Re: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-09-01 Thread tedd

Does anyone have any ideas on what could possibly be causing this?


Matt:

The first thing I would do is to validate the web portion of your 
application. In other words do your forms, as well as the rest of the 
web pages, validate? I've seen instances where people build projects 
around IE when that's not the browser you should be using as a 
foundation, but rather accommodating its shortcomings after you get 
your project working correctly.


The second thing would be to check your javascript for errors. Oddly 
enough, I would use FF (Fire Bug) for that.


You could post a url for us to look at as well.

Cheers,

tedd

--
---
http://sperling.com  http://ancientstones.com  http://earthstones.com

--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



RE: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Williams, Dewey


 -Original Message-
 From: Matt Neimeyer [mailto:m...@neimeyer.org]
 Sent: Monday, August 31, 2009 11:06 AM
 To: php-general
 Subject: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox -
 The impossible!
 
 One of our products allows you to mail merge using an uploaded data
 file and have the results emailed to the recipients. IE works fine,
 always has. Firefox at some point started having the data disappear
 halfway through the merge and now the most recent version of Safari
 ALSO has the data disappear halfway through the merge. This means that
 users of Macs don't really have any options where before at least we
 could point them to Firefox. (Not our favorite option but at least it
 worked)
 
 Here's the process:
 
 1. Menu of what type of merge you want to do. (Initialize working
 table)
 2. Process Data File (Initialize working table and then load in new
 data)
 3. Build Email
 4. Send out Email
 
 
 Matt Neimeyer
 
 --
 PHP General Mailing List (http://www.php.net/)
 To unsubscribe, visit: http://www.php.net/unsub.php

[Dewey Williams]

No answers - just more questions to maybe point you in a direction you
haven't been
 
Is it possible that the query/script is taking too long to build the
response page and FireFox/Safari is asking for an empty query result?

You mention MySQL but is there any JavaScript involved in the
data/result set.  I am not a JS expert, but there could be differences
in IE/FF/Saf JavaScript that could cause this.

When you duplicate this with your data and get the error response, does
the script show a records in the response query, even though it reports
no records?

Sorry but no definitive answers...

Dewey Williams

--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



Re: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Matt Neimeyer
 No answers - just more questions to maybe point you in a direction you
 haven't been

Anything is appreciated...

 Is it possible that the query/script is taking too long to build the
 response page and FireFox/Safari is asking for an empty query result?

I don't think so... the tracer emails I get from the initialize
function are usually a couple seconds apart. It's almost like the
browser is trying to make two simultanious requests which is why I
was thinking maybe prefetching...

 You mention MySQL but is there any JavaScript involved in the
 data/result set.  I am not a JS expert, but there could be differences
 in IE/FF/Saf JavaScript that could cause this.

The only javascript is from the upload screen a popup window is
created that says wait patiently please and then on the next screen
a javascript that closes that popup window. (The email editor is
javascript heavy but the data is already gone by the time you get to
that point. And the only reason the popup exists is because the email
editor requires popups and we have impatient users...)

 When you duplicate this with your data and get the error response, does
 the script show a records in the response query, even though it reports
 no records?

No. By the time the editor screen loads the data really IS gone. If I
sit in a mysql session and keep refreshing the table I see data...
data... data... data... no data... I only mentioned the error
responses as a way (not phrased the best) to indicate that it didn't
see any correlation to this branch of code or that branch of code
(for example if I could narrow it down to it disappears when I
preview then I would know to look at the preview pages)

 Sorry but no definitive answers...

Thanks though... the reason I keep pounding my head on the desk is
that we deliberatly try to keep all the processing on the server side
to avoid cross-browser and cross-platform issues. So sure... maybe a
style sheet breaks... maybe a minor javascript doesn't work quite
right... maybe the font sizes change and things are out of
alignment... but the heavy lifting happens behind the scenes where
they don't need to worry about it. Not very Web 2.0 but it's been
steady (with this exception) so far.

Thanks!

--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



Re: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Robert Cummings

Matt Neimeyer wrote:

No answers - just more questions to maybe point you in a direction you
haven't been


Anything is appreciated...


Is it possible that the query/script is taking too long to build the
response page and FireFox/Safari is asking for an empty query result?


I don't think so... the tracer emails I get from the initialize
function are usually a couple seconds apart. It's almost like the
browser is trying to make two simultanious requests which is why I
was thinking maybe prefetching...


If it were prefetching, or another request clobbering your current 
request then you would see a second hit in your server logs.


Cheers,
Rob.
--
http://www.interjinn.com
Application and Templating Framework for PHP

--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



Re: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Matt Neimeyer
 If it were prefetching, or another request clobbering your current request
 then you would see a second hit in your server logs.

I will admit... I have made (at least) one assumption... Since...

1. This is a difference between IE and Firefox/Safari...
2. I was seeing a tracer email for each hit to the Initialize function...

I have been assuming that it is the browser that has been the cause.
I'm 90% certain that last year we DID check and we DID see multiple
web hits... That said I will make an explicit check and see if there
are multiple hits in the web server logs now for the Safari as well.

Matt

-- 
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



Re: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Paul M Foster
On Mon, Aug 31, 2009 at 11:55:38AM -0400, Matt Neimeyer wrote:

  If it were prefetching, or another request clobbering your current request
  then you would see a second hit in your server logs.
 
 I will admit... I have made (at least) one assumption... Since...
 
 1. This is a difference between IE and Firefox/Safari...
 2. I was seeing a tracer email for each hit to the Initialize function...
 
 I have been assuming that it is the browser that has been the cause.
 I'm 90% certain that last year we DID check and we DID see multiple
 web hits... That said I will make an explicit check and see if there
 are multiple hits in the web server logs now for the Safari as well.

I do something like what you're doing. I process customer mailing lists
into a format which can be used by my Windows mail processing program.
Since I had all this infrastructure already built in Perl/Python/C,
etc., I use system() calls to execute the code and use PHP to stitch the
pieces together. Thus, I have no problems. This could be an alternative
for you, but I don't know enough about your internals to say.

In circumstances like this, I would instrument the code with
echo/print statements all around where you think the problem is. Echo
variables which should be a certain thing, arrays which should exist,
etc. I know, this is probably a blindingly obvious approach, but I've
had circumstances like this, where the problem is completely bizarre and
elusive. And that's how I ultimately solved it (or worked around it).

As with other responders, I have to say that if you're doing the
processing on the server, it doesn't make any sense.

Um... one other wild possibility I almost hate to mention. I don't
recall if you mentioned whether your server was running Windows or not.
But I wouldn't put it past Microsoft to barf under certain circumstances
if the user agent (browser) isn't IE. If it was engineered to happen
only under edge cases, most users would never notice. It may seem
paranoid, but Microsoft has done things like this numerous times before.
If you are running Windows on the server, try exporting the installation
to a Linux server and see if you have the same problem.

Paul

-- 
Paul M. Foster

-- 
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



Re: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Matt Neimeyer
 In circumstances like this, I would instrument the code with
 echo/print statements all around where you think the problem is.

We did that with no luck. We only saw one call to the initialize
function which is why I switched to tracer emails because we knew
the initialize function was the only possible explanation but
couldn't prove it based on any output to the browser.

 As with other responders, I have to say that if you're doing the
 processing on the server, it doesn't make any sense.

 Um... one other wild possibility I almost hate to mention. I don't
 recall if you mentioned whether your server was running Windows or not.

I did not mention... Sorry... according to phpinfo() it's PHP 5.2.5 on
Apache/2.0.52 on CentOS under Plesk 8.2.0. SELECT version() on MySQL
reports 4.1.20-log. The MySQL API version in phpinfo() is 4.1.20. If
it matters from what I can tell the entire toolchain is 64-bit (for
example the php configure command reports
--build=x86_64-redhat-linux-gnu)

 But I wouldn't put it past Microsoft to barf under certain circumstances
 if the user agent (browser) isn't IE.

Neither would I... but unless I have to for some other reason I prefer
PHP 5+, MySQL 5+, Apache 2+ and a *nix of some flavor for my web
serving needs.

Matt

-- 
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



RE: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Bob McConnell
From: Matt Neimeyer

 So far... I can reliably reproduce the problem in Firefox on Windows
 and Mac, Safari on Windows and Mac. But Chrome and IE appear to be
 unaffected.
 
 Finally, have you reproduced the problem while watching the network
 activity using something like tcpdump or Firebug's Net tab?
 
 While I'm familiar with the concepts... :) I have not... this is the
 first problem I've had which could not be directly traced to
 problematic code (php syntax or logic errors, mysql query errors,
 problems with database optimizations, etc)

I would set up Wireshark to capture and compare the http sequences from
each browser. After you capture each stream, use the Follow TCP Stream
option to look at the raw HTTP. If it is the browsers, there should be
some obvious differences in the sequence of requests from them.

Bob McConnell

--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php



Re: [PHP] I'm not crazy I swear it... IE vs Safari and Firefox - The impossible!

2009-08-31 Thread Ben Dunlap
 I would set up Wireshark to capture and compare the http sequences from
 each browser. After you capture each stream, use the Follow TCP Stream
 option to look at the raw HTTP. If it is the browsers, there should be
 some obvious differences in the sequence of requests from them.

This is a good idea, although the Net panel in Firebug would be a very
quick preliminary step to this more systematic approach. And it might
be all you need.

Firebug will parse out the HTTP very nicely for you and the best of it
is that you can watch the requests as the browser sends them, without
even looking aside to another window.

Ben

-- 
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php