Re: [Proto-Scripty] Image cache problem with Ajax

[Richard Quadling]

On 30 September 2011 13:09, Phil Petree  wrote:
> I never meant to imply that onComplete did the same thing as onSuccess, what
> I meant to state was that onComplete ALWAYS gets called last.  onSuccess can
> always set a flag and onComplete can do the UI updates right before turning
> off the spinner.   This is useful if you're experiencing those
> "asynchronicity" problems that the OP thought he had...  Almost a poor mans
> semaphore manager.

I'd be wary of performing activities in onComplete which are based
upon results in onSuccess.

I've patched my Prototype to allow me to create a responder for any
callback (including statuses) as well as the ability for the
responders to block the local callback. Primarily this is for AJAX
error handling. I can return an appropriate status from the server, or
the request can fail (for something out side of my control) and all
the handling is done at the responder level. This makes my specific
AJAX calls much cleaner as they only need the onSuccess callback. This
mechanism works very well for me.




-- 
Richard Quadling
Twitter : EE : Zend : PHPDoc
@RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY : bit.ly/lFnVea

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Phil Petree]

I never meant to imply that onComplete did the same thing as onSuccess, what
I meant to state was that onComplete ALWAYS gets called last.  onSuccess can
always set a flag and onComplete can do the UI updates right before turning
off the spinner.   This is useful if you're experiencing those
"asynchronicity" problems that the OP thought he had...  Almost a poor mans
semaphore manager.

On Fri, Sep 30, 2011 at 7:53 AM, Chris Sansom  wrote:

> On 30 Sep 2011, at 12:32, Richard Quadling wrote:
>
> > onFailure / onSuccess is in response to a working result from the AJAX
> call.
> >
> > onComplete is in response to the AJAX mechanism shutting down.
> >
> > onComplete will always be called, but it isn't its job to deal with
> > the data from the call. That is onSuccess. If the call failed for some
> > reason, onComplete wouldn't have any data to work with and could make
> > things worse.
> >
> > I use onCreate and onComplete to turn on/off a little spinner showing
> > that the app is doing something behind the scenes.
>
> Right - thanks. That could be useful.
>
> --
> Cheers... Chris
> Highway 57 Web Development -- 
>
> --
> You received this message because you are subscribed to the Google Groups
> "Prototype & script.aculo.us" group.
> To post to this group, send email to
> prototype-scriptaculous@googlegroups.com.
> To unsubscribe from this group, send email to
> prototype-scriptaculous+unsubscr...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/prototype-scriptaculous?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Richard Quadling]

On 30 September 2011 12:53, Chris Sansom  wrote:
> On 30 Sep 2011, at 12:32, Richard Quadling wrote:
>
>> onFailure / onSuccess is in response to a working result from the AJAX call.
>>
>> onComplete is in response to the AJAX mechanism shutting down.
>>
>> onComplete will always be called, but it isn't its job to deal with
>> the data from the call. That is onSuccess. If the call failed for some
>> reason, onComplete wouldn't have any data to work with and could make
>> things worse.
>>
>> I use onCreate and onComplete to turn on/off a little spinner showing
>> that the app is doing something behind the scenes.
>
> Right - thanks. That could be useful.
>
> --
> Cheers... Chris
> Highway 57 Web Development -- 
>
> --
> You received this message because you are subscribed to the Google Groups 
> "Prototype & script.aculo.us" group.
> To post to this group, send email to prototype-scriptaculous@googlegroups.com.
> To unsubscribe from this group, send email to 
> prototype-scriptaculous+unsubscr...@googlegroups.com.
> For more options, visit this group at 
> http://groups.google.com/group/prototype-scriptaculous?hl=en.
>
>

And if you use the Ajax.Responders
(http://api.prototypejs.org/ajax/Ajax/Responders/), you can define
that behaviour once for all of the ajax calls you make, rather than
having to add it to each individual ajax instance.



-- 
Richard Quadling
Twitter : EE : Zend : PHPDoc
@RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY : bit.ly/lFnVea

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Chris Sansom]

On 30 Sep 2011, at 12:32, Richard Quadling wrote:

> onFailure / onSuccess is in response to a working result from the AJAX call.
> 
> onComplete is in response to the AJAX mechanism shutting down.
> 
> onComplete will always be called, but it isn't its job to deal with
> the data from the call. That is onSuccess. If the call failed for some
> reason, onComplete wouldn't have any data to work with and could make
> things worse.
> 
> I use onCreate and onComplete to turn on/off a little spinner showing
> that the app is doing something behind the scenes.

Right - thanks. That could be useful.

-- 
Cheers... Chris
Highway 57 Web Development -- 

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Richard Quadling]

On 30 September 2011 10:27, Chris Sansom  wrote:
> Right - time to deal with all these helpful answers I've been getting, but 
> first, to save time, let me say... it's fixed! And a very simple fix it was 
> too, which I /thought/ I'd already tried...
>
> On 29 Sep 2011, at 17:15, Richard Quadling wrote:
>> Create a test case where it goes wrong. Write new clean code that
>> doesn't want/need/use anything from the main project.
>
> I had just prepared such a project when I found the answer among the answers 
> below!
>
> And on 29 Sep 2011, at 18:04, Phil Petree wrote:
>> This is an interesting problem... my first reaction is that you'd want to 
>> use onComplete to update the div's instead of onSuccess.
>>
>> Test this with a couple of alerts and see which one gets called first and 
>> which is last (just as onCreate is the first call, onComplete is the last).
>
> Well I had already tried onComplete and it made no difference, but I tried 
> your suggestion nevertheless and sure enough, wherever I put the alerts in 
> the code of the callbacks, onSuccess always emerged before onComplete. It 
> also didn’t matter (as I assumed it wouldn't) if I put the onComplete before 
> the onSuccess in the code.
>
> Then on 29 Sep 2011, at 18:28, Phil Petree wrote:
>> This guy had a solution that worked for me:
>> http://www.irt.org/script/416.htm
>
> I had already tried using the time in a similar way to this, but it didn’t 
> work. However, in the same message...
>
>> I have also used the cheap trick of adding a random query string on to the 
>> end of the image url:
>> http://www.somedomain.com/images/newname.jpg?id=random_number and since this 
>> will always generate a new url, the browser will refresh the image.
>
> That was the one! As I say, I thought I'd done that - or something very like 
> it - in my numerous attempts at different solutions, but I obviously didn’t 
> do exactly this, and it seems to work reliably every time in a suitably 
> bulletproof fashion. So simple!
>
> It also seems to confirm that my initial suspicions were correct, and it is 
> after all a caching issue, not to do with, er, asynchronicity (if that's a 
> word - if not, Ive just invented it, so there).
>
> So many thanks for your time and thoughts, Richard and Phil, also ncubica and 
> Walter for chipping in.
>
> --
> Cheers... Chris
> Highway 57 Web Development -- 
>
> --
> You received this message because you are subscribed to the Google Groups 
> "Prototype & script.aculo.us" group.
> To post to this group, send email to prototype-scriptaculous@googlegroups.com.
> To unsubscribe from this group, send email to 
> prototype-scriptaculous+unsubscr...@googlegroups.com.
> For more options, visit this group at 
> http://groups.google.com/group/prototype-scriptaculous?hl=en.
>
>

onFailure / onSuccess is in response to a working result from the AJAX call.

onComplete is in response to the AJAX mechanism shutting down.

onComplete will always be called, but it isn't its job to deal with
the data from the call. That is onSuccess. If the call failed for some
reason, onComplete wouldn't have any data to work with and could make
things worse.

I use onCreate and onComplete to turn on/off a little spinner showing
that the app is doing something behind the scenes.

-- 
Richard Quadling
Twitter : EE : Zend : PHPDoc
@RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY : bit.ly/lFnVea

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Chris Sansom]

Right - time to deal with all these helpful answers I've been getting, but 
first, to save time, let me say... it's fixed! And a very simple fix it was 
too, which I /thought/ I'd already tried...

On 29 Sep 2011, at 17:15, Richard Quadling wrote:
> Create a test case where it goes wrong. Write new clean code that
> doesn't want/need/use anything from the main project.

I had just prepared such a project when I found the answer among the answers 
below!

And on 29 Sep 2011, at 18:04, Phil Petree wrote:
> This is an interesting problem... my first reaction is that you'd want to use 
> onComplete to update the div's instead of onSuccess.
>  
> Test this with a couple of alerts and see which one gets called first and 
> which is last (just as onCreate is the first call, onComplete is the last).

Well I had already tried onComplete and it made no difference, but I tried your 
suggestion nevertheless and sure enough, wherever I put the alerts in the code 
of the callbacks, onSuccess always emerged before onComplete. It also didn’t 
matter (as I assumed it wouldn't) if I put the onComplete before the onSuccess 
in the code.

Then on 29 Sep 2011, at 18:28, Phil Petree wrote:
> This guy had a solution that worked for me:
> http://www.irt.org/script/416.htm

I had already tried using the time in a similar way to this, but it didn’t 
work. However, in the same message...

> I have also used the cheap trick of adding a random query string on to the 
> end of the image url:
> http://www.somedomain.com/images/newname.jpg?id=random_number and since this 
> will always generate a new url, the browser will refresh the image.

That was the one! As I say, I thought I'd done that - or something very like it 
- in my numerous attempts at different solutions, but I obviously didn’t do 
exactly this, and it seems to work reliably every time in a suitably 
bulletproof fashion. So simple!

It also seems to confirm that my initial suspicions were correct, and it is 
after all a caching issue, not to do with, er, asynchronicity (if that's a word 
- if not, Ive just invented it, so there).

So many thanks for your time and thoughts, Richard and Phil, also ncubica and 
Walter for chipping in.

-- 
Cheers... Chris
Highway 57 Web Development -- 

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Phil Petree]

I didnt address the caching because I had to look to see if I could find
where I had saved this link off the last time I had this problem... found
it!

The caching is probably on the browser side, not the server side.  Setting
the server side cache variables will only affect the page reload.

The browser knows better than to cache on refresh/reload BUT technically, to
the browser, you are not reloading and since the image has the same name it
doesn't really know that you need a refresh.

This guy had a solution that worked for me:
http://www.irt.org/script/416.htm

I have also used the cheap trick of adding a random query string on to the
end of the image url:
http://www.somedomain.com/images/newname.jpg?id=random_number and since this
will always generate a new url, the browser will refresh the image.


On Thu, Sep 29, 2011 at 1:04 PM, Phil Petree  wrote:

> This is an interesting problem... my first reaction is that you'd want
> to use onComplete to update the div's instead of onSuccess.
>
> Test this with a couple of alerts and see which one gets called first and
> which is last (just as onCreate is the first call, onComplete is the last).
>
> To my way if thinking, if you wait until onComplete gets triggered before
> you do any UI updates, all the images on the server should be properly in
> place.
>
> On Thu, Sep 29, 2011 at 12:15 PM, Richard Quadling wrote:
>
>> On 29 September 2011 16:58, Chris Sansom  wrote:
>> > On 29 Sep 2011, at 15:51, Richard Quadling wrote:
>> >
>> >> Does ALL the JS work take place inside the onSuccess callback?
>> >>
>> >> The "Back in JS" bit has to be part of the onSuccess callback
>> >> otherwise it will happen out of sequence. The A in AJAX is potentially
>> >> the hiccough here.
>> >
>> > That's what I suspected but yes, all the 'back in JS' stuff does indeed
>> happen in the onSuccess. (I also tried onComplete, but got the same result.)
>> I think the problem may be that the div, inevitably, is replaced right at
>> the end of the process (at the end of the onSuccess), and only then is the
>> offending  tag unleashed, calling either the image itself or my little
>> php script... but then I'd have thought preloading it might help, but it
>> doesn’t seem to. I also tried loading it via a php exec() call to the image
>> script in advance of returning the output string to JS, but that didn’t
>> help.
>> >
>> > What also convinces me that you're right about the A in AJAX is that
>> when, for testing, I put a sleep(5) in the image script - which should hold
>> it up by a whole 5 seconds - the div is still replaced immediately. When I
>> first load the page (which also calls this script), I get a broken image
>> icon where the image should have been, replaced after 5 seconds by the
>> image, but when the div is replaced by the ajax call that doesn’t happen - I
>> just get no change of image as before.
>> >
>> > It really would be /so/ nice if I could get this working! It's for a
>> password-protected CMS, so the world at large will never get the benefit,
>> and I could simply reload the whole page instead of just the one div, but
>> it's become a challenge!
>>
>> Create a test case where it goes wrong. Write new clean code that
>> doesn't want/need/use anything from the main project.
>>
>> At best, this will be a small HTML page with some divs and images, a
>> JS file to allow the onclick to fire the AJAX code, along with the
>> onsuccess and the server side code to handle the request and to return
>> the new HTML markup.
>>
>>
>> Without seeing the server and client side code, you are going to be
>> stuck with a limited level of support.
>>
>> If you can't reduce the problem to something that can be read, I doubt
>> anyone can realistically provide any more ideas on this.
>>
>>
>> --
>> Richard Quadling
>> Twitter : EE : Zend : PHPDoc
>> @RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY : bit.ly/lFnVea
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Prototype & script.aculo.us" group.
>> To post to this group, send email to
>> prototype-scriptaculous@googlegroups.com.
>> To unsubscribe from this group, send email to
>> prototype-scriptaculous+unsubscr...@googlegroups.com.
>> For more options, visit this group at
>> http://groups.google.com/group/prototype-scriptaculous?hl=en.
>>
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Phil Petree]

This is an interesting problem... my first reaction is that you'd want
to use onComplete to update the div's instead of onSuccess.

Test this with a couple of alerts and see which one gets called first and
which is last (just as onCreate is the first call, onComplete is the last).

To my way if thinking, if you wait until onComplete gets triggered before
you do any UI updates, all the images on the server should be properly in
place.

On Thu, Sep 29, 2011 at 12:15 PM, Richard Quadling wrote:

> On 29 September 2011 16:58, Chris Sansom  wrote:
> > On 29 Sep 2011, at 15:51, Richard Quadling wrote:
> >
> >> Does ALL the JS work take place inside the onSuccess callback?
> >>
> >> The "Back in JS" bit has to be part of the onSuccess callback
> >> otherwise it will happen out of sequence. The A in AJAX is potentially
> >> the hiccough here.
> >
> > That's what I suspected but yes, all the 'back in JS' stuff does indeed
> happen in the onSuccess. (I also tried onComplete, but got the same result.)
> I think the problem may be that the div, inevitably, is replaced right at
> the end of the process (at the end of the onSuccess), and only then is the
> offending  tag unleashed, calling either the image itself or my little
> php script... but then I'd have thought preloading it might help, but it
> doesn’t seem to. I also tried loading it via a php exec() call to the image
> script in advance of returning the output string to JS, but that didn’t
> help.
> >
> > What also convinces me that you're right about the A in AJAX is that
> when, for testing, I put a sleep(5) in the image script - which should hold
> it up by a whole 5 seconds - the div is still replaced immediately. When I
> first load the page (which also calls this script), I get a broken image
> icon where the image should have been, replaced after 5 seconds by the
> image, but when the div is replaced by the ajax call that doesn’t happen - I
> just get no change of image as before.
> >
> > It really would be /so/ nice if I could get this working! It's for a
> password-protected CMS, so the world at large will never get the benefit,
> and I could simply reload the whole page instead of just the one div, but
> it's become a challenge!
>
> Create a test case where it goes wrong. Write new clean code that
> doesn't want/need/use anything from the main project.
>
> At best, this will be a small HTML page with some divs and images, a
> JS file to allow the onclick to fire the AJAX code, along with the
> onsuccess and the server side code to handle the request and to return
> the new HTML markup.
>
>
> Without seeing the server and client side code, you are going to be
> stuck with a limited level of support.
>
> If you can't reduce the problem to something that can be read, I doubt
> anyone can realistically provide any more ideas on this.
>
>
> --
> Richard Quadling
> Twitter : EE : Zend : PHPDoc
> @RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY : bit.ly/lFnVea
>
> --
> You received this message because you are subscribed to the Google Groups
> "Prototype & script.aculo.us" group.
> To post to this group, send email to
> prototype-scriptaculous@googlegroups.com.
> To unsubscribe from this group, send email to
> prototype-scriptaculous+unsubscr...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/prototype-scriptaculous?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Richard Quadling]

On 29 September 2011 16:58, Chris Sansom  wrote:
> On 29 Sep 2011, at 15:51, Richard Quadling wrote:
>
>> Does ALL the JS work take place inside the onSuccess callback?
>>
>> The "Back in JS" bit has to be part of the onSuccess callback
>> otherwise it will happen out of sequence. The A in AJAX is potentially
>> the hiccough here.
>
> That's what I suspected but yes, all the 'back in JS' stuff does indeed 
> happen in the onSuccess. (I also tried onComplete, but got the same result.) 
> I think the problem may be that the div, inevitably, is replaced right at the 
> end of the process (at the end of the onSuccess), and only then is the 
> offending  tag unleashed, calling either the image itself or my little 
> php script... but then I'd have thought preloading it might help, but it 
> doesn’t seem to. I also tried loading it via a php exec() call to the image 
> script in advance of returning the output string to JS, but that didn’t help.
>
> What also convinces me that you're right about the A in AJAX is that when, 
> for testing, I put a sleep(5) in the image script - which should hold it up 
> by a whole 5 seconds - the div is still replaced immediately. When I first 
> load the page (which also calls this script), I get a broken image icon where 
> the image should have been, replaced after 5 seconds by the image, but when 
> the div is replaced by the ajax call that doesn’t happen - I just get no 
> change of image as before.
>
> It really would be /so/ nice if I could get this working! It's for a 
> password-protected CMS, so the world at large will never get the benefit, and 
> I could simply reload the whole page instead of just the one div, but it's 
> become a challenge!

Create a test case where it goes wrong. Write new clean code that
doesn't want/need/use anything from the main project.

At best, this will be a small HTML page with some divs and images, a
JS file to allow the onclick to fire the AJAX code, along with the
onsuccess and the server side code to handle the request and to return
the new HTML markup.


Without seeing the server and client side code, you are going to be
stuck with a limited level of support.

If you can't reduce the problem to something that can be read, I doubt
anyone can realistically provide any more ideas on this.


-- 
Richard Quadling
Twitter : EE : Zend : PHPDoc
@RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY : bit.ly/lFnVea

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Chris Sansom]

On 29 Sep 2011, at 15:51, Richard Quadling wrote:

> Does ALL the JS work take place inside the onSuccess callback?
> 
> The "Back in JS" bit has to be part of the onSuccess callback
> otherwise it will happen out of sequence. The A in AJAX is potentially
> the hiccough here.

That's what I suspected but yes, all the 'back in JS' stuff does indeed happen 
in the onSuccess. (I also tried onComplete, but got the same result.) I think 
the problem may be that the div, inevitably, is replaced right at the end of 
the process (at the end of the onSuccess), and only then is the offending  
tag unleashed, calling either the image itself or my little php script... but 
then I'd have thought preloading it might help, but it doesn’t seem to. I also 
tried loading it via a php exec() call to the image script in advance of 
returning the output string to JS, but that didn’t help.

What also convinces me that you're right about the A in AJAX is that when, for 
testing, I put a sleep(5) in the image script - which should hold it up by a 
whole 5 seconds - the div is still replaced immediately. When I first load the 
page (which also calls this script), I get a broken image icon where the image 
should have been, replaced after 5 seconds by the image, but when the div is 
replaced by the ajax call that doesn’t happen - I just get no change of image 
as before.

It really would be /so/ nice if I could get this working! It's for a 
password-protected CMS, so the world at large will never get the benefit, and I 
could simply reload the whole page instead of just the one div, but it's become 
a challenge!

-- 
Cheers... Chris
Highway 57 Web Development -- 

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Richard Quadling]

On 29 September 2011 15:00, Chris Sansom  wrote:
> In sequence, this happens:
> -  User clicks image
> -  Ajax request is created, running a php script with various parameters, in 
> which...
> -  ...the necessary renaming takes place (which involves checking for 
> duplicates etc.), then...
> -  ...a php file is included* which does various processing to build up an 
> output string for the replacement div...
> -  ...in which all <> are replaced by [] so as not to confuse the xml return.
> -  Back in JS, the [] are changed back to <> and the containing div is 
> replaced.

Does ALL the JS work take place inside the onSuccess callback?


The "Back in JS" bit has to be part of the onSuccess callback
otherwise it will happen out of sequence. The A in AJAX is potentially
the hiccough here.

Regards,

Richard.
-- 
Richard Quadling
Twitter : EE : Zend : PHPDoc
@RQuadling : e-e.com/M_248814.html : bit.ly/9O8vFY : bit.ly/lFnVea

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

Re: [Proto-Scripty] Image cache problem with Ajax

[Chris Sansom]

On 29 Sep 2011, at 15:00, Chris Sansom wrote:

> I've tried various things to stop it caching, such as loading the top image 
> via a tiny php script

Forgot to mention, before anyone else does, that the first thing I tried, after 
Googling, was to put various query strings [such as '?' + (new 
Date()).getTime()] after the file name in the  tag. Didn’t work.

-- 
Cheers... Chris
Highway 57 Web Development -- 

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.

[Proto-Scripty] Image cache problem with Ajax

[Chris Sansom]

I don't know if this is really something I can fix via Prototype, or if it's in 
any way a Prototype issue, but I'll try here first. Here is my problem - which 
unfortunately I can’t show you because I'm only developing it on my local 
system at the moment - slightly simplified:

I have a pair of images on a page, one above the other. The user should be able 
to click the second one to swap them round, but I don't mean merely swapping 
their position on the page: this involves renaming the files. On the server, 
the first image is given a completely new name and the second is renamed with 
the first's old name. The div containing the two is then replaced by the magic 
of Ajax, except... the problem is that the image with the old name is cached (I 
think), so although the second image is indeed replaced by the first with its 
brand new name, the first image is unchanged, either because the previous image 
with that name is stuck in the cache or because it hasn’t had time to be 
reloaded due to the asynchronous nature of the process. Either way, I end up 
with two of the same image. If I then refresh the page altogether, I get the 
correct images.

In sequence, this happens:
-  User clicks image
-  Ajax request is created, running a php script with various parameters, in 
which...
-  ...the necessary renaming takes place (which involves checking for 
duplicates etc.), then...
-  ...a php file is included* which does various processing to build up an 
output string for the replacement div...
-  ...in which all <> are replaced by [] so as not to confuse the xml return.
-  Back in JS, the [] are changed back to <> and the containing div is replaced.

So is my problem the cache? Or is it simply that the scripts run faster than 
the replacement image can be loaded? I've tried various things to stop it 
caching, such as loading the top image via a tiny php script which sends all 
manner of no-cache type headers then does fpassthru, fread or whatever (I've 
tried it various ways) - this made no difference. I even tried including a 
sleep(1) command in this script! - no good. I've also tried preloading the 
image in JS before the new div is output - no good. Should I be trying to 
subvert (pervert?) Ajax into running synchronously somehow?

I'm quite prepared to accept that I may be approaching this whole problem 
bass-ackwards, and would be grateful for any pointers.

* I did this via an include file because I need to do the same processing when 
the page is initially loaded, so this seemed the most efficient way, code-wise.

-- 
Cheers... Chris
Highway 57 Web Development -- 

-- 
You received this message because you are subscribed to the Google Groups 
"Prototype & script.aculo.us" group.
To post to this group, send email to prototype-scriptaculous@googlegroups.com.
To unsubscribe from this group, send email to 
prototype-scriptaculous+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/prototype-scriptaculous?hl=en.