Re: hard crashes with data from Internet

[Mark Brownell]

On Saturday, December 18, 2004, at 01:04 PM, Richard Gaskin wrote:
I thought about checking the data for the presence of something like 
"", but I can imagine circumstances where the garbage might 
also contain some of the returned data (though I haven't logged enough 
yet to really know how frequently, if ever, that's really the case).

That may be sufficient in most cases, but I'd sure like to find a 100% 
method if possible.  I don't mind script errors, but things that cause 
hard crashes make my work look really sloppy to the user. :)

--
 Richard Gaskin
What I do is to set the htmlText of an off screen field to my input 
variable while the screen is locked then I put the htmlText of that off 
screen field into another variable and unlock the screen. That might 
work for you as a clean up tool. On the other hand if you are 
deliberately working with valid HTML 4.0 then that would not work as a 
fix if your text must remain as HTML 4.0.

Mark
___
use-revolution mailing list
[EMAIL PROTECTED]
http://lists.runrev.com/mailman/listinfo/use-revolution

Re: hard crashes with data from Internet

[Scott Rossi]

Recently, Richard Gaskin wrote:

> I have an app that makes a series of HTTP calls to a server to obtain a
> list of directory contents in nested directories.  I find when I make
> single calls with some time between them the results are almost 100%
> reliable, but when I make these calls in rapid succession sometimes I
> see garbage in the returned result, where the data from the server
> should be.
> 
> The nature of the garbage suggests it may be the result of an errant
> pointer, and this issue is being tracked in Bugzilla:
> 
> 
> But my question for the moment is:  Anyone seen this before and have a
> workaround in place for preventing this errant-pointer garbage?
> 
> To be more specific, the engine doesn't crash until I try to set the
> htmlText of a field to it -- can you think of a way to determine if the
> data is valid HTML?
> 
> I thought about checking the data for the presence of something like
> "", but I can imagine circumstances where the garbage might also
> contain some of the returned data (though I haven't logged enough yet to
> really know how frequently, if ever, that's really the case).
> 
> That may be sufficient in most cases, but I'd sure like to find a 100%
> method if possible.  I don't mind script errors, but things that cause
> hard crashes make my work look really sloppy to the user. :)

I didn't look up the bug you make reference to but I'm assuming it's the one
where trying to grab content from the Web results in partial stack/control
script being returned in the result.  If this is the case, the method I have
been planning to use is checking for the end of the content file, such as
 or whatever is appropriate for the content.  Even better might be to
check for both the beginning and the end if that's possible.  Otherwise,
since every stack is different, I don't know of a reliable check
(unfortunately, the bug is not reliable either).

Regards,

Scott Rossi
Creative Director
Tactile Media, Development & Design
-
E: [EMAIL PROTECTED]
W: http://www.tactilemedia.com

___
use-revolution mailing list
[EMAIL PROTECTED]
http://lists.runrev.com/mailman/listinfo/use-revolution

hard crashes with data from Internet

[Richard Gaskin]

I have an app that makes a series of HTTP calls to a server to obtain a 
list of directory contents in nested directories.  I find when I make 
single calls with some time between them the results are almost 100% 
reliable, but when I make these calls in rapid succession sometimes I 
see garbage in the returned result, where the data from the server 
should be.

The nature of the garbage suggests it may be the result of an errant 
pointer, and this issue is being tracked in Bugzilla:


But my question for the moment is:  Anyone seen this before and have a 
workaround in place for preventing this errant-pointer garbage?

To be more specific, the engine doesn't crash until I try to set the 
htmlText of a field to it -- can you think of a way to determine if the 
data is valid HTML?

I thought about checking the data for the presence of something like 
"", but I can imagine circumstances where the garbage might also 
contain some of the returned data (though I haven't logged enough yet to 
really know how frequently, if ever, that's really the case).

That may be sufficient in most cases, but I'd sure like to find a 100% 
method if possible.  I don't mind script errors, but things that cause 
hard crashes make my work look really sloppy to the user. :)

--
 Richard Gaskin
 Fourth World Media Corporation
 __
 Rev tools and more: http://www.fourthworld.com/rev
___
use-revolution mailing list
[EMAIL PROTECTED]
http://lists.runrev.com/mailman/listinfo/use-revolution