Record Parsers
A record parser breaks a flat file into individual records. When defining a flat file schema, you can specify one of the following record parsers:
Delimited Record Parser. This parser expects a specific delimiter to indicate the end of a record. For a record delimiter, you can specify:
A character (for example, !) or character representation (for example,
\r\n for carriage return)
Hexadecimal value (for example,
0X09)
Octal value (for example,
009)
Unicode characters (for example,
\u XXXX, where
XXXX represents the unicode value of the character)
The occurrence of the record delimiter signals the end of one record and the beginning of the next record.
Fixed Length Record Parser. This parser splits a file into records of the same pre-specified length.
Note:
This parser measures the lengths and positions of records in terms of bytes, rather than characters. Prior to 6.5, Service Pack 2, the parser measured lengths and positions in terms of characters. This change does not affect users who currently parse/compose flat files using single-byte encodings because one byte equals one character. In this case, there is no functional difference between specifying bytes or characters.
The parser supports both single-byte and multi-byte encodings. Multi-byte encoded files must run on JVM version 1.4 or later. (The behavior of delimited fields and records remain the same.) For details, see
Flat File Byte Count Parser.
Variable Length Record Parser. This parser expects each record to be preceded by two bytes that indicate the length of the record. Each record may be a different length.
EDI Document Type Record Parser. This parser is used only for EDI flat files and provides additional functionality needed to properly parse EDI documents. Documents using this record parser should be created using the
webMethods Module for EDI, not the
WmFlatFile package. For more information, see the
webMethods Module for EDI Installation and User’s Guide .