     territory!
>>     http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
>>     <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642>
>>     _______________________________________________
>>     Htmlunit-user mailing list
>>     Htmlunit-user@...
>>     <mailto:Htmlunit-user@...>
>>     https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>>
>>
>>
>>
>> -- 
>> Alessio Pace.
>>     
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live webcast
> and join the prime developer group breaking into this new coding territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>

Hi all,
In our javascript code there is a line that is doing :
fld.onchange ="";
I get a script error...

I try to put
fld.onchange= null;
and I got no error.

Anyone know why ?
Is it a problem with HtmlUnit ?
I dont know where to look if that kind of assignation is valid or not 
with Javascript.
Let me know if I need to enter a bugs!

thanks
anthony

Looks like the set for onchange only allows function references right now, not
strings of javascript, which would be a HtmlUnit problem. Logging a bug (if it's not
there already) would be appropriate, and including a failing unit test or some
examples of javascript that works in a browser would be ideal.

Brad C


--- Anthony Laquerre <anthonyl@...> wrote:

> Hi all,
> In our javascript code there is a line that is doing :
> fld.onchange ="";
> I get a script error...
> 
> I try to put
> fld.onchange= null;
> and I got no error.
> 
> Anyone know why ?
> Is it a problem with HtmlUnit ?
> I dont know where to look if that kind of assignation is valid or not 
> with Javascript.
> Let me know if I need to enter a bugs!
> 
> thanks
> anthony
>

Hi Marc,

I can give you maybe some more infos on htmlunit dependencies. The jsr173
thing was a nasty (transitive) "legacy" dependency (maybe an htmtunit-1.8(.=
0).1
could be nice to depend on, removing also saxpath?), but there are others I
fairly think that can be upgraded (some are just libraries updates):

* commons-lang: commons-lang from 2.0 to 2.1

* xerces: xercesImpl from 2.6.2 to 2.7.1  This is a little trickier because=
,
fro what I saw in practise, xerces 2.7.0 implements the new DOM3 APIs of
JDK5.0, and if you have your own code that extends DOM APis and want to
compile it under JDK5.0 (which added some methods to the DOM APIs indeed),
you do require xerces 2.7.0, otherwise you get problems with methods not
being implemented.

This last point is just something that can be put into account in case
htmlunit aims to implements DOM APIs some day, but it is could be a problem
for people who depend on htmlunit while using Maven2 and JDK5.0 on a projec=
t
which extends DOM things (i.e. MyCustomeDOMNode), and so, because of
htmlunit, get a dependency on xerces-2.6.x which is not in pair with the DO=
M
APIs.  (of course the solution is to explicitly declare to depend on the
newer version)

Mmm sorry if my explanation wasn't that clear, but the JDK5.0 XML thing
added some complexities in that kind of circumstance from my experience..

regards,
Alessio Pace.


On 3/10/06, Ryan Lubke <Ryan.Lubke@...> wrote:
>
> Thanks Marc and Alessio.
>
> When do you expect this to be pushed to ibilio?
>
> Thanks,
>
> -rl
>
> Marc Guillemot wrote:
> > I've upgraded dependency to dom4j 1.6.1 in CVS.
> >
> > Thanks for the info Alessio.
> >
> > Marc.
> >
> > Alessio Pace wrote:
> >
> >> Yes, I know what you mean.
> >>
> >> The problem is the dependency on dom4j-1.5, which in turns declares to
> >> depend on the artifact:
> >>    javax.xml:jsr173:jar:1.0
> >>
> >> as you can read from here:
> >> http://www.ibiblio.org/maven2/dom4j/dom4j/1.5/dom4j-1.5.pom
> >>
> >>
> >> The problem can be solved in two ways:
> >> 1) The htmlunit team changes its pom stating that htmlunit depends on
> >> dom4j-1.6.1 which does *not* depend on the jsr173 thing and is indeed
> >> tha latest stable version of dom4j (at least on ibiblio)
> >>
> >> 2) You "override" yourself the dom4j dependency in your project pom
> >> declaring to depend explicitly on dom4j-1.6.1, in this way:
> >>
> >> <dependency>
> >>     <groupId>dom4j</groupId>
> >>     <artifactId>dom4j</artifactId>
> >>     <version>1.6.1</version>
> >> </dependency>
> >>
> >> Which is what I did in a project, which is maven2 and uses htmlunit.
> >>
> >> Was it of any help?
> >>
> >> Alessio Pace.
> >>
> >>
> >> On 3/7/06, *Ryan Lubke* <Ryan.Lubke@... <mailto:Ryan.Lubke@...
> >>
> >> wrote:
> >>
> >>     Hello Folks,
> >>
> >>     I'm trying to use Antlib for Maven 2.0 [1] to obtain HtmlUnit 1.8
> >>     for a project I'm working on.  However, during the dependency
> >>     resolution process I receive the following:
> >>
> >>         [artifact:dependencies] Unable to resolve artifact: required
> >>     artifacts missing:
> >>         [artifact:dependencies]   javax.xml:jsr173:jar:1.0
> >>
> >>     If I browse the pom for this dependency, I see a URL to JWSDP.  Is
> >>     this right?  It seems overkill to download JWSDP just to use
> HtmlUnit
> >>     (I certainly don't need to when I use the standalone distribute of
> >>     HtmlUnit).
> >>
> >>     Ideas?
> >>
> >>     Thanks,
> >>     -rl
> >>
> >>     [1] http://maven.apache.org/ant-tasks.html
> >>
> >>
> >>     -------------------------------------------------------
> >>     This SF.Net email is sponsored by xPML, a groundbreaking scripting
> >>     language
> >>     that extends applications into web and mobile media. Attend the
> live
> >>     webcast
> >>     and join the prime developer group breaking into this new coding
> >>     territory!
> >>
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
> >>     <
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642>
> >>     _______________________________________________
> >>     Htmlunit-user mailing list
> >>     Htmlunit-user@...
> >>     <mailto:Htmlunit-user@...>
> >>     https://lists.sourceforge.net/lists/listinfo/htmlunit-user
> >>
> >>
> >>
> >>
> >> --
> >> Alessio Pace.
> >>
> >
> >
> >
> > -------------------------------------------------------
> > This SF.Net email is sponsored by xPML, a groundbreaking scripting
> language
> > that extends applications into web and mobile media. Attend the live
> webcast
> > and join the prime developer group breaking into this new coding
> territory!
> > http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
> > _______________________________________________
> > Htmlunit-user mailing list
> > Htmlunit-user@...
> > https://lists.sourceforge.net/lists/listinfo/htmlunit-user
> >
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting
> language
> that extends applications into web and mobile media. Attend the live
> webcast
> and join the prime developer group breaking into this new coding
> territory!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>

Brad Clarke <yourgod <at> bradclarke.com> writes:

> 
> Looks like the set for onchange only allows function references right now, not
> strings of javascript, which would be a HtmlUnit problem. Logging a bug (if
it's not
> there already) would be appropriate, and including a failing unit test or some
> examples of javascript that works in a browser would be ideal.

in fact handlers should accept anything and gets executed only if they are a
function (if you assign a string, nothing will be executed). 
Should not be very complicated to implement but requires unit tests ;-)

Marc.

Alessio Pace <alessio.pace <at> gmail.com> writes:
> 
> 
> Hi Marc,
> I can give you maybe some more infos on htmlunit dependencies. The
> jsr173 thing was a nasty (transitive) "legacy" dependency (maybe an
> htmtunit-1.8(.0).1 could be nice to depend on, removing also saxpath?),
> but there are others I fairly think that can be upgraded (some are just
> libraries updates):
> * commons-lang: commons-lang from 2.0 to 2.1 

is already done in CVS since a few days

> * xerces: xercesImpl from 2.6.2 to 2.7.1  This is a little
> trickier because, fro what I saw in practise, xerces 2.7.0 implements
> the new DOM3 APIs of JDK5.0, and if you have your own code that extends
> DOM APis and want to compile it under JDK5.0 (which added some methods
> to the DOM APIs indeed), you do require xerces 2.7.0, otherwise you get
> problems with methods not being implemented. 
> This last point is just something that can be put into account in case
> htmlunit aims to implements DOM APIs some day, but it is could be a
> problem for people who depend on htmlunit while using Maven2 and JDK5.0
> on a project which extends DOM things (i.e. MyCustomeDOMNode), and so,
> because of htmlunit, get a dependency on xerces-2.6.x which is not in
> pair with the DOM APIs.  (of course the solution is to explicitly
> declare to depend on the newer version)
> Mmm sorry if my explanation wasn't that clear, but the JDK5.0 XML thing
> added some complexities in that kind of circumstance from my
> experience..

interesting. Why 2.71 and not not the recently released 2.8?
Brad said for a few days that it may requires some changes. I haven't looked
at it but I think that it would be interesting, particularly due to the
performance improvements. I've already mailed Andy Clark (the author of
NekoHtml) to knwo if he plans to upgrade the parser too.

Marc.
--
View this message in context: http://www.nabble.com/Obtaining-HtmlUnit-1.8-via-Maven-t1237089.html#a3338801
Sent from the HtmlUnit - General forum at Nabble.com.

On 3/10/06, Marc Guillemot <mguillemot@...> wrote:
>
>
> Alessio Pace <alessio.pace <at> gmail.com> writes:
> >
> >
> > Hi Marc,
> > I can give you maybe some more infos on htmlunit dependencies. The
> > jsr173 thing was a nasty (transitive) "legacy" dependency (maybe an
> > htmtunit-1.8(.0).1 could be nice to depend on, removing also saxpath?),
> > but there are others I fairly think that can be upgraded (some are just
> > libraries updates):
> > * commons-lang: commons-lang from 2.0 to 2.1
>
> is already done in CVS since a few days



sorry, I saw the dependency that was declared in the 1.8 revision, not the
cvs

> * xerces: xercesImpl from 2.6.2 to 2.7.1  This is a little
> > trickier because, fro what I saw in practise, xerces 2.7.0 implements
> > the new DOM3 APIs of JDK5.0, and if you have your own code that extends
> > DOM APis and want to compile it under JDK5.0 (which added some methods
> > to the DOM APIs indeed), you do require xerces 2.7.0, otherwise you get
> > problems with methods not being implemented.
> > This last point is just something that can be put into account in case
> > htmlunit aims to implements DOM APIs some day, but it is could be a
> > problem for people who depend on htmlunit while using Maven2 and JDK5.0
> > on a project which extends DOM things (i.e. MyCustomeDOMNode), and so,
> > because of htmlunit, get a dependency on xerces-2.6.x which is not in
> > pair with the DOM APIs.  (of course the solution is to explicitly
> > declare to depend on the newer version)
> > Mmm sorry if my explanation wasn't that clear, but the JDK5.0 XML thing
> > added some complexities in that kind of circumstance from my
> > experience..
>
> interesting. Why 2.71 and not not the recently released 2.8?


because the 2.7.1 is the latest on ibiblio :)
I guess someone should push the xerces team to release the 2.8 on ibiblio


Brad said for a few days that it may requires some changes. I haven't looke=
d
> at it but I think that it would be interesting, particularly due to the
> performance improvements. I've already mailed Andy Clark (the author of
> NekoHtml) to knwo if he plans to upgrade the parser too.



Yes, nekohtml indeed depends on the old versions:

  <dependencies>
    <dependency>
      <groupId>xerces</groupId>
      <artifactId>xerces</artifactId>
      <version>2.4.0</version>
  </dependency>

Which is fairly old and forces maven2 projects (of the kind I
explained before, like mine..)
to exclude explicitly that transitive dependency (weird enough
xerces:xerces has been
replaced -I guess,looking at version numbers- by xerces:xercesImpl so
you have a
maven2 runtime classpath messed up!).

That is a problem related to all libraries which depend on xerces:xerces
(also commons-jelly:commons-jelly that is required by htmltunit), and
that should be
updated to depend on xerces:xercesImpl.


Alessio Pace.



Marc.
> --
> View this message in context:
> http://www.nabble.com/Obtaining-HtmlUnit-1.8-via-Maven-t1237089.html#a333=
8801
> Sent from the HtmlUnit - General forum at Nabble.com.
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting
> language
> that extends applications into web and mobile media. Attend the live
> webcast
> and join the prime developer group breaking into this new coding
> territory!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>



--
Alessio Pace.

Okay,
I will try to get involve. but I am not sure how to fix it .
It's the first time I jump into the source code :-)

I will see if I can do something!
thanks
anthony


> Brad Clarke <yourgod <at> bradclarke.com> writes:
>
>   
>> Looks like the set for onchange only allows function references right now, not
>> strings of javascript, which would be a HtmlUnit problem. Logging a bug (if
>>     
> it's not
>   
>> there already) would be appropriate, and including a failing unit test or some
>> examples of javascript that works in a browser would be ideal.
>>     
>
> in fact handlers should accept anything and gets executed only if they are a
> function (if you assign a string, nothing will be executed). 
> Should not be very complicated to implement but requires unit tests ;-)
>
> Marc.
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live webcast
> and join the prime developer group breaking into this new coding territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>

well bad news,
I understand the problem, but I dont know how to fully implement it.

Heres the problems :
The problem is the assignation of the field.
fld.onchange = "";

js
>
> anthony
>
>
>> Brad Clarke <yourgod <at> bradclarke.com> writes:
>>
>>  
>>> Looks like the set for onchange only allows function references 
>>> right now, not
>>> strings of javascript, which would be a HtmlUnit problem. Logging a 
>>> bug (if
>>>     
>> it's not
>>  
>>> there already) would be appropriate, and including a failing unit 
>>> test or some
>>> examples of javascript that works in a browser would be ideal.
>>>     
>>
>> in fact handlers should accept anything and gets executed only if 
>> they are a
>> function (if you assign a string, nothing will be executed). Should 
>> not be very complicated to implement but requires unit tests ;-)
>>
>> Marc.
>>
>>
>>
>> -------------------------------------------------------
>> This SF.Net email is sponsored by xPML, a groundbreaking scripting 
>> language
>> that extends applications into web and mobile media. Attend the live 
>> webcast
>> and join the prime developer group breaking into this new coding 
>> territory!
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
>> _______________________________________________
>> Htmlunit-user mailing list
>> Htmlunit-user@...
>> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>>   
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting 
> language
> that extends applications into web and mobile media. Attend the live 
> webcast
> and join the prime developer group breaking into this new coding 
> territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user

oups sorry I send too fast ;-)

Heres the problems :
The problem is the assignation of the field.
fld.onchange = "";
the call is :
setterMethod.invoke(a1,a2)
a2 is a string but it should be Function.
and it throw an exception.


How to fix it now.
1- Verify that all args parameter are of the good class type. 
(myParameter.isInstance(a2))
if everything is correct then invoke the methode.
if not, what should we do ???
We dont want to redefined all function with a String as parameter.

any help ?

anthony

> Okay,
> I will try to get involve. but I am not sure how to fix it .
> It's the first time I jump into the source code :-)
>
> I will see if I can do something!
> thanks
> anthony
>
>
>> Brad Clarke <yourgod <at> bradclarke.com> writes:
>>
>>  
>>> Looks like the set for onchange only allows function references 
>>> right now, not
>>> strings of javascript, which would be a HtmlUnit problem. Logging a 
>>> bug (if
>>>     
>> it's not
>>  
>>> there already) would be appropriate, and including a failing unit 
>>> test or some
>>> examples of javascript that works in a browser would be ideal.
>>>     
>>
>> in fact handlers should accept anything and gets executed only if 
>> they are a
>> function (if you assign a string, nothing will be executed). Should 
>> not be very complicated to implement but requires unit tests ;-)
>>
>> Marc.
>>
>>
>>
>> -------------------------------------------------------
>> This SF.Net email is sponsored by xPML, a groundbreaking scripting 
>> language
>> that extends applications into web and mobile media. Attend the live 
>> webcast
>> and join the prime developer group breaking into this new coding 
>> territory!
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
>> _______________________________________________
>> Htmlunit-user mailing list
>> Htmlunit-user@...
>> https://lists.sourceforge.net/lists/listinfo/htmlunit-user
>>   
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting 
> language
> that extends applications into web and mobile media. Attend the live 
> webcast
> and join the prime developer group breaking into this new coding 
> territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Htmlunit-user mailing list
> Htmlunit-user@...
> https://lists.sourceforge.net/lists/listinfo/htmlunit-user

According to my tests with Firefox, you can set any value to the
property onchange and a function is called only if the value of onchange
is... a function. This means that the getter/setter of the JS host
object should accept/return an Object and not a Function and that the
code dispatching the onchange event should look at the type of the property.

Marc.

Anthony Laquerre wrote:
> oups sorry I send too fast ;-)
> 
> Heres the problems :
> The problem is the assignation of the field.
> fld.onchange = "";
> the call is :
> setterMethod.invoke(a1,a2)
> a2 is a string but it should be Function.
> and it throw an exception.
> 
> 
> How to fix it now.
> 1- Verify that all args parameter are of the good class type.
> (myParameter.isInstance(a2))
> if everything is correct then invoke the methode.
> if not, what should we do ???
> We dont want to redefined all function with a String as parameter.
> 
> any help ?
> 
> anthony

Alessio Pace wrote:
> ...
> >  interesting. Why 2.71 and not not the recently released 2.8?
>
>
> because the 2.7.1 is the latest on ibiblio :)
> I guess someone should push the xerces team to release the 2.8 on ibiblio

you can do it as well if you need it. It is not reserved to the xerces team.

> ...
> That is a problem related to all libraries which depend on xerces:xerces
> (also commons-jelly:commons-jelly that is required by htmltunit), and
that should be
> updated to depend on xerces:xercesImpl.

this is a problem you will ever have. I doubt that you will get only
dependencies to the latest version of each jar when a project uses many
libraries.

Marc.