Merge branch 'gawk-4.1-stable'

14 files changed, 303 insertions, 155 deletions

diff --git a/ChangeLog b/ChangeLog
index 4ec71357..d9942268 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,17 @@
+2014-03-10         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* dfa.h, dfa.c: Sync with grep. Yet again.
+	* regex_internal.c (built_wcs_upper_buffer, build_upper_buffer):
+	Fixes from GNULIB for mixed case matching on Mac OS X.
+
+	Unrelated:
+
+	* builtin.c (format_tree): Smarten handling of %' flag. Always
+	pass it in for floating point formats. Then only add the
+	thousands_sep if there is one. Also, allow for thousands_sep
+	to be a string, not just one character.  Thanks to Michal Jaegermann
+	for the report.
+
 2014-03-08         Andrew J. Schorr     <aschorr@telemetry-investments.com>
 
 	* gawkapi.c (api_impl): Add memory allocation function pointers.
diff --git a/NEWS b/NEWS
index 8d919351..cf741b18 100644
--- a/NEWS
+++ b/NEWS
@@ -52,6 +52,11 @@ extension facility only works on Alpha and Itanium.
 realloc() and free(), to insure that the same memory allocation
 functions are always used. This bumps the minor version by one.
 
+13. The printf quote flag now works correctly in locales with a different
+decimal point character but without a thousands separator character.
+If the thousands separator is a string, it will be correctly added
+to decimal numbers.
+
 XXX. A number of bugs have been fixed. See the ChangeLog.
 
 Changes from 4.0.2 to 4.1.0
diff --git a/builtin.c b/builtin.c
index 8b1f9868..ebd0fc99 100644
--- a/builtin.c
+++ b/builtin.c
@@ -994,9 +994,7 @@ check_pos:
 			goto check_pos;
 		case '\'':
 #if defined(HAVE_LOCALE_H)       
-			/* allow quote_flag if there is a thousands separator. */
-			if (loc.thousands_sep[0] != '\0')
-				quote_flag = true;
+			quote_flag = true;
 			goto check_pos;
 #else
 			goto retry;  
@@ -1196,6 +1194,9 @@ out0:
 			}
 			if (i < 1)
 				goto out_of_range;
+#if defined(HAVE_LOCALE_H)
+			quote_flag = (quote_flag && loc.thousands_sep[0] != 0);
+#endif
 			chp = &cpbufs[1].buf[i-1];
 			ii = jj = 0;
 			do {
@@ -1203,8 +1204,14 @@ out0:
 				chp--; i--;
 #if defined(HAVE_LOCALE_H)
 				if (quote_flag && loc.grouping[ii] && ++jj == loc.grouping[ii]) {
-					if (i)	/* only add if more digits coming */
-						PREPEND(loc.thousands_sep[0]);	/* XXX - assumption it's one char */
+					if (i) {	/* only add if more digits coming */
+						int k;
+						const char *ts = loc.thousands_sep;
+
+						for (k = strlen(ts) - 1; k >= 0; k--) {
+							PREPEND(ts[k]);
+						}
+					}
 					if (loc.grouping[ii+1] == 0)
 						jj = 0;		/* keep using current val in loc.grouping[ii] */
 					else if (loc.grouping[ii+1] == CHAR_MAX)
@@ -1360,6 +1367,9 @@ mpf1:
 #ifdef HAVE_MPFR
 	int0:
 #endif
+#if defined(HAVE_LOCALE_H)
+			quote_flag = (quote_flag && loc.thousands_sep[0] != 0);
+#endif
 			/*
 			 * When to fill with zeroes is of course not simple.
 			 * First: No zero fill if left-justifying.
@@ -1378,8 +1388,14 @@ mpf1:
 				uval /= base;
 #if defined(HAVE_LOCALE_H)
 				if (base == 10 && quote_flag && loc.grouping[ii] && ++jj == loc.grouping[ii]) {
-					if (uval)	/* only add if more digits coming */
-						PREPEND(loc.thousands_sep[0]);	/* XXX --- assumption it's one char */
+					if (uval) {	/* only add if more digits coming */
+						int k;
+						const char *ts = loc.thousands_sep;
+
+						for (k = strlen(ts) - 1; k >= 0; k--) {
+							PREPEND(ts[k]);
+						}
+					}
 					if (loc.grouping[ii+1] == 0)                                          
 						jj = 0;     /* keep using current val in loc.grouping[ii] */
 					else if (loc.grouping[ii+1] == CHAR_MAX)                        
diff --git a/dfa.c b/dfa.c
index 3dd3c209..8771bbee 100644
--- a/dfa.c
+++ b/dfa.c
@@ -45,6 +45,11 @@
 
 #include "dfa.h"
 
+/* Gawk doesn't use Gnulib, so don't assume static_assert is present.  */
+#ifndef static_assert
+# define static_assert(cond, diagnostic) \
+    extern int (*foo (void)) [!!sizeof (struct { int foo: (cond) ? 8 : -1; })]
+#endif
 
 #define STREQ(a, b) (strcmp (a, b) == 0)
 
@@ -748,34 +753,16 @@ setbit_wc (wint_t wc, charclass c)
 #endif
 }
 
-/* Set a bit for B in the charclass C, if B is a valid single byte
-   character in the current character set.  If case is folded, set B's
-   lower and upper case variants similarly.  If MB_CUR_MAX > 1, the
-   resulting charset is used only as an optimization, and the caller
-   should set the appropriate field of struct mb_char_classes.  */
+/* Set a bit for B and its case variants in the charclass C.
+   MB_CUR_MAX must be 1.  */
 static void
 setbit_case_fold_c (int b, charclass c)
 {
-  if (MB_CUR_MAX > 1)
-    {
-      wint_t wc = btowc (b);
-      if (wc == WEOF)
-        return;
-      if (case_fold)
-        {
-          setbit_wc (towlower (wc), c);
-          setbit_wc (towupper (wc), c);
-        }
-    }
-  else
-    {
-      if (case_fold)
-        {
-          setbit (tolower (b), c);
-          setbit (toupper (b), c);
-        }
-    }
-  setbit (b, c);
+  int ub = toupper (b);
+  int i;
+  for (i = 0; i < NOTCHAR; i++)
+    if (toupper (i) == ub)
+      setbit (i, c);
 }
 
 
@@ -940,6 +927,50 @@ static unsigned char const *buf_end;    /* reference to end in dfaexec.  */
 # define MIN(a,b) ((a) < (b) ? (a) : (b))
 #endif
 
+/* The set of wchar_t values C such that there's a useful locale
+   somewhere where C != towupper (C) && C != towlower (towupper (C)).
+   For example, 0x00B5 (U+00B5 MICRO SIGN) is in this table, because
+   towupper (0x00B5) == 0x039C (U+039C GREEK CAPITAL LETTER MU), and
+   towlower (0x039C) == 0x03BC (U+03BC GREEK SMALL LETTER MU).  */
+static short const lonesome_lower[] =
+  {
+    0x00B5, 0x0131, 0x017F, 0x01C5, 0x01C8, 0x01CB, 0x01F2, 0x0345,
+    0x03C2, 0x03D0, 0x03D1, 0x03D5, 0x03D6, 0x03F0, 0x03F1,
+
+    /* U+03F2 GREEK LUNATE SIGMA SYMBOL lacks a specific uppercase
+       counterpart in locales predating Unicode 4.0.0 (April 2003).  */
+    0x03F2,
+
+    0x03F5, 0x1E9B, 0x1FBE,
+  };
+
+static_assert ((sizeof lonesome_lower / sizeof *lonesome_lower + 2
+                == CASE_FOLDED_BUFSIZE),
+               "CASE_FOLDED_BUFSIZE is wrong");
+
+/* Find the characters equal to C after case-folding, other than C
+   itself, and store them into FOLDED.  Return the number of characters
+   stored.  */
+int
+case_folded_counterparts (wchar_t c, wchar_t folded[CASE_FOLDED_BUFSIZE])
+{
+  int i;
+  int n = 0;
+  wint_t uc = towupper (c);
+  wint_t lc = towlower (uc);
+  if (uc != c)
+    folded[n++] = uc;
+  if (lc != uc && lc != c && towupper (lc) == uc)
+    folded[n++] = lc;
+  for (i = 0; i < sizeof lonesome_lower / sizeof *lonesome_lower; i++)
+    {
+      wint_t li = lonesome_lower[i];
+      if (li != lc && li != uc && li != c && towupper (li) == uc)
+        folded[n++] = li;
+    }
+  return n;
+}
+
 typedef int predicate (int);
 
 /* The following list maps the names of the Posix named character classes
@@ -1100,7 +1131,7 @@ parse_bracket_exp (void)
 
                   for (c2 = 0; c2 < NOTCHAR; ++c2)
                     if (pred->func (c2))
-                      setbit_case_fold_c (c2, ccl);
+                      setbit (c2, ccl);
                 }
               else
                 known_bracket_exp = false;
@@ -1167,8 +1198,21 @@ parse_bracket_exp (void)
                     }
                 }
               else if (using_simple_locale ())
-                for (; c <= c2; c++)
-                  setbit_case_fold_c (c, ccl);
+                {
+                  for (c1 = c; c1 <= c2; c1++)
+                    setbit (c1, ccl);
+                  if (case_fold)
+                    {
+                      int uc = toupper (c);
+                      int uc2 = toupper (c2);
+                      for (c1 = 0; c1 < NOTCHAR; c1++)
+                        {
+                          int uc1 = toupper (c1);
+                          if (uc <= uc1 && uc1 <= uc2)
+                            setbit (c1, ccl);
+                        }
+                    }
+                }
               else
                 known_bracket_exp = false;
 
@@ -1187,26 +1231,22 @@ parse_bracket_exp (void)
 
       if (MB_CUR_MAX == 1)
         {
-          setbit_case_fold_c (c, ccl);
+          if (case_fold)
+            setbit_case_fold_c (c, ccl);
+          else
+            setbit (c, ccl);
           continue;
         }
 
       if (case_fold)
         {
-          wint_t folded = towlower (wc);
-          if (folded != wc && !setbit_wc (folded, ccl))
-            {
-              REALLOC_IF_NECESSARY (work_mbc->chars, chars_al,
-                                    work_mbc->nchars + 1);
-              work_mbc->chars[work_mbc->nchars++] = folded;
-            }
-          folded = towupper (wc);
-          if (folded != wc && !setbit_wc (folded, ccl))
-            {
-              REALLOC_IF_NECESSARY (work_mbc->chars, chars_al,
-                                    work_mbc->nchars + 1);
-              work_mbc->chars[work_mbc->nchars++] = folded;
-            }
+          wchar_t folded[CASE_FOLDED_BUFSIZE];
+          int i, n = case_folded_counterparts (wc, folded);
+          REALLOC_IF_NECESSARY (work_mbc->chars, chars_al,
+                                work_mbc->nchars + n);
+          for (i = 0; i < n; i++)
+            if (!setbit_wc (folded[i], ccl))
+              work_mbc->chars[work_mbc->nchars++] = folded[i];
         }
       if (!setbit_wc (wc, ccl))
         {
@@ -1552,7 +1592,7 @@ lex (void)
           if (MB_CUR_MAX > 1)
             return lasttok = WCHAR;
 
-          if (case_fold && (tolower (c) != c || toupper (c) != c))
+          if (case_fold && isalpha (c))
             {
               zeroset (ccl);
               setbit_case_fold_c (c, ccl);
@@ -1799,18 +1839,14 @@ atom (void)
   if (MBS_SUPPORT && tok == WCHAR)
     {
       addtok_wc (wctok);
+
       if (case_fold)
         {
-          wint_t folded = towlower (wctok);
-          if (folded != wctok)
-            {
-              addtok_wc (folded);
-              addtok (OR);
-            }
-          folded = towupper (wctok);
-          if (folded != wctok)
+          wchar_t folded[CASE_FOLDED_BUFSIZE];
+          int i, n = case_folded_counterparts (wctok, folded);
+          for (i = 0; i < n; i++)
             {
-              addtok_wc (folded);
+              addtok_wc (folded[i]);
               addtok (OR);
             }
         }
diff --git a/dfa.h b/dfa.h
index 7e0674fc..24fbcbe7 100644
--- a/dfa.h
+++ b/dfa.h
@@ -101,3 +101,11 @@ extern void dfawarn (const char *);
 extern _Noreturn void dfaerror (const char *);
 
 extern int using_utf8 (void);
+
+/* Maximum number of characters that can be the case-folded
+   counterparts of a single character, not counting the character
+   itself.  This is 1 for towupper, 1 for towlower, and 1 for each
+   entry in LONESOME_LOWER; see dfa.c.  */
+enum { CASE_FOLDED_BUFSIZE = 1 + 1 + 19 };
+
+extern int case_folded_counterparts (wchar_t, wchar_t[CASE_FOLDED_BUFSIZE]);
diff --git a/doc/ChangeLog b/doc/ChangeLog
index ac48a973..0b34c55f 100644
--- a/doc/ChangeLog
+++ b/doc/ChangeLog
@@ -1,3 +1,12 @@
+2014-03-10         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* gawktexi.in: Finish indexing improvements. (For now, anyway.)
+
+	Unrelated:
+
+	* gawk.1: Document the quote flag! (Better late than never.)
+	* awkcard.in: Update documentation of quote flag.
+
 2014-03-08         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* gawktexi.in: Minor edits to the discussion of the memory allocation
diff --git a/doc/awkcard.in b/doc/awkcard.in
index 610032b7..5f3a9735 100644
--- a/doc/awkcard.in
+++ b/doc/awkcard.in
@@ -1,7 +1,7 @@
 .\" AWK Reference Card --- Arnold Robbins, arnold@skeeve.com
 .\"
 .\" Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002,
-.\" 2003, 2004, 2005, 2007, 2009, 2010, 2011, 2012, 2013
+.\" 2003, 2004, 2005, 2007, 2009, 2010, 2011, 2012, 2013, 2014
 .\" Free Software Foundation, Inc.
 .\" 
 .\" Permission is granted to make and distribute verbatim copies of
@@ -1493,7 +1493,8 @@ Only has an effect when the field width is wider
 than the value to be printed.
 T}
 \*(CB\*(FC'\*(FR	T{
-Use the locale's thousands separator for \*(FC%d\fP, \*(FC%i\fP, and \*(FC%u\fP.\*(CD
+Use the locale's thousands separator and decimal
+point characters.\*(CD
 T}
 \*(FIwidth\fP	T{
 Pad the field to this width. The field is normally
diff --git a/doc/gawk.1 b/doc/gawk.1
index bd58b10c..a879b527 100644
--- a/doc/gawk.1
+++ b/doc/gawk.1
@@ -13,7 +13,7 @@
 .		if \w'\(rq' .ds rq "\(rq
 .	\}
 .\}
-.TH GAWK 1 "Jan 28 2014" "Free Software Foundation" "Utility Commands"
+.TH GAWK 1 "Mar 08 2014" "Free Software Foundation" "Utility Commands"
 .SH NAME
 gawk \- pattern scanning and processing language
 .SH SYNOPSIS
@@ -2454,6 +2454,15 @@ This applies only to the numeric output formats.
 This flag only has an effect when the field width is wider than the
 value to be printed.
 .TP
+.B '
+A single quote character instructs
+.I gawk
+to insert the locale's thousands-separator character
+into decimal numbers, and to also use the locale's
+decimal point character with floating point formats.
+This requires correct locale support in the C library
+and in the definition of the current locale.
+.TP
 .I width
 The field should be padded to this width.  The field is normally padded
 with spaces.  With the
diff --git a/doc/gawk.info b/doc/gawk.info
index 74773fcb..13f5ea77 100644
--- a/doc/gawk.info
+++ b/doc/gawk.info
@@ -25937,6 +25937,9 @@ Info file, in approximate chronological order:
 
    * Anders Wallin helped keep the VMS port going for several years.
 
+   * Assaf Gordon contributed the code to implement the `--sandbox'
+     option.
+
    * John Haque made the following contributions:
 
         - The modifications to convert `gawk' into a byte-code
@@ -31301,12 +31304,12 @@ Index
 * for statement, looping over arrays:    Scanning an Array.   (line  20)
 * fork() extension function:             Extension Sample Fork.
                                                               (line  11)
+* format specifiers:                     Basic Printf.        (line  15)
 * format specifiers, mixing regular with positional specifiers: Printf Ordering.
                                                               (line  57)
 * format specifiers, printf statement:   Control Letters.     (line   6)
 * format specifiers, strftime() function (gawk): Time Functions.
                                                               (line  88)
-* format strings:                        Basic Printf.        (line  15)
 * formats, numeric output:               OFMT.                (line   6)
 * formatting output:                     Printf.              (line   6)
 * forward slash (/):                     Regexp.              (line  10)
@@ -31531,6 +31534,12 @@ Index
 * gettext() function (C library):        Explaining gettext.  (line  62)
 * gettimeofday() extension function:     Extension Sample Time.
                                                               (line  13)
+* git utility <1>:                       Adding Code.         (line 111)
+* git utility <2>:                       Accessing The Source.
+                                                              (line  10)
+* git utility <3>:                       Other Versions.      (line  29)
+* git utility:                           gawkextlib.          (line  29)
+* git, use of for gawk source code:      Derived Files.       (line   6)
 * GMP:                                   Arbitrary Precision Arithmetic.
                                                               (line   6)
 * GNITS mailing list:                    Acknowledgments.     (line  52)
@@ -31547,6 +31556,7 @@ Index
 * GNU/Linux <1>:                         Glossary.            (line 624)
 * GNU/Linux <2>:                         I18N Example.        (line  55)
 * GNU/Linux:                             Manual History.      (line  28)
+* Gordon, Assaf:                         Contributors.        (line 105)
 * GPL (General Public License) <1>:      Glossary.            (line 314)
 * GPL (General Public License):          Manual History.      (line  11)
 * GPL (General Public License), printing: Options.            (line  88)
@@ -31564,7 +31574,7 @@ Index
                                                               (line  66)
 * Hankerson, Darrel <1>:                 Contributors.        (line  60)
 * Hankerson, Darrel:                     Acknowledgments.     (line  60)
-* Haque, John:                           Contributors.        (line 105)
+* Haque, John:                           Contributors.        (line 108)
 * Hartholz, Elaine:                      Acknowledgments.     (line  38)
 * Hartholz, Marshall:                    Acknowledgments.     (line  38)
 * Hasegawa, Isamu:                       Contributors.        (line  94)
@@ -31652,9 +31662,9 @@ Index
 * installing gawk:                       Installation.        (line   6)
 * INT signal (MS-Windows):               Profiling.           (line 214)
 * int() function:                        Numeric Functions.   (line  23)
-* integer, arbitrary precision:          Arbitrary Precision Integers.
-                                                              (line   6)
 * integers:                              General Arithmetic.  (line   6)
+* integers, arbitrary precision:         Arbitrary Precision Integers.
+                                                              (line   6)
 * integers, unsigned:                    General Arithmetic.  (line  15)
 * interacting with other programs:       I/O Functions.       (line  72)
 * internationalization <1>:              I18N and L10N.       (line   6)
@@ -31800,10 +31810,15 @@ Index
 * login information:                     Passwd Functions.    (line  16)
 * long options:                          Command Line.        (line  13)
 * loops:                                 While Statement.     (line   6)
+* loops, break statement and:            Break Statement.     (line   6)
 * loops, continue statements and:        For Statement.       (line  64)
 * loops, count for header:               Profiling.           (line 131)
+* loops, do-while:                       Do Statement.        (line   6)
 * loops, exiting:                        Break Statement.     (line   6)
+* loops, for, array scanning:            Scanning an Array.   (line   6)
+* loops, for, iterative:                 For Statement.       (line   6)
 * loops, See Also while statement:       While Statement.     (line   6)
+* loops, while:                          While Statement.     (line   6)
 * ls utility:                            More Complex.        (line  15)
 * lshift() function (gawk):              Bitwise Functions.   (line  46)
 * lvalues/rvalues:                       Assignment Ops.      (line  32)
@@ -32200,10 +32215,10 @@ Index
 * programming conventions, private variable names: Library Names.
                                                               (line  23)
 * programming language, recipe for:      History.             (line   6)
-* Programming languages, Ada:            Glossary.            (line  20)
+* programming languages, Ada:            Glossary.            (line  20)
 * programming languages, data-driven vs. procedural: Getting Started.
                                                               (line  12)
-* Programming languages, Java:           Glossary.            (line 388)
+* programming languages, Java:           Glossary.            (line 388)
 * programming, basic steps:              Basic High Level.    (line  20)
 * programming, concepts:                 Basic Concepts.      (line   6)
 * pwcat program:                         Passwd Functions.    (line  23)
@@ -32340,7 +32355,7 @@ Index
 * RLENGTH variable, match() function and: String Functions.   (line 220)
 * Robbins, Arnold <1>:                   Future Extensions.   (line   6)
 * Robbins, Arnold <2>:                   Bugs.                (line  32)
-* Robbins, Arnold <3>:                   Contributors.        (line 132)
+* Robbins, Arnold <3>:                   Contributors.        (line 135)
 * Robbins, Arnold <4>:                   General Data Types.  (line   6)
 * Robbins, Arnold <5>:                   Alarm Program.       (line   6)
 * Robbins, Arnold <6>:                   Passwd Functions.    (line  90)
@@ -32381,7 +32396,7 @@ Index
                                                               (line  68)
 * sandbox mode:                          Options.             (line 279)
 * scalar values:                         Basic Data Typing.   (line  13)
-* Schorr, Andrew <1>:                    Contributors.        (line 128)
+* Schorr, Andrew <1>:                    Contributors.        (line 131)
 * Schorr, Andrew:                        Acknowledgments.     (line  60)
 * Schreiber, Bert:                       Acknowledgments.     (line  38)
 * Schreiber, Rita:                       Acknowledgments.     (line  38)
@@ -32516,6 +32531,7 @@ Index
 * source code, mawk:                     Other Versions.      (line  44)
 * source code, mixing:                   Options.             (line 117)
 * source code, pawk:                     Other Versions.      (line  78)
+* source code, pawk (Python version):    Other Versions.      (line 124)
 * source code, QSE Awk:                  Other Versions.      (line 130)
 * source code, QuikTrim Awk:             Other Versions.      (line 134)
 * source code, Solaris awk:              Other Versions.      (line  96)
@@ -32816,7 +32832,7 @@ Index
 * xgettext utility:                      String Extraction.   (line  13)
 * XOR bitwise operation:                 Bitwise Functions.   (line   6)
 * xor() function (gawk):                 Bitwise Functions.   (line  55)
-* Yawitz, Efraim:                        Contributors.        (line 126)
+* Yawitz, Efraim:                        Contributors.        (line 129)
 * Zaretskii, Eli <1>:                    Bugs.                (line  70)
 * Zaretskii, Eli <2>:                    Contributors.        (line  55)
 * Zaretskii, Eli:                        Acknowledgments.     (line  60)
@@ -33317,62 +33333,62 @@ Ref: Ranges and Locales-Footnote-11034931
 Ref: Ranges and Locales-Footnote-21034958
 Ref: Ranges and Locales-Footnote-31035192
 Node: Contributors1035413
-Node: Installation1040558
-Node: Gawk Distribution1041452
-Node: Getting1041936
-Node: Extracting1042762
-Node: Distribution contents1044454
-Node: Unix Installation1050159
-Node: Quick Installation1050776
-Node: Additional Configuration Options1053222
-Node: Configuration Philosophy1054958
-Node: Non-Unix Installation1057312
-Node: PC Installation1057770
-Node: PC Binary Installation1059069
-Node: PC Compiling1060917
-Node: PC Testing1063861
-Node: PC Using1065037
-Node: Cygwin1069205
-Node: MSYS1070014
-Node: VMS Installation1070528
-Node: VMS Compilation1071292
-Ref: VMS Compilation-Footnote-11072544
-Node: VMS Dynamic Extensions1072602
-Node: VMS Installation Details1073975
-Node: VMS Running1076226
-Node: VMS GNV1079060
-Node: VMS Old Gawk1079783
-Node: Bugs1080253
-Node: Other Versions1084171
-Node: Notes1090255
-Node: Compatibility Mode1091055
-Node: Additions1091838
-Node: Accessing The Source1092765
-Node: Adding Code1094205
-Node: New Ports1100250
-Node: Derived Files1104385
-Ref: Derived Files-Footnote-11109706
-Ref: Derived Files-Footnote-21109740
-Ref: Derived Files-Footnote-31110340
-Node: Future Extensions1110438
-Node: Implementation Limitations1111021
-Node: Extension Design1112273
-Node: Old Extension Problems1113427
-Ref: Old Extension Problems-Footnote-11114935
-Node: Extension New Mechanism Goals1114992
-Ref: Extension New Mechanism Goals-Footnote-11118357
-Node: Extension Other Design Decisions1118543
-Node: Extension Future Growth1120649
-Node: Old Extension Mechanism1121485
-Node: Basic Concepts1123225
-Node: Basic High Level1123906
-Ref: figure-general-flow1124177
-Ref: figure-process-flow1124776
-Ref: Basic High Level-Footnote-11128005
-Node: Basic Data Typing1128190
-Node: Glossary1131545
-Node: Copying1157007
-Node: GNU Free Documentation License1194564
-Node: Index1219701
+Node: Installation1040640
+Node: Gawk Distribution1041534
+Node: Getting1042018
+Node: Extracting1042844
+Node: Distribution contents1044536
+Node: Unix Installation1050241
+Node: Quick Installation1050858
+Node: Additional Configuration Options1053304
+Node: Configuration Philosophy1055040
+Node: Non-Unix Installation1057394
+Node: PC Installation1057852
+Node: PC Binary Installation1059151
+Node: PC Compiling1060999
+Node: PC Testing1063943
+Node: PC Using1065119
+Node: Cygwin1069287
+Node: MSYS1070096
+Node: VMS Installation1070610
+Node: VMS Compilation1071374
+Ref: VMS Compilation-Footnote-11072626
+Node: VMS Dynamic Extensions1072684
+Node: VMS Installation Details1074057
+Node: VMS Running1076308
+Node: VMS GNV1079142
+Node: VMS Old Gawk1079865
+Node: Bugs1080335
+Node: Other Versions1084253
+Node: Notes1090337
+Node: Compatibility Mode1091137
+Node: Additions1091920
+Node: Accessing The Source1092847
+Node: Adding Code1094287
+Node: New Ports1100332
+Node: Derived Files1104467
+Ref: Derived Files-Footnote-11109788
+Ref: Derived Files-Footnote-21109822
+Ref: Derived Files-Footnote-31110422
+Node: Future Extensions1110520
+Node: Implementation Limitations1111103
+Node: Extension Design1112355
+Node: Old Extension Problems1113509
+Ref: Old Extension Problems-Footnote-11115017
+Node: Extension New Mechanism Goals1115074
+Ref: Extension New Mechanism Goals-Footnote-11118439
+Node: Extension Other Design Decisions1118625
+Node: Extension Future Growth1120731
+Node: Old Extension Mechanism1121567
+Node: Basic Concepts1123307
+Node: Basic High Level1123988
+Ref: figure-general-flow1124259
+Ref: figure-process-flow1124858
+Ref: Basic High Level-Footnote-11128087
+Node: Basic Data Typing1128272
+Node: Glossary1131627
+Node: Copying1157089
+Node: GNU Free Documentation License1194646
+Node: Index1219783
 
 End Tag Table
diff --git a/doc/gawk.texi b/doc/gawk.texi
index b9bf2170..1b610c42 100644
--- a/doc/gawk.texi
+++ b/doc/gawk.texi
@@ -8359,7 +8359,7 @@ parentheses are necessary if any of the item expressions use the @samp{>}
 relational operator; otherwise, it can be confused with an output redirection
 (@pxref{Redirection}).
 
-@cindex format strings
+@cindex format specifiers
 The difference between @code{printf} and @code{print} is the @var{format}
 argument.  This is an expression whose value is taken as a string; it
 specifies how to output each of the other arguments.  It is called the
@@ -12706,6 +12706,7 @@ the first thing on its line.
 @subsection The @code{while} Statement
 @cindex @code{while} statement
 @cindex loops
+@cindex loops, @code{while}
 @cindex loops, See Also @code{while} statement
 
 In programming, a @dfn{loop} is a part of a program that can
@@ -12766,6 +12767,7 @@ program is harder to read without it.
 @node Do Statement
 @subsection The @code{do}-@code{while} Statement
 @cindex @code{do}-@code{while} statement
+@cindex loops, @code{do}-@code{while}
 
 The @code{do} loop is a variation of the @code{while} looping statement.
 The @code{do} loop executes the @var{body} once and then repeats the
@@ -12811,6 +12813,7 @@ occasionally is there a real use for a @code{do} statement.
 @node For Statement
 @subsection The @code{for} Statement
 @cindex @code{for} statement
+@cindex loops, @code{for}, iterative
 
 The @code{for} statement makes it more convenient to count iterations of a
 loop.  The general form of the @code{for} statement looks like this:
@@ -12983,6 +12986,7 @@ it is not available.
 @subsection The @code{break} Statement
 @cindex @code{break} statement
 @cindex loops, exiting
+@cindex loops, @code{break} statement and
 
 The @code{break} statement jumps out of the innermost @code{for},
 @code{while}, or @code{do} loop that encloses it.  The following example
@@ -14628,6 +14632,7 @@ END @{
 @subsection Scanning All Elements of an Array
 @cindex elements in arrays, scanning
 @cindex arrays, scanning
+@cindex loops, @code{for}, array scanning
 
 In programs that use arrays, it is often necessary to use a loop that
 executes once for each element of an array.  In other languages, where
@@ -29318,7 +29323,7 @@ the problem at hand is often the correct approach in such situations.
 
 @node Arbitrary Precision Integers
 @section Arbitrary Precision Integer Arithmetic with @command{gawk}
-@cindex integer, arbitrary precision
+@cindex integers, arbitrary precision
 
 If one of the options @option{--bignum} or @option{-M} is specified,
 @command{gawk} performs all
@@ -33068,6 +33073,7 @@ The @code{time} extension described earlier (@pxref{Extension Sample
 Time}) was originally from this project but has been moved in to the
 main @command{gawk} distribution.
 
+@cindex @command{git} utility
 You can check out the code for the @code{gawkextlib} project
 using the @uref{http://git-scm.com, GIT} distributed source
 code control system.  The command is as follows:
@@ -34698,6 +34704,11 @@ environments.
 Anders Wallin helped keep the VMS port going for several years.
 
 @item
+@cindex Gordon, Assaf
+Assaf Gordon contributed the code to implement the
+@option{--sandbox} option.
+
+@item
 @cindex Haque, John
 John Haque made the following contributions:
 
@@ -36110,6 +36121,7 @@ It is available in several archive formats:
 @uref{http://www.cs.princeton.edu/~bwk/btl.mirror/awk.zip}
 @end table
 
+@cindex @command{git} utility
 You can also retrieve it from Git Hub:
 
 @example
@@ -36238,6 +36250,7 @@ This is an embeddable @command{awk} interpreter derived from
 @uref{http://repo.hu/projects/libmawk/}.
 
 @item @code{pawk}
+@cindex source code, @command{pawk} (Python version)
 @cindex @code{pawk}, @command{awk}-like facilities for Python
 This is a Python module that claims to bring @command{awk}-like
 features to Python. See @uref{https://github.com/alecthomas/pawk}
@@ -36343,6 +36356,7 @@ As @command{gawk} is Free Software, the source code is always available.
 @ref{Gawk Distribution}, describes how to get and build the formal,
 released versions of @command{gawk}.
 
+@cindex @command{git} utility
 However, if you want to modify @command{gawk} and contribute back your
 changes, you will probably wish to work with the development version.
 To do so, you will need to access the @command{gawk} source code
@@ -36518,6 +36532,7 @@ If possible, please update the @command{man} page as well.
 
 You will also have to sign paperwork for your documentation changes.
 
+@cindex @command{git} utility
 @item
 Submit changes as unified diffs.
 Use @samp{diff -u -r -N} to compare
@@ -36651,6 +36666,8 @@ coding style and brace layout that suits your taste.
 @node Derived Files
 @appendixsubsec Why Generated Files Are Kept In @command{git}
 
+@c STARTOFRANGE gawkgit
+@cindex @command{git}, use of for @command{gawk} source code
 @c From emails written March 22, 2012, to the gawk developers list.
 
 If you look at the @command{gawk} source in the @command{git}
@@ -36830,7 +36847,7 @@ wget http://git.savannah.gnu.org/cgit/gawk.git/snapshot/gawk-@var{branchname}.ta
 
 @noindent
 to retrieve a snapshot of the given branch.
-
+@c ENDOFRANGE gawkgit
 
 @node Future Extensions
 @appendixsec Probable Future Extensions
@@ -37409,7 +37426,7 @@ better written in another language.
 You can get it from @uref{http://awk.info/?awk100/aaa}.
 
 @cindex Ada programming language
-@cindex Programming languages, Ada
+@cindex programming languages, Ada
 @item Ada
 A programming language originally defined by the U.S.@: Department of
 Defense for embedded programming. It was designed to enforce good
@@ -37892,7 +37909,7 @@ information about the name of the organization and its language-independent
 three-letter acronym.
 
 @cindex Java programming language
-@cindex Programming languages, Java
+@cindex programming languages, Java
 @item Java
 A modern programming language originally developed by Sun Microsystems
 (now Oracle) supporting Object-Oriented programming.  Although usually
diff --git a/doc/gawktexi.in b/doc/gawktexi.in
index 6054ed40..e9a31935 100644
--- a/doc/gawktexi.in
+++ b/doc/gawktexi.in
@@ -7977,7 +7977,7 @@ parentheses are necessary if any of the item expressions use the @samp{>}
 relational operator; otherwise, it can be confused with an output redirection
 (@pxref{Redirection}).
 
-@cindex format strings
+@cindex format specifiers
 The difference between @code{printf} and @code{print} is the @var{format}
 argument.  This is an expression whose value is taken as a string; it
 specifies how to output each of the other arguments.  It is called the
@@ -12083,6 +12083,7 @@ the first thing on its line.
 @subsection The @code{while} Statement
 @cindex @code{while} statement
 @cindex loops
+@cindex loops, @code{while}
 @cindex loops, See Also @code{while} statement
 
 In programming, a @dfn{loop} is a part of a program that can
@@ -12143,6 +12144,7 @@ program is harder to read without it.
 @node Do Statement
 @subsection The @code{do}-@code{while} Statement
 @cindex @code{do}-@code{while} statement
+@cindex loops, @code{do}-@code{while}
 
 The @code{do} loop is a variation of the @code{while} looping statement.
 The @code{do} loop executes the @var{body} once and then repeats the
@@ -12188,6 +12190,7 @@ occasionally is there a real use for a @code{do} statement.
 @node For Statement
 @subsection The @code{for} Statement
 @cindex @code{for} statement
+@cindex loops, @code{for}, iterative
 
 The @code{for} statement makes it more convenient to count iterations of a
 loop.  The general form of the @code{for} statement looks like this:
@@ -12360,6 +12363,7 @@ it is not available.
 @subsection The @code{break} Statement
 @cindex @code{break} statement
 @cindex loops, exiting
+@cindex loops, @code{break} statement and
 
 The @code{break} statement jumps out of the innermost @code{for},
 @code{while}, or @code{do} loop that encloses it.  The following example
@@ -13959,6 +13963,7 @@ END @{
 @subsection Scanning All Elements of an Array
 @cindex elements in arrays, scanning
 @cindex arrays, scanning
+@cindex loops, @code{for}, array scanning
 
 In programs that use arrays, it is often necessary to use a loop that
 executes once for each element of an array.  In other languages, where
@@ -28459,7 +28464,7 @@ the problem at hand is often the correct approach in such situations.
 
 @node Arbitrary Precision Integers
 @section Arbitrary Precision Integer Arithmetic with @command{gawk}
-@cindex integer, arbitrary precision
+@cindex integers, arbitrary precision
 
 If one of the options @option{--bignum} or @option{-M} is specified,
 @command{gawk} performs all
@@ -32209,6 +32214,7 @@ The @code{time} extension described earlier (@pxref{Extension Sample
 Time}) was originally from this project but has been moved in to the
 main @command{gawk} distribution.
 
+@cindex @command{git} utility
 You can check out the code for the @code{gawkextlib} project
 using the @uref{http://git-scm.com, GIT} distributed source
 code control system.  The command is as follows:
@@ -33839,6 +33845,11 @@ environments.
 Anders Wallin helped keep the VMS port going for several years.
 
 @item
+@cindex Gordon, Assaf
+Assaf Gordon contributed the code to implement the
+@option{--sandbox} option.
+
+@item
 @cindex Haque, John
 John Haque made the following contributions:
 
@@ -35251,6 +35262,7 @@ It is available in several archive formats:
 @uref{http://www.cs.princeton.edu/~bwk/btl.mirror/awk.zip}
 @end table
 
+@cindex @command{git} utility
 You can also retrieve it from Git Hub:
 
 @example
@@ -35379,6 +35391,7 @@ This is an embeddable @command{awk} interpreter derived from
 @uref{http://repo.hu/projects/libmawk/}.
 
 @item @code{pawk}
+@cindex source code, @command{pawk} (Python version)
 @cindex @code{pawk}, @command{awk}-like facilities for Python
 This is a Python module that claims to bring @command{awk}-like
 features to Python. See @uref{https://github.com/alecthomas/pawk}
@@ -35484,6 +35497,7 @@ As @command{gawk} is Free Software, the source code is always available.
 @ref{Gawk Distribution}, describes how to get and build the formal,
 released versions of @command{gawk}.
 
+@cindex @command{git} utility
 However, if you want to modify @command{gawk} and contribute back your
 changes, you will probably wish to work with the development version.
 To do so, you will need to access the @command{gawk} source code
@@ -35659,6 +35673,7 @@ If possible, please update the @command{man} page as well.
 
 You will also have to sign paperwork for your documentation changes.
 
+@cindex @command{git} utility
 @item
 Submit changes as unified diffs.
 Use @samp{diff -u -r -N} to compare
@@ -35792,6 +35807,8 @@ coding style and brace layout that suits your taste.
 @node Derived Files
 @appendixsubsec Why Generated Files Are Kept In @command{git}
 
+@c STARTOFRANGE gawkgit
+@cindex @command{git}, use of for @command{gawk} source code
 @c From emails written March 22, 2012, to the gawk developers list.
 
 If you look at the @command{gawk} source in the @command{git}
@@ -35971,7 +35988,7 @@ wget http://git.savannah.gnu.org/cgit/gawk.git/snapshot/gawk-@var{branchname}.ta
 
 @noindent
 to retrieve a snapshot of the given branch.
-
+@c ENDOFRANGE gawkgit
 
 @node Future Extensions
 @appendixsec Probable Future Extensions
@@ -36550,7 +36567,7 @@ better written in another language.
 You can get it from @uref{http://awk.info/?awk100/aaa}.
 
 @cindex Ada programming language
-@cindex Programming languages, Ada
+@cindex programming languages, Ada
 @item Ada
 A programming language originally defined by the U.S.@: Department of
 Defense for embedded programming. It was designed to enforce good
@@ -37033,7 +37050,7 @@ information about the name of the organization and its language-independent
 three-letter acronym.
 
 @cindex Java programming language
-@cindex Programming languages, Java
+@cindex programming languages, Java
 @item Java
 A modern programming language originally developed by Sun Microsystems
 (now Oracle) supporting Object-Oriented programming.  Although usually
diff --git a/helpers/ChangeLog b/helpers/ChangeLog
index 14affebc..e5087f37 100644
--- a/helpers/ChangeLog
+++ b/helpers/ChangeLog
@@ -1,3 +1,8 @@
+2014-03-10         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* quoteconvert2.sh: Use .UTF-8 locales per request from
+	Michal Jaegermann.
+
 2014-03-08         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* quoteconvert2.sh, tryfmt.c, scanfmt.c: New files.
diff --git a/helpers/quoteconvert2.sh b/helpers/quoteconvert2.sh
index e33a4d05..63750a37 100755
--- a/helpers/quoteconvert2.sh
+++ b/helpers/quoteconvert2.sh
@@ -17,17 +17,17 @@ fi
 llist="
 C
 en_US
-en_US.utf8
+en_US.UTF-8
 de_DE
-de_DE.utf8
+de_DE.UTF-8
 fr_FR
-fr_FR.utf8
+fr_FR.UTF-8
 pt_PT
-pt_PT.utf8
+pt_PT.UTF-8
 pt_BR
-pt_BR.utf8
+pt_BR.UTF-8
 ru_RU
-ru_RU.utf8
+ru_RU.UTF-8
 pl_PX
 "
 
diff --git a/regex_internal.c b/regex_internal.c
index 10dd6e00..056cff3d 100644
--- a/regex_internal.c
+++ b/regex_internal.c
@@ -320,12 +320,11 @@ build_wcs_upper_buffer (re_string_t *pstr)
 			       + byte_idx), remain_len, &pstr->cur_state);
 	  if (BE (mbclen + 2 > 2, 1))
 	    {
-	      wchar_t wcu = wc;
-	      if (iswlower (wc))
+	      wchar_t wcu = towupper (wc);
+	      if (wcu != wc)
 		{
 		  size_t mbcdlen;
 
-		  wcu = towupper (wc);
 		  mbcdlen = wcrtomb (buf, wcu, &prev_st);
 		  if (BE (mbclen == mbcdlen, 1))
 		    memcpy (pstr->mbs + byte_idx, buf, mbclen);
@@ -390,12 +389,11 @@ build_wcs_upper_buffer (re_string_t *pstr)
 	mbclen = __mbrtowc (&wc, p, remain_len, &pstr->cur_state);
 	if (BE (mbclen + 2 > 2, 1))
 	  {
-	    wchar_t wcu = wc;
-	    if (iswlower (wc))
+	    wchar_t wcu = towupper (wc);
+	    if (wcu != wc)
 	      {
 		size_t mbcdlen;
 
-		wcu = towupper (wc);
 		mbcdlen = wcrtomb ((char *) buf, wcu, &prev_st);
 		if (BE (mbclen == mbcdlen, 1))
 		  memcpy (pstr->mbs + byte_idx, buf, mbclen);
@@ -547,10 +545,7 @@ build_upper_buffer (re_string_t *pstr)
       int ch = pstr->raw_mbs[pstr->raw_mbs_idx + char_idx];
       if (BE (pstr->trans != NULL, 0))
 	ch = pstr->trans[ch];
-      if (islower (ch))
-	pstr->mbs[char_idx] = toupper (ch);
-      else
-	pstr->mbs[char_idx] = ch;
+      pstr->mbs[char_idx] = toupper (ch);
     }
   pstr->valid_len = char_idx;
   pstr->valid_raw_len = char_idx;