| Author |
Message
|
| jefflowrey |
 Posted: Tue May 04, 2004 11:30 am Post subject: |
 |
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
Your input message should be
MSG_CONTREPEAT_SEGEL_1AAEL_2BB
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
 |
| kimbert |
| Posted: Wed May 05, 2004 12:29 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5543
Location: Southampton
|
Your input TDS bitstream does not contain tags, so why are you providing a TDS tag length? (and why are you setting the DES to tagged fixed length?).
Your XML output will contain tags, but those tags will be taken from the names of the elements in the tree (the XML domain has to work that way because it does not use the message model).
Assuming that your example input bitstream is correct, your innermost complex type should have DES of 'Fixed Length'.
By the way, did you try my previous suggestion? Did it work? |
|
| Back to top |
|
|
| sebastian |
| Posted: Wed May 05, 2004 4:04 am Post subject: |
|
|
Centurion
Joined: 12 Nov 2003
Posts: 110
Location: Philadelphia
|
You were right with your previous suggestion. I was confused on how to represent the repeating elements and thought they needed to be defined individually instead of as a single repeating element pointing to a compound type that contain elements.
MSG_CONTREPEAT_SEGAABB
The input TDS bitstream is correct, I just did not understand how to define it correctly. The individual fields will not have a tag by their data so I changed the DES of REPEAT_SEG to fixed length and removed the tag length.
I am now getting a message that there is a missing group indicator in the bit stream.
"( PRWAPMQADM01DBK.seb_ex_group ) Group Indicator missing in the bitstream
The TDF parser could not parse the data in the bitstream for Fixed Length compound type, %%1, no Group Indicator, %%2, was present in the bitstream at offset %%3.
The message model definition does not match the bitstream. Check that your model definition is correct and if so, update your bitstream data accordingly. "
This is a bit confusing since I only have two group indicators and I can see them both in the input data.
I appreciate any suggestions....I really think I am close to getting this
Sebastian
_________________
sebastian signature |
|
| Back to top |
|
|
| kimbert |
| Posted: Wed May 05, 2004 6:21 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5543
Location: Southampton
|
| Quote: |
| MSG_CONTREPEAT_SEGAABB |
Your message should have a TDS tag of 'MSG_CONT' (or a group indicator may work, but I think a tag is simpler and more obvious).
The message's complex type should have DES set to 'Tagged Fixed Length', and a Tag Length of 10. (for parsing the 'REPEAT_SEG' tag described below).
It should contain a single element called 'subsection1'. This element should have a tag of 'REPEAT_SEG'. The element's type ( REPEAT_SEG_TYPE, perhaps ) should have DES of Fixed Length. It should contain two elements, as you described in your earlier post.
Note that I have avoided the use of Group Indicators here, and replaced them with tags. Wherever a piece of text precedes some data, a tag is the most natural way to model it. Group Indicator/Group Terminator are only truly necessary when you have an embedded compound type (so there's no element to hold the tag definition) or when you need to model some trailing text (like CR/LF at the end of a line). |
|
| Back to top |
|
|
| sebastian |
| Posted: Wed May 05, 2004 7:08 am Post subject: |
|
|
Centurion
Joined: 12 Nov 2003
Posts: 110
Location: Philadelphia
|
I apologize for seeming dense here but I am not clear on where the TDS tag for MSG_CONT would be.
I have taken a crude shot at a message set visual reproduction here:
Messages
MSG_CONT
Subsection1 REPEAT_SEG
EL_1 STRING
EL_2 STRING
Types
(compound) MSG_CONTAINER
subsection1 REPEAT_SEG
EL_1 STRING
EL_2 STRING
(compound) REPEAT_SEG
EL_1 STRING
EL_2 STRING
Are you saying that under the messages folder, for the first message icon of 'MSG_CONT' that the message key in the TDS tab should be defined as MSG_CONT?
Under my types folder, in the MSG_CONTAINER compound type, my group indicator is MSG_CONT but I do not see where I could define a tag for this. I am now wondering if my message set definition needs to be changed.
I can email you the message set exort if you have the time to look at it.
thanks,
Seb
_________________
sebastian signature |
|
| Back to top |
|
|
| kimbert |
| Posted: Thu May 06, 2004 12:48 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5543
Location: Southampton
|
Tags are associated with elements, not with compound types. To set a tag, just click on an element or a message, and go to the TDS property page.
You seem to be confusing elements and types. Elements are named pieces of data which become nodes in the tree. The type of an element describes its value (if it is a simple element) or its structure.
If you don't understand this, you should re-read the Working with Messages manual.
You should certainly not be setting the message key - that is used to resolve multipart messages.
You do not need two compound types ( or are they elements? ) with identical members ('subsection1' and 'compound'). |
|
| Back to top |
|
|
| sebastian |
| Posted: Fri May 07, 2004 6:00 am Post subject: |
|
|
Centurion
Joined: 12 Nov 2003
Posts: 110
Location: Philadelphia
|
I think this discussion has drifted too far off from the original subject definition.
I will open a new subject called TDS modeling - repeating segments.
Seb[/img][/url][/quote]
_________________
sebastian signature |
|
| Back to top |
|
|
|
|
