Merge branch 'gawk-4.1-stable'

12 files changed, 79 insertions, 10 deletions

diff --git a/doc/ChangeLog b/doc/ChangeLog
index 3385ba09..c1754161 100644
--- a/doc/ChangeLog
+++ b/doc/ChangeLog
@@ -1,3 +1,8 @@
+2013-06-27         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* texinfo.tex: Update from Karl, fixes a formating problem.
+	* gawktexi.in (Conversions): Undo @w{} around @option{--posix}.
+
 2013-06-22         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* gawktexi.in (Type Functions): Add more explanation to isarray(),
diff --git a/doc/gawk.texi b/doc/gawk.texi
index c9211007..fa3e5871 100644
--- a/doc/gawk.texi
+++ b/doc/gawk.texi
@@ -10121,7 +10121,7 @@ point, so the default behavior was restored to use a period as the
 decimal point character.  You can use the @option{--use-lc-numeric}
 option (@pxref{Options}) to force @command{gawk} to use the locale's
 decimal point character.  (@command{gawk} also uses the locale's decimal
-point character when in POSIX mode, either via @w{@option{--posix}}, or the
+point character when in POSIX mode, either via @option{--posix}, or the
 @env{POSIXLY_CORRECT} environment variable, as shown previously.)
 
 @ref{table-locale-affects} describes the cases in which the locale's decimal
diff --git a/doc/gawktexi.in b/doc/gawktexi.in
index 68316d1c..59ee1a69 100644
--- a/doc/gawktexi.in
+++ b/doc/gawktexi.in
@@ -9618,7 +9618,7 @@ point, so the default behavior was restored to use a period as the
 decimal point character.  You can use the @option{--use-lc-numeric}
 option (@pxref{Options}) to force @command{gawk} to use the locale's
 decimal point character.  (@command{gawk} also uses the locale's decimal
-point character when in POSIX mode, either via @w{@option{--posix}}, or the
+point character when in POSIX mode, either via @option{--posix}, or the
 @env{POSIXLY_CORRECT} environment variable, as shown previously.)
 
 @ref{table-locale-affects} describes the cases in which the locale's decimal
diff --git a/doc/texinfo.tex b/doc/texinfo.tex
index a0c9d080..9cf29f2e 100644
--- a/doc/texinfo.tex
+++ b/doc/texinfo.tex
@@ -3,7 +3,7 @@
 % Load plain if necessary, i.e., if running under initex.
 \expandafter\ifx\csname fmtname\endcsname\relax\input plain\fi
 %
-\def\texinfoversion{2013-03-19.11}
+\def\texinfoversion{2013-06-21.17}
 %
 % Copyright 1985, 1986, 1988, 1990, 1991, 1992, 1993, 1994, 1995,
 % 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
@@ -2475,14 +2475,14 @@ end
 }
 
 % We *must* turn on hyphenation at `-' and `_' in @code.
+% (But see \codedashfinish below.)
 % Otherwise, it is too hard to avoid overfull hboxes
 % in the Emacs manual, the Library manual, etc.
 %
 % Unfortunately, TeX uses one parameter (\hyphenchar) to control
 % both hyphenation at - and hyphenation within words.
 % We must therefore turn them both off (\tclose does that)
-% and arrange explicitly to hyphenate at a dash.
-%  -- rms.
+% and arrange explicitly to hyphenate at a dash. -- rms.
 {
   \catcode`\-=\active \catcode`\_=\active
   \catcode`\'=\active \catcode`\`=\active
@@ -2499,14 +2499,33 @@ end
      \let-\normaldash
      \let_\realunder
     \fi
+    % Given -foo (with a single dash), we do not want to allow a break
+    % after the hyphen.
+    \global\let\codedashprev=\codedash
+    %
     \codex
   }
+  %
+  \gdef\codedash{\futurelet\next\codedashfinish}
+  \gdef\codedashfinish{%
+    \normaldash % always output the dash character itself.
+    % 
+    % Now, output a discretionary to allow a line break, unless
+    % (a) the next character is a -, or
+    % (b) the preceding character is a -.
+    % E.g., given --posix, we do not want to allow a break after either -.
+    % Given --foo-bar, we do want to allow a break between the - and the b.
+    \ifx\next\codedash \else
+      \ifx\codedashprev\codedash 
+      \else \discretionary{}{}{}\fi
+    \fi
+    \global\let\codedashprev=\next
+  }
 }
-
+\def\normaldash{-}
+%
 \def\codex #1{\tclose{#1}\endgroup}
 
-\def\normaldash{-}
-\def\codedash{-\discretionary{}{}{}}
 \def\codeunder{%
   % this is all so @math{@code{var_name}+1} can work.  In math mode, _
   % is "active" (mathcode"8000) and \normalunderscore (or \char95, etc.)
diff --git a/helpers/ChangeLog b/helpers/ChangeLog
new file mode 100644
index 00000000..4b720787
--- /dev/null
+++ b/helpers/ChangeLog
@@ -0,0 +1,12 @@
+2013-06-27         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* ChangeLog: Created.
+	* testdfa.c: Add ifdef around xalloc.h, for use with grep dfa.
+
+2013-06-03         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* testdfa.c: New file.
+
+2013-06-03         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* testnet.awk, testnet.c, fixdump.awk: New files.
diff --git a/helpers/testdfa.c b/helpers/testdfa.c
index 653a28f8..c644fb04 100644
--- a/helpers/testdfa.c
+++ b/helpers/testdfa.c
@@ -732,6 +732,7 @@ char casetable[] = {
 
 #undef C
 
+#ifdef GREP_DFA	/* not needed for gawk */
 /* xalloc.h -- malloc with out-of-memory checking
 
    Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
@@ -1069,3 +1070,4 @@ xmemdup (T const *p, size_t s)
 
 
 #endif /* !XALLOC_H_ */
+#endif /* GREP_DFA */
diff --git a/test/ChangeLog b/test/ChangeLog
index 2eada5aa..fcd922c1 100644
--- a/test/ChangeLog
+++ b/test/ChangeLog
@@ -1,3 +1,9 @@
+2013-06-27         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* Makefile.am (dfamb1): New test.
+	* dfamb1.awk, dfamb1.in, dfamb1.ok: New files.
+	Test case from Steven Daniels <stevendaniels88@gmail.com>.
+
 2013-06-24         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* Makefile.am (clos1way): Move to here since Maketests gets
diff --git a/test/Makefile.am b/test/Makefile.am
index fb73a76f..fe6d3a59 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -177,6 +177,9 @@ EXTRA_DIST = \
 	devfd1.awk \
 	devfd1.ok \
 	devfd2.ok \
+	dfamb1.awk \
+	dfamb1.in \
+	dfamb1.ok \
 	dfastress.awk \
 	dfastress.ok \
 	double1.awk \
@@ -925,7 +928,7 @@ BASIC_TESTS = \
 	back89 backgsub \
 	childin clobber closebad clsflnam compare compare2 concat1 concat2 \
 	concat3 concat4 convfmt \
-	datanonl defref delargv delarpm2 delarprm delfunc dfastress dynlj \
+	datanonl defref delargv delarpm2 delarprm delfunc dfamb1 dfastress dynlj \
 	eofsplit exit2 exitval1 exitval2 \
 	fcall_exit fcall_exit2 fldchg fldchgnf fnamedat fnarray fnarray2 \
 	fnaryscl fnasgnm fnmisc fordel forref forsimp fsbs fsrs fsspcoln \
@@ -1838,6 +1841,12 @@ clos1way:
 	@AWKPATH=$(srcdir) LC_ALL=C $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
 
+dfamb1:
+	@echo $@
+	@[ -z "$$GAWKLOCALE" ] && GAWKLOCALE=en_US.UTF-8; \
+	AWKPATH=$(srcdir) $(AWK) -f $@.awk $(srcdir)/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
+
 # Targets generated for other tests:
 include Maketests
 
diff --git a/test/Makefile.in b/test/Makefile.in
index 5bb3e201..06946c0c 100644
--- a/test/Makefile.in
+++ b/test/Makefile.in
@@ -395,6 +395,9 @@ EXTRA_DIST = \
 	devfd1.awk \
 	devfd1.ok \
 	devfd2.ok \
+	dfamb1.awk \
+	dfamb1.in \
+	dfamb1.ok \
 	dfastress.awk \
 	dfastress.ok \
 	double1.awk \
@@ -1142,7 +1145,7 @@ BASIC_TESTS = \
 	back89 backgsub \
 	childin clobber closebad clsflnam compare compare2 concat1 concat2 \
 	concat3 concat4 convfmt \
-	datanonl defref delargv delarpm2 delarprm delfunc dfastress dynlj \
+	datanonl defref delargv delarpm2 delarprm delfunc dfamb1 dfastress dynlj \
 	eofsplit exit2 exitval1 exitval2 \
 	fcall_exit fcall_exit2 fldchg fldchgnf fnamedat fnarray fnarray2 \
 	fnaryscl fnasgnm fnmisc fordel forref forsimp fsbs fsrs fsspcoln \
@@ -2231,6 +2234,12 @@ clos1way:
 	@echo $@
 	@AWKPATH=$(srcdir) LC_ALL=C $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
+
+dfamb1:
+	@echo $@
+	@[ -z "$$GAWKLOCALE" ] && GAWKLOCALE=en_US.UTF-8; \
+	AWKPATH=$(srcdir) $(AWK) -f $@.awk $(srcdir)/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
 Gt-dummy:
 # file Maketests, generated from Makefile.am by the Gentests program
 addcomma:
diff --git a/test/dfamb1.awk b/test/dfamb1.awk
new file mode 100644
index 00000000..7647adf5
--- /dev/null
+++ b/test/dfamb1.awk
@@ -0,0 +1,4 @@
+# The first of these used to core dump gawk. As of 7/2013 it's fixed.
+# This file is to make sure that remains true.
+{match($0, /(([^ \?.]*\?pos=ad |([^ \?.]*\?pos=(jj|va) )地\?pos=dev ){0,2})/ , arr)}  { if(arr[0]) print arr[1], arr[4], $6}
+{match($0, /(([^ \?.]*\?pos=ad |([^ \?.]*\?pos=(jj|va) )[地]\?pos=dev ){0,2})/ , arr)}  { if(arr[0]) print arr[1], arr[4], $6} 
diff --git a/test/dfamb1.in b/test/dfamb1.in
new file mode 100644
index 00000000..8347590f
--- /dev/null
+++ b/test/dfamb1.in
@@ -0,0 +1 @@
+很?pos=ad 宽广?pos=va ， 更?pos=ad 是?pos=vc1
diff --git a/test/dfamb1.ok b/test/dfamb1.ok
new file mode 100644
index 00000000..ad443a23
--- /dev/null
+++ b/test/dfamb1.ok
@@ -0,0 +1,2 @@
+很?pos=ad   
+很?pos=ad