Rails spinoffs - Jul 2005 - Autocompletion not working in Konqueror

I''ve just had a look again at the autocompletion demo that Thomas put
up
at

http://script.aculo.us/demos/ajax/autocompleter

Unfortunately, in Konqueror (3.4.1) it doesn''t work at all. Not even an
error is triggered. It does work nicely in Firefox. How about Safari?

I don''t have the time right now to track this down systematically
(maybe
latter), but just from appearances it looks as if the listeners 
attached to the text field aren''t doing anything.

Michael

-- 
Michael Schuerig                             Those who call the shots
mailto:michael@schuerig.de              Are never in the line of fire
http://www.schuerig.de/michael/           --Ani DiFranco, Not So Soft

Works and tested in Safari, Firefox, IE 6.

Thomas

Am 06.07.2005 um 14:57 schrieb Michael Schuerig:
>
> I''ve just had a look again at the autocompletion demo that Thomas
> put up
> at
>
> http://script.aculo.us/demos/ajax/autocompleter
>
> Unfortunately, in Konqueror (3.4.1) it doesn''t work at all. Not  
> even an
> error is triggered. It does work nicely in Firefox. How about Safari?
>
> I don''t have the time right now to track this down systematically
> (maybe
> latter), but just from appearances it looks as if the listeners
> attached to the text field aren''t doing anything.
>
> Michael
>
> -- 
> Michael Schuerig                             Those who call the shots
> mailto:michael@schuerig.de              Are never in the line of fire
> http://www.schuerig.de/michael/           --Ani DiFranco, Not So Soft
> _______________________________________________
> Rails-spinoffs mailing list
> Rails-spinoffs@lists.rubyonrails.org
> http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
>

On Wednesday 06 July 2005 15:18, Thomas Fuchs wrote:
> Works and tested in Safari, Firefox, IE 6.
The workaround in Prototype.js for Safari needs to be applied to 
khtml-based browsers more generally. Patch:

http://dev.rubyonrails.com/ticket/1641

Michael

-- 
Michael Schuerig                       Face reality and stare it down
mailto:michael@schuerig.de        --Jethro Tull, Silver River Turning
http://www.schuerig.de/michael/

Hi, a bit late but, I''ve come up with this:

// KHTML/Safari: http://developer.apple.com/internet/safari/ 
faq.html#anchor2
// Gecko: http://developer-test.mozilla.org/en/docs/ 
Browser_Detection_and_Cross_Browser_Support
// Opera: http://www.opera.com/support/search/supsearch.dml?index=570

var Engine = {
   detect: function() {
     var UA = navigator.userAgent;
     this.isKHTML = /Konqueror|Safari|KHTML/.test(UA);
     this.isGecko = (/Gecko/.test(UA) && !this.isKHTML);
     this.isOpera = /Opera/.test(UA);
     this.isIE    = (/MSIE/.test(UA) && !this.isOpera);
   }
}
Engine.detect();

Note that the Gecko test not on navigator.product because Safari  
returns "Gecko" too here. :P

Can you test this on Konqueror? Only Engine.isKHTML should be true.

Thomas

Am 07.07.2005 um 17:21 schrieb Michael Schuerig:
> On Wednesday 06 July 2005 15:18, Thomas Fuchs wrote:
>
>> Works and tested in Safari, Firefox, IE 6.
>>
>
> The workaround in Prototype.js for Safari needs to be applied to
> khtml-based browsers more generally. Patch:
>
> http://dev.rubyonrails.com/ticket/1641
>
> Michael
>
> -- 
> Michael Schuerig                       Face reality and stare it down
> mailto:michael@schuerig.de        --Jethro Tull, Silver River Turning
> http://www.schuerig.de/michael/
> _______________________________________________
> Rails-spinoffs mailing list
> Rails-spinoffs@lists.rubyonrails.org
> http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
>

On Saturday 16 July 2005 20:34, Thomas Fuchs wrote:
> Hi, a bit late but, I''ve come up with this:
>
> // KHTML/Safari: http://developer.apple.com/internet/safari/
> faq.html#anchor2
> // Gecko: http://developer-test.mozilla.org/en/docs/
> Browser_Detection_and_Cross_Browser_Support
> // Opera: http://www.opera.com/support/search/supsearch.dml?index=570
>
> var Engine = {
>    detect: function() {
>      var UA = navigator.userAgent;
>      this.isKHTML = /Konqueror|Safari|KHTML/.test(UA);
>      this.isGecko = (/Gecko/.test(UA) && !this.isKHTML);
>      this.isOpera = /Opera/.test(UA);
>      this.isIE    = (/MSIE/.test(UA) && !this.isOpera);
>    }
> }
> Engine.detect();
>
> Note that the Gecko test not on navigator.product because Safari
> returns "Gecko" too here. :P
>
> Can you test this on Konqueror? Only Engine.isKHTML should be true.
Works. Thanks.

Michael

-- 
Michael Schuerig                  There is no matrix,
mailto:michael@schuerig.de        only reality.
http://www.schuerig.de/michael/   --Lawrence Fishburn

On Saturday 16 July 2005 20:34, Thomas Fuchs wrote:
> Hi, a bit late but, I''ve come up with this:
>
> // KHTML/Safari: http://developer.apple.com/internet/safari/
> faq.html#anchor2
> // Gecko: http://developer-test.mozilla.org/en/docs/
> Browser_Detection_and_Cross_Browser_Support
> // Opera: http://www.opera.com/support/search/supsearch.dml?index=570
>
> var Engine = {
>    detect: function() {
>      var UA = navigator.userAgent;
>      this.isKHTML = /Konqueror|Safari|KHTML/.test(UA);
>      this.isGecko = (/Gecko/.test(UA) && !this.isKHTML);
>      this.isOpera = /Opera/.test(UA);
>      this.isIE    = (/MSIE/.test(UA) && !this.isOpera);
>    }
> }
> Engine.detect();
Thomas,

has anything become of this? As I wrote back then, the detection works 
correctly on Konqueror.

Michael

-- 
Michael Schuerig                      They tell you that the darkness
mailto:michael@schuerig.de                  Is a blessing in disguise
http://www.schuerig.de/michael/           --Janis Ian, From Me To You

On 19/08/05, Michael Schuerig <michael@schuerig.de> wrote:

I do not know the broader context of this, but I am in favor of
feature testing whereever possible. The danger is that old code using
this Engine thing excludes browser unnecessarily when they gain
previously unavailable features.

Testing the user agent is unreliable. Better test for features
specific to browsers.
> >      this.isGecko = (/Gecko/.test(UA) && !this.isKHTML);
Is better done by testing for window.pkcs11
> >      this.isOpera = /Opera/.test(UA);
Is better done by testing for window.opera
> >      this.isIE    = (/MSIE/.test(UA) && !this.isOpera);
Is better done by using conditional compiles.
>      this.isKHTML = /Konqueror|Safari|KHTML/.test(UA);
Konquerer has document.clientWidth, maybe Safari has it as well?


Bye,
  Martin

On Saturday 20 August 2005 12:06, Martin Bialasinski
wrote:
> On 19/08/05, Michael Schuerig <michael@schuerig.de> wrote:
>
> I do not know the broader context of this, but I am in favor of
> feature testing whereever possible.
The "wherever possible" is the catch. There are cases where it is not 
possible, notably buggy implementations. The original context was that 
Safari and Konqueror don''t handle listening for keypress events 
correctly. For Safari, there is a "sniffing" workaround in
prototype.js
(Event.observe), Konqueror is left in the cold.
> The danger is that old code using 
> this Engine thing excludes browser unnecessarily when they gain
> previously unavailable features.
Agreed, but what can be done about it? Sometimes it is possible to check 
if the bug is present (see below for an example). At other times it is 
not.

Michael


StdExt.native_date = Date;

// khtml Date copy constructor doesn''t work for dates before 1970-01-01
if (StdExt.isKhtml && (new Date(new Date(1969,1,1)).getFullYear()) == 
1970) {
  Date = function() {
    if (arguments.length == 0) {
      return new StdExt.native_date();
    }
    if (arguments.length == 1) {
      d = arguments[0];
      if (d instanceof StdExt.native_date) {
        return new StdExt.native_date(
          d.getFullYear(), d.getMonth(), d.getDate(),
          d.getHours(), d.getMinutes(), d.getSeconds());
      }
    }
    return new StdExt.native_date(arguments[0], arguments[1],
      arguments[2], arguments[3], arguments[4], arguments[5]);
  };
  Date.prototype = StdExt.native_date.prototype
}

-- 
Michael Schuerig                  Failures to use one''s frontal lobes
mailto:michael@schuerig.de        can result in the loss of them.
http://www.schuerig.de/michael/   --William H. Calvin

On 20/08/05, Michael Schuerig <michael@schuerig.de>
wrote:
> On Saturday 20 August 2005 12:06, Martin Bialasinski wrote:
> > I do not know the broader context of this, but I am in favor of
> > feature testing whereever possible.
> 
> The "wherever possible" is the catch. There are cases where it is
not
> possible, notably buggy implementations.
I see. But then, I would still not check the user agent string but
test for the features I  posted.

The decision is then to a) assume broken in any newer version of the
browser, or b) assume fixed in the next version of the browser. Both
have their negative impacts :-(

a) Freeing the web developer from greve, but leaving the users in the
cold when the browser creator fixes a bug
b) Forcing the web developer to update the test, if the browser
creator does not fix the bug in the next version
> Sometimes it is possible to check
> if the bug is present (see below for an example).
Neat example.


Bye,
  Martin

I am trying to use the Ajax.Updater function to perform server side 
validation on a form I have. This works great, but I want the form to submit 
if the form validates and give the user a message if it doesn''t. I can
get
it to give the user a message, but then the form always submits. Is there 
anyway to stop the form from submitting or allow the form to submit based on 
the results coming back to me? Any help would be appreciated.

--
Eric Fleming
efleming@gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails-spinoffs/attachments/20050820/454a5ca2/attachment.html

On 8/21/05, Eric Fleming <efleming@gmail.com>
wrote:
> I am trying to use the Ajax.Updater function to perform server side
> validation on a form I have.  This works great, but I want the form to
> submit if the form validates and give the user a message if it
doesn''t.  I
> can get it to give the user a message, but then the form always submits.
If you are calling the validation from onsubmit, returning true(or
nothing) submits the form, returning false stops the form being
submitted.

HTH,
Rob.


-- 
Rob Sharp

  email/jabber: rob@sharp.id.au
  web: http://sharp.id.au
  pgp: 0E2C C63B BA04 DEB4 7CC0 84FD 17E3 6AA4 87FB 62DF

I am a newbie and wasn''t aware of the Ajax.Request. Sounds like that is
what
I need, thanks for the info.

On 8/20/05, Nic Schlueter <schlueter@gmail.com>
wrote:
> 
> I am sort of perplexed I don''t know a whole lot about the
> Ajax.Updater. But I was just reading over the code for the
> Ajax.Updater and it looks like that is for updating a container
> (correct me if I am wrong). Maybe I just do things the hard way but I
> usually just use Ajax.Request (I pass in a function for onComplete and
> look at the responseText and act accordingly). If you have any
> questions or I could look at your js to see if I can help.
> 
> Here is a URL for the Updater:
> http://wiki.script.aculo.us/scriptaculous/show/Ajax.Updater
> 
> Here is a URL for the Request:
> http://wiki.script.aculo.us/scriptaculous/show/Ajax.Request
> 
> Nicholas
> 
> On 8/20/05, Eric Fleming <efleming@gmail.com> wrote:
> > I am trying to use the Ajax.Updater function to perform server side
> > validation on a form I have. This works great, but I want the form to
> > submit if the form validates and give the user a message if it
doesn''t.
> I
> > can get it to give the user a message, but then the form always
submits.
> Is
> > there anyway to stop the form from submitting or allow the form to 
> submit
> > based on the results coming back to me? Any help would be appreciated.
> >
> > --
> > Eric Fleming
> > efleming@gmail.com
> >
> > _______________________________________________
> > Rails-spinoffs mailing list
> > Rails-spinoffs@lists.rubyonrails.org
> > http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> >
> >
> >
> 


-- 
Eric Fleming
efleming@gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails-spinoffs/attachments/20050821/45b82b32/attachment.html

Still have a question, if I have my Ajax.Request function in the onsubmit 
event for the form and I have an onComplete event in the Ajax.Request, how 
do I return a false if the validation fails. Here is a sample of the code:

<script>
function doValidation(){

}
</script>

<form action="someurl" onsubmit="doValidation()"
method="post">
</form>




On 8/21/05, Eric Fleming <efleming@gmail.com>
wrote:
> 
> I am a newbie and wasn''t aware of the Ajax.Request. Sounds like
that is
> what I need, thanks for the info.
> 
> On 8/20/05, Nic Schlueter < schlueter@gmail.com> wrote:
> > 
> > I am sort of perplexed I don''t know a whole lot about the 
> > Ajax.Updater. But I was just reading over the code for the
> > Ajax.Updater and it looks like that is for updating a container
> > (correct me if I am wrong). Maybe I just do things the hard way but I
> > usually just use Ajax.Request (I pass in a function for onComplete and
> > look at the responseText and act accordingly). If you have any
> > questions or I could look at your js to see if I can help.
> > 
> > Here is a URL for the Updater:
> > http://wiki.script.aculo.us/scriptaculous/show/Ajax.Updater
> > 
> > Here is a URL for the Request:
> > http://wiki.script.aculo.us/scriptaculous/show/Ajax.Request
> > 
> > Nicholas
> > 
> > On 8/20/05, Eric Fleming <efleming@gmail.com> wrote:
> > > I am trying to use the Ajax.Updater function to perform server
side
> > > validation on a form I have. This works great, but I want the
form to
> > > submit if the form validates and give the user a message if it 
> > doesn''t. I
> > > can get it to give the user a message, but then the form always 
> > submits. Is 
> > > there anyway to stop the form from submitting or allow the form
to
> > submit
> > > based on the results coming back to me? Any help would be
appreciated.
> > >
> > > --
> > > Eric Fleming
> > > efleming@gmail.com
> > >
> > > _______________________________________________
> > > Rails-spinoffs mailing list
> > > Rails-spinoffs@lists.rubyonrails.org 
> > > http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> > >
> > >
> > >
> > 
> 
> 
> 
> -- 
> Eric Fleming
> efleming@gmail.com 
> 


-- 
Eric Fleming
efleming@gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails-spinoffs/attachments/20050821/055d35b6/attachment.html

Here''s some pseudo-code for ya:

1) change the onsubmit event to be "return(doValidation());"

function doValidation(){
var bValid = true;

// perform validation routines

// if rules fail, set bValid = false;

return(bValid);
}

On 8/21/05, Eric Fleming <efleming@gmail.com>
wrote:
> 
> Still have a question, if I have my Ajax.Request function in the onsubmit 
> event for the form and I have an onComplete event in the Ajax.Request, how 
> do I return a false if the validation fails. Here is a sample of the code:
> 
> <script>
> function doValidation(){
> 
> }
> </script>
> 
> <form action="someurl" onsubmit="doValidation()"
method="post">
> </form>
> 
> 
> 
> 
> On 8/21/05, Eric Fleming <efleming@gmail.com> wrote:
> > 
> > I am a newbie and wasn''t aware of the Ajax.Request. Sounds
like that is
> > what I need, thanks for the info.
> > 
> > On 8/20/05, Nic Schlueter < schlueter@gmail.com> wrote:
> > > 
> > > I am sort of perplexed I don''t know a whole lot about
the
> > > Ajax.Updater. But I was just reading over the code for the
> > > Ajax.Updater and it looks like that is for updating a container
> > > (correct me if I am wrong). Maybe I just do things the hard way
but I
> > > usually just use Ajax.Request (I pass in a function for
onComplete and
> > > look at the responseText and act accordingly). If you have any
> > > questions or I could look at your js to see if I can help.
> > > 
> > > Here is a URL for the Updater: 
> > > http://wiki.script.aculo.us/scriptaculous/show/Ajax.Updater
> > > 
> > > Here is a URL for the Request: 
> > > http://wiki.script.aculo.us/scriptaculous/show/Ajax.Request
> > > 
> > > Nicholas
> > > 
> > > On 8/20/05, Eric Fleming < efleming@gmail.com> wrote:
> > > > I am trying to use the Ajax.Updater function to perform
server side
> > > > validation on a form I have. This works great, but I want
the form
> > > to
> > > > submit if the form validates and give the user a message if
it
> > > doesn''t. I
> > > > can get it to give the user a message, but then the form
always
> > > submits. Is 
> > > > there anyway to stop the form from submitting or allow the
form to
> > > submit
> > > > based on the results coming back to me? Any help would be 
> > > appreciated.
> > > >
> > > > --
> > > > Eric Fleming
> > > > efleming@gmail.com
> > > >
> > > > _______________________________________________
> > > > Rails-spinoffs mailing list
> > > > Rails-spinoffs@lists.rubyonrails.org 
> > > > http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> > > >
> > > >
> > > >
> > > 
> > 
> > 
> > 
> > -- 
> > Eric Fleming
> > efleming@gmail.com 
> > 
> 
> 
> 
> -- 
> Eric Fleming
> efleming@gmail.com 
> 
> _______________________________________________
> Rails-spinoffs mailing list
> Rails-spinoffs@lists.rubyonrails.org
> http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> 
> 
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails-spinoffs/attachments/20050821/8492ed21/attachment-0001.html

I tried that, but the doValidation function will return true all the time, 
because it doesn''t wait for the oncomplete event to complete before 
returning the result.

On 8/21/05, Ed C. <defeated2k4@gmail.com> wrote:
> 
> Here''s some pseudo-code for ya:
> 
> 1) change the onsubmit event to be "return(doValidation());"
> 
> function doValidation(){
> var bValid = true;
> 
> // perform validation routines
> 
> // if rules fail, set bValid = false;
> 
> return(bValid);
> }
> 
> On 8/21/05, Eric Fleming <efleming@gmail.com> wrote:
> > 
> > Still have a question, if I have my Ajax.Request function in the 
> > onsubmit event for the form and I have an onComplete event in the 
> > Ajax.Request, how do I return a false if the validation fails. Here is
a
> > sample of the code:
> > 
> > <script>
> > function doValidation(){
> > 
> > }
> > </script>
> > 
> > <form action="someurl"
onsubmit="doValidation()" method="post">
> > </form>
> > 
> > 
> > 
> > 
> > On 8/21/05, Eric Fleming < efleming@gmail.com> wrote:
> > > 
> > > I am a newbie and wasn''t aware of the Ajax.Request.
Sounds like that
> > > is what I need, thanks for the info.
> > > 
> > > On 8/20/05, Nic Schlueter < schlueter@gmail.com> wrote:
> > > > 
> > > > I am sort of perplexed I don''t know a whole lot
about the
> > > > Ajax.Updater. But I was just reading over the code for the
> > > > Ajax.Updater and it looks like that is for updating a
container
> > > > (correct me if I am wrong). Maybe I just do things the hard
way but
> > > > I
> > > > usually just use Ajax.Request (I pass in a function for
onComplete
> > > > and
> > > > look at the responseText and act accordingly). If you have
any
> > > > questions or I could look at your js to see if I can help.
> > > > 
> > > > Here is a URL for the Updater: 
> > > > http://wiki.script.aculo.us/scriptaculous/show/Ajax.Updater
> > > > 
> > > > Here is a URL for the Request: 
> > > > http://wiki.script.aculo.us/scriptaculous/show/Ajax.Request
> > > > 
> > > > Nicholas
> > > > 
> > > > On 8/20/05, Eric Fleming < efleming@gmail.com> wrote:
> > > > > I am trying to use the Ajax.Updater function to perform
server
> > > > side
> > > > > validation on a form I have. This works great, but I
want the form
> > > > to
> > > > > submit if the form validates and give the user a
message if it
> > > > doesn''t. I
> > > > > can get it to give the user a message, but then the
form always
> > > > submits. Is 
> > > > > there anyway to stop the form from submitting or allow
the form to
> > > > submit
> > > > > based on the results coming back to me? Any help would
be
> > > > appreciated.
> > > > >
> > > > > --
> > > > > Eric Fleming
> > > > > efleming@gmail.com
> > > > >
> > > > > _______________________________________________
> > > > > Rails-spinoffs mailing list
> > > > > Rails-spinoffs@lists.rubyonrails.org 
> > > > >
http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> > > > >
> > > > >
> > > > >
> > > > 
> > > 
> > > 
> > > 
> > > -- 
> > > Eric Fleming
> > > efleming@gmail.com 
> > > 
> > 
> > 
> > 
> > -- 
> > Eric Fleming
> > efleming@gmail.com 
> > 
> > _______________________________________________
> > Rails-spinoffs mailing list
> > Rails-spinoffs@lists.rubyonrails.org 
> > http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> > 
> > 
> > 
> 

-- 
Eric Fleming
efleming@gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://wrath.rubyonrails.org/pipermail/rails-spinoffs/attachments/20050821/98889fa1/attachment.html

Just have it return false all the time then in your onComplete
callback if it is successful, submit the form with js.  But in my
opinion it would be better to put onBlur events on your form elements
and validate them as the user is filling the form out.  Then you can
highlight the fields, display feedback quicker, or do whatever you
want to make the WebApp seem more responsive.

Nicholas

On 8/21/05, Eric Fleming <efleming@gmail.com>
wrote:
> I tried that, but the doValidation function will return true all the time,
> because it doesn''t wait for the oncomplete event to complete
before
> returning the result.
> 
> 
> On 8/21/05, Ed C. <defeated2k4@gmail.com> wrote:
> > Here''s some pseudo-code for ya:
> > 
> > 1) change the onsubmit event to be "return(doValidation());"
> > 
> > function doValidation(){
> >   var bValid = true;
> > 
> >   // perform validation routines
> > 
> >   // if rules fail, set bValid = false;
> > 
> >   return(bValid);
> > 
> > }
> > 
> > 
> > On 8/21/05, Eric Fleming <efleming@gmail.com > wrote:
> > > Still have a question, if I have my Ajax.Request function in the
> onsubmit event for the form and I have an onComplete event in the
> Ajax.Request, how do I return a false if the validation fails. Here is a
> sample of the code:
> > > 
> > > <script>
> > > function doValidation(){
> > > 
> > > }
> > > </script>
> > > 
> > > <form action="someurl"
onsubmit="doValidation()" method="post">
> > > </form>
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > On 8/21/05, Eric Fleming < efleming@gmail.com> wrote:
> > > > I am a newbie and wasn''t aware of the Ajax.Request.
Sounds like that
> is what I need, thanks for the info.
> > > > 
> > > > 
> > > > 
> > > > On 8/20/05, Nic Schlueter < schlueter@gmail.com>
wrote:
> > > > > I am sort of perplexed I don''t know a whole
lot about the
> > > > > Ajax.Updater.  But I was just reading over the code for
the
> > > > > Ajax.Updater and it looks like that is for updating a
container
> > > > > (correct me if I am wrong).  Maybe I just do things the
hard way but
> I
> > > > > usually just use Ajax.Request (I pass in a function for
onComplete
> and
> > > > > look at the responseText and act accordingly).  If you
have any
> > > > > questions or I could look at your js to see if I can
help.
> > > > > 
> > > > > Here is a URL for the Updater: 
> > > > >
> http://wiki.script.aculo.us/scriptaculous/show/Ajax.Updater
> > > > > 
> > > > > Here is a URL for the Request: 
> > > > >
> http://wiki.script.aculo.us/scriptaculous/show/Ajax.Request
> > > > > 
> > > > > Nicholas
> > > > > 
> > > > > On 8/20/05, Eric Fleming < efleming@gmail.com>
wrote:
> > > > > > I am trying to use the Ajax.Updater function to
perform server
> side
> > > > > > validation on a form I have.  This works great,
but I want the
> form to
> > > > > > submit if the form validates and give the user a
message if it
> doesn''t.  I
> > > > > > can get it to give the user a message, but then
the form always
> submits.  Is 
> > > > > > there anyway to stop the form from submitting or
allow the form to
> submit
> > > > > > based on the results coming back to me?  Any help
would be
> appreciated.
> > > > > >
> > > > > >  --
> > > > > >  Eric Fleming
> > > > > >   efleming@gmail.com
> > > > > >
> > > > > > _______________________________________________
> > > > > > Rails-spinoffs mailing list
> > > > > > Rails-spinoffs@lists.rubyonrails.org 
> > > > > >
> http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> > > > > >
> > > > > >
> > > > > >
> > > > > 
> > > > 
> > > > 
> > > > 
> > > > -- 
> > > > 
> > > > Eric Fleming
> > > > efleming@gmail.com 
> > > 
> > > 
> > > 
> > > -- 
> > > 
> > > Eric Fleming
> > > efleming@gmail.com 
> > > _______________________________________________
> > > Rails-spinoffs mailing list
> > > Rails-spinoffs@lists.rubyonrails.org 
> > >
> http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> > > 
> > > 
> > > 
> > 
> > 
> 
> 
> 
> -- 
> 
> Eric Fleming
> efleming@gmail.com 
> _______________________________________________
> Rails-spinoffs mailing list
> Rails-spinoffs@lists.rubyonrails.org
> http://lists.rubyonrails.org/mailman/listinfo/rails-spinoffs
> 
> 
>

On 21/08/05, Nic Schlueter <schlueter@gmail.com> wrote:
> Just have it return false all the time then in your onComplete
> callback if it is successful, submit the form with js.
The routine should return false, if Ajax is available and true otherwise.

Or in prototype.js style:

No onsubmit in the <form> tag, but after the form:

Event.observe(formId, "submit", doValidation.bindAsEventListener());

and 

function doValidation(event){
  if (ajax is available) {
   initiate checks
   Event.stop(event);
 }
}
> But in my
> opinion it would be better to put onBlur events on your form elements
> and validate them as the user is filling the form out.  Then you can
> highlight the fields, display feedback quicker, or do whatever you
> want to make the WebApp seem more responsive.
I agree with this. But the event to monitor should be onchange. Well,
for check- and radioboxes, it would be onclick. The easiest way to do
the right thing is
new Form.Element.EventObserver(fieldId, checkroutine);

checkroutine will get called with two parameter: the element changed,
and the new value.


Bye,
  Martin