Asian rules for text formatting

Some characters should not be placed at the beginning of a line, and some should not be placed at the end of a line. These beginning-inhibited and ending-inhibited characters are different among the languages but the required process is the same. Thus, ISPF uses the same text formatting process for these Asian languages, but uses a different beginning-and-ending-inhibitor character table for each of the languages.

The text is first split into words. An SBCS word is delimited by blanks, or SO/SI characters. Then any beginning inhibitors are stripped from the beginning of the word and treated as separate words, and any ending inhibitors are stripped from the end of the word and treated as separate words.

Adjoining DBCS alphanumeric characters (that is, Ward 42 characters) are treated as one DBCS word. Then any beginning inhibitors are stripped from the beginning of the word and treated as separate words, and any ending inhibitors are stripped from the end of the word and treated as separate words. All other non-Ward 42 double-byte characters are treated as separate DBCS words.

If a word exceeds the available panel space, then the word splits and continues on the next line. If the text consists of mixed data and does not fit in one line within the specified width, the first position is always reserved for a SO character (if first word is double-byte) or for a blank (if the first word is single-byte). This allows the text to be aligned properly.

Words that exceed the width of the available panel space are wrapped to the next line according to following rules:

Figure 1. Text formatting rules

This table shows what happens with word overflow.

Where:

CE-1 and CE: Last two words that fit on line
CB and CB+1: First two words on next line
E: Ending inhibitor
B: Beginning inhibitor
X: Neither
Forward: Move CE to next line
Backward: Move CB to previous line
No process: Split as is

Note: If words CE or CB are single-byte words and are more than 1 character, or if CE or CB are double-byte words and are more than 1 double-byte character, then no special processing is used; the line is split as is.

When your panel contains several successive lines of mixed data from different tags, the alignment of a short text string can appear to be shifted 1 byte further left than the surrounding text. This occurs because a text string that fits on one line does not have the leading position reserved for the SO character to use as many positions on the screen as possible.

You can control the alignment of successive lines of mixed data by adding a string of DBCS blanks to the end of a short text string. This forces the SO character position to be reserved during formatting.

SBCS and DBCS blanks that end or begin a line are deleted.