added Arnold's fixes, updates to awktest.tar

FIXES

@@ -25,6 +25,37 @@ THIS SOFTWARE.
 This file lists all bug fixes, changes, etc., made since the AWK book
 was sent to the printers in August, 1987.
 
+Aug 23, 2018:
+        A long list of fixes courtesy of Arnold Robbins,
+        to whom profound thanks.
+
+        1. ofs-rebuild: OFS value used to rebuild the record was incorrect.
+        Fixed August 19, 2014. Revised fix August 2018.
+
+        2. system-status: Instead of a floating-point division by 256, use
+        the wait(2) macros to create a reasonable exit status.
+        Fixed March 12, 2016.
+
+        3. space: Use provided xisblank() function instead of ispace() for
+        matching [[:blank:]].
+
+        4. a-format: Add POSIX standard %a and %A to supported formats. Check
+        at runtime that this format is available.
+
+        5. decr-NF: Decrementing NF did not change $0. This is a decades-old
+        bug. There are interactions with the old and new value of OFS as well.
+        Most of the fix came from the NetBSD awk.
+
+        6. string-conv: String conversions of scalars were sticky.  Once a
+        conversion to string happened, even with OFMT, that value was used until
+        a new numeric value was assigned, even if OFMT differed from CONVFMT,
+        and also if CONVFMT changed.
+
+        7. unary-plus: Unary plus on a string constant returned the string.
+        Instead, it should convert the value to numeric and give that value.
+
+	Also added Arnold's tests for these to awktest.tar as T.arnold.
+
 Aug 15, 2018:
 	fixed mangled awktest.tar (thanks, Arnold), posted all
 	current (very minor) fixes to github / onetrueawk

TODO

@@ -1,20 +0,0 @@
-Fri Aug 10 11:11:11 IDT 2018
-============================
-
-A semi-random list of things to look into.
-
-1. DONE. Make sure the FIXES file has everything after the date order
-was reversed.
-
-2. DONE. Look into the problem that NF-- doesn't throw away the last field.
-This a long-standing issue.
-
-3. DONE. Add suport for %a and %A printf formats.
-
-4. DONE. Look into the problem that string conversions are sticky; once
-done they never change, even if OFMT or CONVFMT do. Similarly, a
-conversion from OFMT becomes the permanent string value, which is
-incorrect.
-
-5. MAYBE: Improve the test suite enough so that it can be cleaned up after
-it was run, and old results can be saved for comparison.

main.c

@@ -22,7 +22,7 @@ ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
 THIS SOFTWARE.
 ****************************************************************/
 
-const char	*version = "version 20130105 + fixes by arnold@skeeve.com 20180822";
+const char	*version = "version 20180823";
 
 #define DEBUG
 #include <stdio.h>

makefile

@@ -23,17 +23,19 @@
 # ****************************************************************/
 
 CFLAGS = -g
-CFLAGS = -O2
 CFLAGS =
+CFLAGS = -O2
 
+# compiler options
 #CC = gcc -Wall -g -Wwrite-strings
-CC = gcc -g -Wall -pedantic 
 #CC = gcc -O4 -Wall -pedantic -fno-strict-aliasing
 #CC = gcc -fprofile-arcs -ftest-coverage # then gcov f1.c; cat f1.c.gcov
+CC = gcc -g -Wall -pedantic 
 
-YACC = bison -d -y
+# yacc options.  pick one; this varies a lot by system.
-# YACC = yacc -d
 #YFLAGS = -d -S
+#YACC = bison -d -y
+YACC = yacc -d
 		# -S uses sprintf in yacc parser instead of sprint
 
 OFILES = b.o main.o parse.o proctab.o tran.o lib.o run.o lex.o