Cyril,
Finding your line breaks in gentext
may not be all that easy. It looks like it may be the processing
instruction _subchars, but if you look the attached, you will see this
may not always hold true.
I am attaching a screen shot of some
gentext and then using the oid_gentext() (see help 5708), the markup around
the gentext, a screen shot of the published output and the FOSI fragment
that generated this gentext.
FOSI fragment for the <prtitle>
element
<e-i-c
gi="prtitle"
context="* frntcover">
<charlist
envname="tmcoverpgenv">
<ruling
thick="1pt"
lentype="rel"
rellen="col"
voffset="-10pt"
placemnt="before"
type="single">
<presp
minimum="0"
nominal="0"
maximum="0"></ruling>
<usetext
source="\TECHNICAL
MANUAL\" placemnt="before">
<subchars>
<font
style="sanserif"
size="14pt"
weight="bold">
<leading
lead="16pt">
<quadding
quad="center"
lastquad="lcenter">
<presp
minimum="20"
nominal="24"
maximum="36"
priority="force">
<postsp
minimum="12pt"
nominal="16pt"
maximum="24pt"
priority="high">
<textbrk
startln="1"></subchars>
</usetext>
<usetext
source="typelvltitle"
placemnt="before">
<subchars>
<font
style="sanserif"
size="14pt"
weight="bold">
<leading
lead="16pt">
<hyphen
hyph="0">
<quadding
quad="center"
lastquad="lcenter">
<textbrk
startln="1"
endln="1"></subchars>
</usetext>
<usetext
source="\ FOR\"
placemnt="before">
<subchars>
<font
style="sanserif"
size="14pt"
weight="bold">
<leading
lead="16pt">
<quadding
quad="center"
lastquad="lcenter">
<postsp
minimum="20"
nominal="24"
maximum="36"
priority="force">
<textbrk
startln="1"
endln="1"></subchars>
</usetext>
</charlist>
</e-i-c>
This <att> is from another <e-i-c>,
but it sets the "typelvltitle" for the <usetext> in the
<prtitle> <e-i-c> The 'attval' is the 'maintlvls' attribute
you can see in the <paper.manual> tag in the screen capture.
<att>
<specval
attname="maintlvls"
attval="14">
<charsubset>
<savetext
textid="typelvltitle"
conrule="\OPERATOR'S,
UNIT, DIRECT SUPPORT, AND GENERAL SUPPORT MAINTENANCE MANUAL\"
append="0"></charsubset>
</att>
This <att> adds the "including
repair parts" line. It has a pseudo element for a new line.
<att>
<specval
attname="rpstl"
attval="#NONZERO">
<charsubset>
<savetext
textid="typelvltitle"
conrule="<newln>,\INCLUDING
REPAIR PARTS AND SPECIAL TOOLS LIST\,</newln>"
append="1"></charsubset>
</att>
Eval ouput for the oid_gentext of the
<prtitle>
offset="-1310720" rellen="0" len="0" ruleclr="default"?>
scid="56"?>TECHNICAL MANUAL
OPERATOR'S,
UNIT, DIRECT SUPPORT, AND GENERAL SUPPORT MAINTENANCE MANUAL
<newln?>INCLUDING REPAIR PARTS
AND SPECIAL TOOLS LIST
</newln?>
FOR/_subchars?>
What you need to notice is that where
I have a <textbrk startln="1"> or <textbrk endln="1">
(or both) in the FOSI, the line breaks are _subchar PIs. Now if you
look closely, you will notice that in some places the _subchars wraps
the text and in others, it either precedes or follows the text (this does
not appear in my screen shots, but you can do this with the set gentexttagdisplay
command set to full (see help 282)). This appears to be based on
how you set the 'startln' and 'endln' attributes in the <textbrk>.
If you have both set to 1, the _subchars wraps. If 'startln'
is set to 1, the _subchars precedes and if 'endln' is 1 then it will follow.
Notice that the published output has a line break in the "typelvltitle"
string that is not in the string or specified in the FOSI. This occurred
because of other formatting constraints.
So even if you succeed in finding your
line breaks in the edited gentext, there may be others that will appear
in the processed output. This is especially true if you have a long
generated text string.
Lynn