| Author |
Message |
    
Anonymous
| | Posted on Wednesday, Mar 3, 2004 - 14:20: | |
I have files with up to 25 different "record
types" per file a la:
rec_type_A.field1
rec_type_A.field2
rec_type_A.field3
rec_type_A.field4
rec_type_B.field1
rec_type_B.field2
rec_type_B.field3
rec_type_B.field4
rec_type_B.field5
rec_type_B.field6
...
if none of the record types fits into the
remaining BLOCK_NO*BLOCK_SIZE it is put in the
next block, and so forth.
It needs to recognize the "record types"
which can be done by the first 3 bytes of
each record type. This "key" consists of
2 bytes for the length of the record and
1 byte for the "type".
For example, the file starts with a header, so
I started with this template.
requires 0 "290000"
begin
// HEA
hex 2 "RECORD_LENGTH"
hex 1 "RECORD_TYPE"
hex 1 "CHARGING_BLOCK_SIZE"
hex 2 "TAPE_BLOCK_TYPE"
hex 2 "DATA_LENGTH_IN_BLOCK"
hex 10 "EXCHANGE_ID"
hex 4 "FIRST_RECORD_NUMBER"
hex 4 "BATCH_SEQUENCE_NUMBER"
hex 2 "BLOCK_SEQUENCE_NUMBER"
hex 3 "START_TIME"
hex 4 "START_DATE"
hex 6 "FORMAT_VERSION"
end
The record type following the header may be
any of the other 24 types. The "signature"
for the header is "290000" and the next
record has 33 fields and the "signature"
"880009".
How can I best reflect that in a template?
Thanks in advance. |
|