Text record

This GOFF definition describes three basic mechanisms or styles of supplying text information:

Byte-oriented data is an unstructured stream of bytes to be placed at a specified offset ("address", for Assembler users) within a designated element of the program object.
Byte-oriented data is typified by machine language instructions and data (as in current OBJ TXT records). (These can be thought of as analogous to "U-format" inputs.)
Structured-record data is in the form of fixed-length records whose internal structure is known to and defined by the program management binder, and which will be placed in the C'B_IDRL' class. (These can be thought of as analogous to "F-format" inputs.)
The only currently acceptable form is 19-byte IDR data. (See Figure 1.) Data records of any other length or content must be defined by the program management binder.

An example of structured-record data is current OBJ IDR data, which is in the form of one or two 19-byte structures on an OBJ END record.
Unstructured-record data is in the form of records whose internal structure is (and will remain) unknown to the program management binder, which will simply append such records (prefixed by a length field) at the current end of the designated element. The length of each such record must be supplied by its provider. (These can be thought of as analogous to "V-format" inputs.)
Examples of unstructured-record data are current OBJ SYM data and GOFF ADATA records.

A text record has the following format:

Table 1. Text record
Field	Offset	Type	Description
PTV	0-2	Byte(3)	Type of record X'031000' If not continued X'031100' If continued
	3.0-3	Bit(4)	Reserved.
Text Record Style	3.4-7	Bit(4)	Designates the style of text data on this record: `B'0000'` Byte-oriented data. The "Address" field provides the position where the data bytes are to be placed. `B'0001'` Structured-record data Note: All structured-record data in a text record must have the same format. `B'0010'` Unstructured-record data All other values are reserved. Note: This field replicates a similar field in the element definition, so that text records can be checked for consistency.
Element ESDID	4-7	Binary(4)	ESDID of the element or part to which the data on this (and any subsequent continuation) records belongs. Note: Since the ED or PR record contains both the class name and the ESDID of the Section Definition (SD) record, this field uniquely identifies the program object element to which this data belongs.
	8-11	Binary(4)	Reserved.
Offset	12-15	Binary(4)	Starting offset from the element or part origin of the text on this (and any subsequent continuation) records. This field must be zero for structured-record and unstructured-record styles of data. Note: The offset for Assembler-produced text can be relative to a nonzero element origin.
Text Field True Length	16-19	Binary(4)	If the Text Encoding Type in the following field is zero, this field must be zero. If the Text Encoding Type is nonzero, this field specifies the length of the text after expansion.
Text Encoding	20-21	Binary(2)	If the text on this record is not compressed or not encoded, this field must be zero. A nonzero value indicates that the data is encoded, and will require decoding, expansion, or other treatment. See Text encoding and compression.
Data Length	22-23	Binary(2)	Total length in bytes of the data on this (and any following continuation) record. This length is unsigned and cannot be zero.
Data	24-*	Byte(n)	The next `n` bytes of data. Note: The IDR data must follow conventions and rules for translator-produced IDR data. (See Figure 1.) The format for emitting IDR data is given in Identification record data field below.
Trailer		Byte(m)	Unused space at the end of a record is reserved, and cannot be used for any other purpose.