[Laszlo-user] Could someone elaborate on this? XML binary encoding:

Andrew Chandler andy at riftware.com
Wed Mar 8 23:54:39 EST 2006


Thanks Henry, however I wasn't asking anyone to debug it - we have a
perfectly working solution as far as I'm aware.  I only wanted explanation
of a statement that is made on the laszlo overview or summary pages that I
quoted below.  I wanted to know what it meant - optional implies user or
programmer control and its not explained anywhere I could see.  Again the
caption right off the openlaszlo site is
"Data transcoding
Optionally transcode XML data responses to a compact binary format."  
The above statement can be found on this page:
http://www.openlaszlo.org/architecture    down under the laszlo servlet
section.
 


  _____  

From: laszlo-user-bounces at openlaszlo.org
[mailto:laszlo-user-bounces at openlaszlo.org] On Behalf Of Henry Minsky
Sent: Wednesday, March 08, 2006 5:56 PM
To: Andy Chandler
Cc: Laszlo-user at openlaszlo.org
Subject: Re: [Laszlo-user] Could someone elaborate on this? XML binary
encoding:


The Flash instruction set cannot support strings longer than 64 kbytes, but
the runtime can certainly have longer strings in memory. We have code in the
XML encoder to split longer strings and reassemble them using byte codes for
push and string append. However, it is possible that the Java rpc encoder,
which shares code with the XML encoder,  is generating some kind of overly
long instruction byte codes. It would be best if you could send us a test
case with long strings that are failing, otherwise it's difficult for anyone
to get the motivation to debug this. 





On 3/8/06, Andy Chandler <achandler at visionael.com> wrote: 

I believe (and I'll have to double check with the tech lead on the other
project) that the way they have things working is that we have java classes
that return data as a single xml string on the tomcat server.   This string
is retrieved using javarpc from inside an lzx file - the "string" that is
retrieved is then assigned to a dataset and then the laszlo controls are
bound to the dataset.   The limit seems to be in the length of that string -
it can break something either in the browser or somewhere else if its too
long so they developed a mechanism to send it in "chunks" of say 1024
characters and then reassemble it at the client before assigning it to the
dataset.  Is there a fundamentally better way?   I'm also still curious
about the stament below I found on the architecture page.


  _____  

From: Henry Minsky [mailto:henry.minsky at gmail.com] 
Sent: Wednesday, March 08, 2006 4:22 PM
To: Andy Chandler
Cc: Laszlo-user at openlaszlo.org
Subject: Re: [Laszlo-user] Could someone elaborate on this? XML binary
encoding:






On 3/8/06, Andy Chandler <achandler at visionael.com> wrote: 

I found the following tidbit in the architecture document - could someone
point me to where this is elaborated ?    I know we've hit XML string
response limits and implemented a "chunking" mechanism that retrieves long 
responses in pieces but this sounds interesting and perhaps more appropriate
-



What context are you talking about XML string limits? Laszlo currently 
proxies XML data by encoding the XML as a Flash byte-coded instruction
stream, but we're moving to use the newer Flash client-side XML parsing,
because it is now fast enough. For DHTML, we use the XMLHTTPRequest API, so
the XML is sent as text, but it can be gzipped by the server , which is a
better idea all around than a separate binary encoding for XML. 




Data transcoding
Optionally transcode XML data responses to a compact binary format. 


Andrew Chandler

Manager DataCollection Group


Visionael Corporation             Phone : (918) 388-8308

9717 E. 42nd Street               Fax   : (918) 663-1456
<mailto:  <mailto:achandler at visionael.com> achandler at visionael.com>

Suite 200                         Email : andy at visionael.com
<mailto:  <mailto:achandler at visionael.com> achandler at visionael.com >
Tulsa, OK 74146-3680              WEB   : www.visionael.com
<  <http://www.visionael.com/> http://www.visionael.com/>


_______________________________________________ 
Laszlo-user mailing list
Laszlo-user at openlaszlo.org
http://www.openlaszlo.org/mailman/listinfo/laszlo-user
<http://www.openlaszlo.org/mailman/listinfo/laszlo-user> 





-- 
Henry Minsky
Software Architect
hminsky at laszlosystems.com






-- 
Henry Minsky
Software Architect
hminsky at laszlosystems.com



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.openlaszlo.org/pipermail/laszlo-user/attachments/20060308/fae675ab/attachment-0001.html


More information about the Laszlo-user mailing list