| Author |
Message
|
| pcelari |
 Posted: Mon Jul 24, 2006 9:56 am Post subject: XML to CWF from C header - end of string question |
 |
|
Chevalier
Joined: 31 Mar 2006
Posts: 411
Location: New York
|
it must be a trivial question.
I have C structure like this:
struct Order {
char ONumber[11];
char ItemId[13];
};
under C, the last char is always reserved for the '\0'. However, I need the end msg to be fix length string of 23-char long, with 0-9 for ONum, and 10-21 for ItemId, followed by a single '\0' at the end.
Do I need to alter the length of each item after importing the structure?
thanks a lot.
_________________
pcelari
-----------------------------------------
- a master of always being a newbie |
|
| Back to top |
|
 |
| kimbert |
| Posted: Mon Jul 24, 2006 11:45 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
| Quote: |
| I need the end msg to be fix length string of 23-char long, with 0-9 for ONum, and 10-21 for ItemId, followed by a single '\0' at the end. |
It's very rare to find a fixed-length field which is null-terminated. Are you sure that this is what you want to do?
| Quote: |
| Do I need to alter the length of each item after importing the structure? |
Your problem is bigger than that, I think. If you define these strings as 'fixed-length' the CWF writer will not null-terminate the second string. (Why would it?). If you want a fixed-length field with a \0 at the end, you will have to insert the \0 yourself.
I'll leave you to do the byte-counting yourself. Or you could always just give it a try and see what happens  |
|
| Back to top |
|
|
| pcelari |
| Posted: Mon Jul 24, 2006 12:21 pm Post subject: |
|
|
Chevalier
Joined: 31 Mar 2006
Posts: 411
Location: New York
|
thanks for the response. I think I should include comment to my structure:
struct Order {
char ONumber[11]; /* 10-digit OrderNumber + 1 for terminating null */
char ItemId[13]; /* 12-digit ItemId + 1-char for terminating null */
};
The protocol for the msg is based on this structure, but without the '\0' in between. Otherwise, the recipient won't receive the whole msg rather only the first one and treat the msg as 10-char long when it read the '\0' at position 11.
an example of such a msg:
"1234567890ABCDEFGHIJKL'\0'"
| Quote: |
| If you want a fixed-length field with a \0 at the end, you will have to insert the \0 yourself. |
You're right. I can always add one-character filler to the CWF msg and set it to be a fixed '\0'.
_________________
pcelari
-----------------------------------------
- a master of always being a newbie |
|
| Back to top |
|
|
| pcelari |
| Posted: Mon Jul 24, 2006 12:27 pm Post subject: |
|
|
Chevalier
Joined: 31 Mar 2006
Posts: 411
Location: New York
|
and it would also mean I'll need to edit the length field in the msg set editor to reduce the length of each element by one, and add one field. This corresponds to importing the following altered structure:
struct Order {
char ONumber[10]; /* 10-digit OrderNumber, no terminating null */
char ItemId[12]; /* 12-digit ItemId, no terminating null */
char filler[1]; /*** only for terminating null ***/
};
Am I right? Any other input?
_________________
pcelari
-----------------------------------------
- a master of always being a newbie |
|
| Back to top |
|
|
| kimbert |
| Posted: Mon Jul 24, 2006 1:39 pm Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
| Quote: |
| You're right. I can always add one-character filler to the CWF msg and set it to be a fixed '\0' |
If you feel brave you could use the TDS format instead, and define a Group Terminator of \0 for the structure. That way, you don't have to write any ESQL, and all the information about your message structure is in your message set ( which is a Very Good Idea ). |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Mon Jul 24, 2006 1:58 pm Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
pcelari might run into some issues using TDS if all of his data is not represented as strings of some kind or another.
pcelari - can you be a bit more specific about the source application and the destination application? Are you trying to process this message as an input, or produce it as an output, or both? Is the structure the same on both sides, with some differences in things like null-terminators - or is the structure completely different?
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| pcelari |
| Posted: Tue Jul 25, 2006 9:31 am Post subject: |
|
|
Chevalier
Joined: 31 Mar 2006
Posts: 411
Location: New York
|
| jefflowrey wrote: |
pcelari might run into some issues using TDS if all of his data is not represented as strings of some kind or another.
pcelari - can you be a bit more specific about the source application and the destination application? Are you trying to process this message as an input, or produce it as an output, or both? Is the structure the same on both sides, with some differences in things like null-terminators - or is the structure completely different? |
Yes, they are all strings. the CWF from C structure is at the legacy end. the input msg is MRM XML. What I'm doing is to use a Mapping Node to perform the mapping, while let the CWF to construct the physical byte stream, plus the terminating zero.
_________________
pcelari
-----------------------------------------
- a master of always being a newbie |
|
| Back to top |
|
|
| kimbert |
| Posted: Tue Jul 25, 2006 2:09 pm Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
| You can get away without the char filler[1]. Just set the Physical Type of ItemId to 'null-terminated string' and the CWF writer will put in the \0 for you. |
|
| Back to top |
|
|
|
|
