Improve the texnl macro and its use.

3 files changed, 16 insertions, 13 deletions

diff --git a/doc/libunistring.texi b/doc/libunistring.texi
index 6818d57..aa09f49 100644
--- a/doc/libunistring.texi
+++ b/doc/libunistring.texi
@@ -54,16 +54,19 @@
 @code{\func\}
 @end macro
 
-@c Macro for forcing a line break in TeX mode.
+@c Macro for an advisory ragged line break in TeX mode.
 @c Needed because there are long unbreakable pieces of text (such as URLs or
 @c formulas), TeX is too shy to move them to a new line. TeX considers only
 @c two choices: a line break in aligned mode (which it rejects due to aesthetic
 @c reasons) and writing into the margin. What we want in many cases is a line
-@c break without filling the first line. Like what @* delivers.
-@c But we need it only in TeX mode. info and HTML modes are fine.
+@c break without filling the first line. Like what @* delivers. But we want it
+@c only when needed, so that it disappears when unrelated changes in the same
+@c paragraph cause a line break in a nearby position. And we need it only in
+@c TeX mode. info and HTML modes are fine.
+@c This trick is from Karl Berry.
 @iftex
 @macro texnl
-@*
+@hfil@penalty9000@hfilneg
 @end macro
 @end iftex
 @ifnottex
@@ -348,13 +351,13 @@ Japanese, even the processing of pure Japanese text was error prone.
 References:
 @itemize @bullet
 @item
-The Unicode standard: @url{http://www.unicode.org/}
+The Unicode standard:@texnl{} @url{http://www.unicode.org/}
 @item
-Definition of UTF-8: @url{http://www.rfc-editor.org/rfc/rfc3629.txt}
+Definition of UTF-8:@texnl{} @url{http://www.rfc-editor.org/rfc/rfc3629.txt}
 @item
-Definition of UTF-16: @url{http://www.rfc-editor.org/rfc/rfc2781.txt}
+Definition of UTF-16:@texnl{} @url{http://www.rfc-editor.org/rfc/rfc2781.txt}
 @item
-Markus Kuhn's UTF-8 and Unicode FAQ:@texnl
+Markus Kuhn's UTF-8 and Unicode FAQ:@texnl{}
 @url{http://www.cl.cam.ac.uk/~mgk25/unicode.html}
 @end itemize
 
@@ -849,7 +852,7 @@ that uses @code{libunistring} was installed.
 
 @cindex autoconf macro
 GNU Gnulib provides an autoconf macro that test for the availability
-of @code{libunistring}.  It is contained in the Gnulib module@texnl
+of @code{libunistring}.  It is contained in the Gnulib module@texnl{}
 @url{http://www.gnu.org/software/gnulib/MODULES.html#module=libunistring}.
 
 @amindex gl_LIBUNISTRING
diff --git a/doc/unictype.texi b/doc/unictype.texi
index cee4f45..a154ead 100644
--- a/doc/unictype.texi
+++ b/doc/unictype.texi
@@ -377,7 +377,7 @@ combining classes are attached "first" or "closer" to the base character.
 
 The canonical combining class of a character is a number in the range
 0..255.  The possible values are described in the Unicode Character Database
-@texnl @url{http://www.unicode.org/Public/UNIDATA/UCD.html}.  The list here is
+@texnl{}@url{http://www.unicode.org/Public/UNIDATA/UCD.html}.  The list here is
 not definitive; more values can be added in future versions.
 
 @deftypevr Constant int UC_CCC_NR
@@ -471,7 +471,7 @@ Returns the canonical combining class of a Unicode character.
 Every Unicode character or code point has a @emph{bidirectional category}
 assigned to it.
 
-The bidirectional category guides the bidirectional algorithm@texnl
+The bidirectional category guides the bidirectional algorithm@texnl{}
 (@url{http://www.unicode.org/reports/tr9/}).  The possible values are
 the following.
 
diff --git a/doc/uniwbrk.texi b/doc/uniwbrk.texi
index ea0ec89..6f06b92 100644
--- a/doc/uniwbrk.texi
+++ b/doc/uniwbrk.texi
@@ -43,8 +43,8 @@ word break to be present at the beginning of the string (before
 
 This is a more low-level API.  The word break property is a property defined
 in Unicode Standard Annex #29, section ``Word Boundaries'', see
-@url{http://www.unicode.org/reports/tr29/#Word_Boundaries}.@texnl  It is used
-for determining the word breaks in a string.
+@url{http://www.unicode.org/reports/tr29/#Word_Boundaries}.@texnl{}  It is
+used for determining the word breaks in a string.
 
 The following are the possible values of the word break property.  More values
 may be added in the future.