| Author |
Message
|
| shalabh1976 |
 Posted: Thu Jun 12, 2003 11:14 am Post subject: LFCR in input XML message |
 |
|
Partisan
Joined: 18 Jul 2002
Posts: 381
Location: Gurgaon, India
|
It is well known that the MQSI XML parser fails when the input XML message contains LFCR characters.
To get over this I have a small compute node after the input node that parses the whole message bit and bit and copies the input bit to the output if it is not a LFCR character.
This is a big processing overhead and has a major performance impact because each message must be parsed twice.
Is there a way around this or must I continue doing the current processing ?
Do most XML generating applications put in this LFCR on the message ? |
|
| Back to top |
|
 |
| kirani |
| Posted: Thu Jun 12, 2003 11:25 am Post subject: |
|
|
Jedi Knight
Joined: 05 Sep 2001
Posts: 3779
Location: Torrance, CA, USA
|
I guess it's not just supported by the XML parser that WMQI uses internally. I wonder how other XML parsers handle this kind of messages.
In our environment we have not seen even a single application, which generates XML messages with LFCR, so we are safe!
_________________
Kiran
IBM Cert. Solution Designer & System Administrator - WBIMB V5
IBM Cert. Solutions Expert - WMQI
IBM Cert. Specialist - WMQI, MQSeries
IBM Cert. Developer - MQSeries
|
|
| Back to top |
|
|
| shalabh1976 |
| Posted: Thu Jun 12, 2003 12:05 pm Post subject: |
|
|
Partisan
Joined: 18 Jul 2002
Posts: 381
Location: Gurgaon, India
|
Kiran,
My input XML is created by ARGO systems who say that they use some form of a Microsoft XML parser and it has the LFCR in the input message.
Is there a work around to this from IBM ? |
|
| Back to top |
|
|
| shalabh1976 |
| Posted: Fri Jun 13, 2003 7:04 am Post subject: |
|
|
Partisan
Joined: 18 Jul 2002
Posts: 381
Location: Gurgaon, India
|
|
| Back to top |
|
|
| kirani |
| Posted: Fri Jun 13, 2003 9:44 am Post subject: |
|
|
Jedi Knight
Joined: 05 Sep 2001
Posts: 3779
Location: Torrance, CA, USA
|
If I remember correctly, in the conference they were talking about "improving" the XML parser to allow parsing of message with LF,CR chars in the next release.
_________________
Kiran
IBM Cert. Solution Designer & System Administrator - WBIMB V5
IBM Cert. Solutions Expert - WMQI
IBM Cert. Specialist - WMQI, MQSeries
IBM Cert. Developer - MQSeries
|
|
| Back to top |
|
|
| jefflowrey |
| Posted: Fri Jun 13, 2003 11:43 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
In 2.1 CSD04, I can parse XML that has <CR><LF> in it. I get trace output like
| Code: |
(0x2000000) = '
'
(0x1000000)PARAMETER = (
(0x3000000)NAME = 'Status'
(0x2000000) = 'Success'
)
|
Now, parsing <LF><CR> I don't know about.
And I know things get stuck if the first characters in the bitstream are <CR><LF>. (or fun  with parsing xml extracted from non XML bitstreams). |
|
| Back to top |
|
|
| shalabh1976 |
| Posted: Fri Jun 13, 2003 1:45 pm Post subject: |
|
|
Partisan
Joined: 18 Jul 2002
Posts: 381
Location: Gurgaon, India
|
I am on CSD 02 and if CSD 04 solves the problem, ill try using it.
Also recent performance tests here show that the compute node that removes the LFCR from the input is taking 0.002 sec.(approx.);
Thats not too bad if the customer accepts it. |
|
| Back to top |
|
|
|
|
