The use of this segment is not well understand but there are some facts about it:

A-segment. Item properties.

This segment contains the description of the properties for each item in the form. Notice that some properties need a "link" to others segment (eg. "Initial value" to "K-segment").

Initial values for each item are stored (if needed) in a structure separated from others by character 0x03.

Each structure contains this information:

• Character 0x04.
• Length of the string used to assign the initial value to the item. Notice that length is "in charHex code" (see Numerical constants). Notice than length is not physical because "escaped" character as %%^ (0x07 0x2D) is count as 1 char.
• Length and "string" separator (char 0x06).
• Character 0x03 finishing the structure

All objects and their properties are enumerated here.

The structure of this segment consists in a set of values enclosed by the character 0x03. Between two 0x03 we will found the item "in charHex code" (see Numerical constants), the character 0x06 and the item name. Notice than length is not physical because "escaped" character as %%^ (0x07 0x2D) is count as 1 char.

0x03 0x37 0x06 0x45 0x4E 0x54 0x41 0x54 0x03 is the definition for an object called "ENTITAT".

The object is enumerated according to the position in the component (left to right and up to down).

The component variables are defined in this segment in the same way than other objects.

The P-segment (proc code segment) use, basically two characters as separator between the different structures. The first character of this segment is the character "P"

The character 0xC8 indicates that the following characters, until the end of the token, are a constant string.

The substitution (it means something like "The current date is %$DATE$ ") is codified using, in the position where the substitution must be done, the character 0x09 followed by the variable address (2 bytes) and type (1 byte).

Notice that LIST are stored as normal string (the only difference is that it can contain <GOLD => and <GOLD ;>

0x10: Group wildcard (<GOLD *> or *)

0x11: Single wildcard (<GOLD ?> or?)

0x12: Value and representation separator (<GOLD => or ?)

0x1B: List item separator (<GOLD ;> or ;)

Carriage Return %%^ (0x07 0x2D)

Numeric constants are codified in different ways (depending on their values). The first char indicates how is stored and it can be:

Global registers are codified using codes for small numbers (see table below) without 0xFD

The first BCB we always found after the segment identification (the character "P") is composed by several tokens:

First: The value 0x06.suspicion -"the classical 1 done line at the beginning of the proc listing" or maybe is the BCB type).

Following this BCB, the next ones correspond with any trigger/entry (sentence) or parameter/variable declaration (data definition). The BCB for parameter/variable declaration for a trigger/entry is stored next to the "sentences BCB".

Each trigger contains an automatically generated declaration of each local ($V$) and global ($$V) used in the proc code for this trigger. The declaration (enumeration may be better) consists in the two token: name and "visibility". The visibility can be "0x0C" for global variables and "0x0D" for locals.

The parameter definition has this token sequence:

Direction and type The first (and maybe second – see * in the table) byte. This byte is calculated assigning a value depending on the parameter type plus and value according to the direction. See the table below:

(*) The value is stored using the "Low Numbers" method exposed in the "Numeric constants" section. Then the types marked with "*" in the "in" direction are codified as: 0x07 0x28; 0x07 0x29; 0x07 0x2A; 0x07 0x2B; 0x07 0x2D

End of the parameter. next token :0x0E

Variables are codified exactly as " IN parameters" (including raw, Boolean, lineardate, lineartime, lineardatetime and image). The only difference is that next token "end of variable" is 0x0F (instead "0x0E" in parameters).

This segment contains a list with de window properties (each delimited by 0x1b). These items are:

The rest of properties (More… command button) are added as other items at end of this list.

This segment contains the list of operations available in the component.

This segment contain the character 0x00 repeated until the component reached a 512 multiple size.

Anyway, remember that a trigger can appears in a BCB or not depending on its functionality (eg. A LPMX can not be generate a Yxx code because it is not a piece of code, it contains "entries" and these entries are able of be generated).

The numeration of these triggers has a particularity: If the trigger "Operation" is defined the Y-code specified is correct but if there are not OPER trigger then the code for triggers 7 to 17 must be decreased in 1 (CLR becomes 6, FRLF 10, and so on).

See the note and example about numbering entities and field triggers at the end of this appendix

See the note and example about numbering entities and field triggers at the end of this appendix

For entities and fields the specified code is the "Offset" you must to add to the "object number". Think that a gap of twenty codes are reserved for every object", and that the codes 1 to 19 are for the component level triggers.

Example: Consider a Form with 3 entities (with the very original name ENTA, ENTB and ENTC). ENTA has two field FLDA_1 and FLDA_2; ENTB has only the FLDB_1 field and FLDC_1 and FLDC_2 are ENTC fields.