buildsys: update unifdef

sync up to 55501a61dd257e941d53f175350ee52fc6ff2866
(2.9 + refactor keyword and symbol matching)

Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>

5 files changed, 934 insertions, 371 deletions

diff --git a/Makefile.in b/Makefile.in
index 500b8f37d..946e14e9f 100644
--- a/Makefile.in
+++ b/Makefile.in
@@ -67,6 +67,10 @@ MAKEFLAGS += -L
 $(top_builddir)include/config/linuxthreads/old.h $(top_builddir)include/config/linuxthreads/new.h:
 	@true
 
+$(top_builddir)include/generated/unifdef_config.h: $(top_builddir)include/bits/uClibc_config.h| $(top_builddir)include/generated
+	@$(disp_gen)
+	$(Q)$(SED) -e '1,3d' $^ > $@
+
 # For the moment, we have to keep re-running this target
 # because the fix includes scripts rely on pre-processers
 # in order to generate the headers correctly :(.  That
@@ -331,7 +335,9 @@ $(PREFIX)$(RUNTIME_PREFIX)$(MULTILIB_DIR):
 	$(do_mkdir)
 endif
 endif
-install_headers: headers $(top_builddir)extra/scripts/unifdef | $(PREFIX)$(DEVEL_PREFIX)include
+
+
+install_headers: headers $(top_builddir)extra/scripts/unifdef $(top_builddir)include/generated/unifdef_config.h | $(PREFIX)$(DEVEL_PREFIX)include
 	@$(call disp_install,"include -> $(PREFIX)$(DEVEL_PREFIX)include")
 	$(Q)top_builddir=$(top_builddir) \
 	$(top_srcdir)extra/scripts/install_headers.sh \
diff --git a/extra/scripts/install_headers.sh b/extra/scripts/install_headers.sh
index 14d64dc9d..5a966066d 100755
--- a/extra/scripts/install_headers.sh
+++ b/extra/scripts/install_headers.sh
@@ -32,7 +32,6 @@ if ! test -x "$top_builddir/extra/scripts/unifdef"; then
 	exit 1
 fi
 
-
 # Sanitize and copy uclibc headers
 (
 # We must cd, or else we'll prepend "${srcdir}" to filenames!
@@ -56,6 +55,9 @@ while read -r filename; do
 	# Do not abort the script if unifdef "fails"!
 	# NB2: careful with sed command arguments, they contain tab character
 	"$top_builddir/extra/scripts/unifdef" \
+		-B \
+		-t \
+		-f "$top_builddir/include/generated/unifdef_config.h" \
 		-U_LIBC \
 		-U__UCLIBC_GEN_LOCALE \
 		-U__NO_CTYPE \
diff --git a/extra/scripts/unifdef.c b/extra/scripts/unifdef.c
index abc39966c..b159df0a6 100644
--- a/extra/scripts/unifdef.c
+++ b/extra/scripts/unifdef.c
@@ -1,13 +1,5 @@
 /*
- * Copyright (c) 2002 - 2005 Tony Finch <dot@dotat.at>.  All rights reserved.
- *
- * This code is derived from software contributed to Berkeley by Dave Yost.
- * It was rewritten to support ANSI C by Tony Finch. The original version of
- * unifdef carried the following copyright notice. None of its code remains
- * in this version (though some of the names remain).
- *
- * Copyright (c) 1985, 1993
- *	The Regents of the University of California.  All rights reserved.
+ * Copyright (c) 2002 - 2013 Tony Finch <dot@dotat.at>
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
@@ -31,27 +23,15 @@
  * SUCH DAMAGE.
  */
 
-#include <sys/cdefs.h>
-
-#ifndef lint
-#if 0
-static const char copyright[] =
-"@(#) Copyright (c) 1985, 1993\n\
-	The Regents of the University of California.  All rights reserved.\n";
-#endif
-#ifdef __IDSTRING
-__IDSTRING(Berkeley, "@(#)unifdef.c	8.1 (Berkeley) 6/6/93");
-__IDSTRING(NetBSD, "$NetBSD: unifdef.c,v 1.8 2000/07/03 02:51:36 matt Exp $");
-__IDSTRING(dotat, "$dotat: things/unifdef.c,v 1.171 2005/03/08 12:38:48 fanf2 Exp $");
-#endif
-#endif /* not lint */
-#ifdef __FBSDID
-__FBSDID("$FreeBSD: /repoman/r/ncvs/src/usr.bin/unifdef/unifdef.c,v 1.20 2005/05/21 09:55:09 ru Exp $");
-#endif
-
 /*
  * unifdef - remove ifdef'ed lines
  *
+ * This code was derived from software contributed to Berkeley by Dave Yost.
+ * It was rewritten to support ANSI C by Tony Finch. The original version
+ * of unifdef carried the 4-clause BSD copyright licence. None of its code
+ * remains in this version (though some of the names remain) so it now
+ * carries a more liberal licence.
+ *
  *  Wishlist:
  *      provide an option which will append the name of the
  *        appropriate symbol after #else's and #endif's
@@ -59,26 +39,17 @@ __FBSDID("$FreeBSD: /repoman/r/ncvs/src/usr.bin/unifdef/unifdef.c,v 1.20 2005/05
  *        #else's and #endif's to see that they match their
  *        corresponding #ifdef or #ifndef
  *
- *   The first two items above require better buffer handling, which would
- *     also make it possible to handle all "dodgy" directives correctly.
+ *   These require better buffer handling, which would also make
+ *   it possible to handle all "dodgy" directives correctly.
  */
 
-#include <errno.h>
-#include <ctype.h>
-#include <stdarg.h>
-#include <stdbool.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-#include <unistd.h>
-
-/* Avoid err.h since uClibc can disable these things */
-#define vwarnx(fmt, args)   ({ fprintf(stderr, "unifdef: "); vfprintf(stderr, fmt, args); fprintf(stderr, "\n"); })
-#define warnx(fmt, args...) fprintf(stderr, "unifdef: " fmt "\n", ## args)
-#define errx(exit_code, fmt, args...) ({ warnx(fmt, ## args); exit(exit_code); })
-#define err(exit_code, fmt, args...)  errx(exit_code, fmt ": %s", ## args, strerror(errno))
+#include "unifdef.h"
 
-size_t strlcpy(char *dst, const char *src, size_t siz);
+static const char copyright[] =
+    #include "version.h"
+    "@(#) $Author: Tony Finch (dot@dotat.at) $\n"
+    "@(#) $URL: http://dotat.at/prog/unifdef $\n"
+;
 
 /* types of input lines: */
 typedef enum {
@@ -96,6 +67,7 @@ typedef enum {
 	LT_DODGY_LAST = LT_DODGY + LT_ENDIF,
 	LT_PLAIN,		/* ordinary line */
 	LT_EOF,			/* end of file */
+	LT_ERROR,		/* unevaluable #if */
 	LT_COUNT
 } Linetype;
 
@@ -106,9 +78,12 @@ static char const * const linetype_name[] = {
 	"DODGY IF", "DODGY TRUE", "DODGY FALSE",
 	"DODGY ELIF", "DODGY ELTRUE", "DODGY ELFALSE",
 	"DODGY ELSE", "DODGY ENDIF",
-	"PLAIN", "EOF"
+	"PLAIN", "EOF", "ERROR"
 };
 
+#define linetype_if2elif(lt) ((Linetype)(lt - LT_IF + LT_ELIF))
+#define linetype_2dodgy(lt) ((Linetype)(lt + LT_DODGY))
+
 /* state of #if processing */
 typedef enum {
 	IS_OUTSIDE,
@@ -162,7 +137,7 @@ static char const * const linestate_name[] = {
  */
 #define	MAXDEPTH        64			/* maximum #if nesting */
 #define	MAXLINE         4096			/* maximum length of line */
-#define	MAXSYMS         4096			/* maximum number of symbols */
+#define	MAXSYMS         16384			/* maximum number of symbols */
 
 /*
  * Sometimes when editing a keyword the replacement text is longer, so
@@ -174,13 +149,17 @@ static char const * const linestate_name[] = {
  * Globals.
  */
 
+static bool             compblank;		/* -B: compress blank lines */
+static bool             lnblank;		/* -b: blank deleted lines */
 static bool             complement;		/* -c: do the complement */
 static bool             debugging;		/* -d: debugging reports */
+static bool             inplace;		/* -m: modify in place */
 static bool             iocccok;		/* -e: fewer IOCCC errors */
+static bool             strictlogic;		/* -K: keep ambiguous #ifs */
 static bool             killconsts;		/* -k: eval constant #ifs */
-static bool             lnblank;		/* -l: blank deleted lines */
 static bool             lnnum;			/* -n: add #line directives */
 static bool             symlist;		/* -s: output symbol list */
+static bool             symdepth;		/* -S: output symbol depth */
 static bool             text;			/* -t: this is a text file */
 
 static const char      *symname[MAXSYMS];	/* symbol name */
@@ -191,10 +170,28 @@ static int              nsyms;			/* number of symbols */
 static FILE            *input;			/* input file pointer */
 static const char      *filename;		/* input file name */
 static int              linenum;		/* current line number */
+static const char      *linefile;		/* file name for #line */
+static FILE            *output;			/* output file pointer */
+static const char      *ofilename;		/* output file name */
+static const char      *backext;		/* backup extension */
+static char            *tempname;		/* avoid splatting input */
 
 static char             tline[MAXLINE+EDITSLOP];/* input buffer plus space */
 static char            *keyword;		/* used for editing #elif's */
 
+/*
+ * When processing a file, the output's newline style will match the
+ * input's, and unifdef correctly handles CRLF or LF endings whatever
+ * the platform's native style. The stdio streams are opened in binary
+ * mode to accommodate platforms whose native newline style is CRLF.
+ * When the output isn't a processed input file (when it is error /
+ * debug / diagnostic messages) then unifdef uses native line endings.
+ */
+
+static const char      *newline;		/* input file format */
+static const char       newline_unix[] = "\n";
+static const char       newline_crlf[] = "\r\n";
+
 static Comment_state    incomment;		/* comment parser state */
 static Line_state       linestate;		/* #if line parser state */
 static Ifstate          ifstate[MAXDEPTH];	/* #if processor state */
@@ -202,31 +199,52 @@ static bool             ignoring[MAXDEPTH];	/* ignore comments state */
 static int              stifline[MAXDEPTH];	/* start of current #if */
 static int              depth;			/* current #if nesting */
 static int              delcount;		/* count of deleted lines */
-static bool             keepthis;		/* don't delete constant #if */
+static unsigned         blankcount;		/* count of blank lines */
+static unsigned         blankmax;		/* maximum recent blankcount */
+static bool             constexpr;		/* constant #if expression */
+static bool             zerosyms;		/* to format symdepth output */
+static bool             firstsym;		/* ditto */
 
+static int              exitmode;		/* exit status mode */
 static int              exitstat;		/* program exit status */
 
-static void             addsym(bool, bool, char *);
+static void             addsym1(bool, bool, char *);
+static void             addsym2(bool, const char *, const char *);
+static char            *astrcat(const char *, const char *);
+static void             cleantemp(void);
+static void             closeio(void);
 static void             debug(const char *, ...);
+static void             debugsym(const char *, int);
+static bool             defundef(void);
+static void             defundefile(const char *);
 static void             done(void);
 static void             error(const char *);
-static int              findsym(const char *);
+static int              findsym(const char **);
 static void             flushline(bool);
-static Linetype         get_line(void);
+static void             hashline(void);
+static void             help(void);
 static Linetype         ifeval(const char **);
 static void             ignoreoff(void);
 static void             ignoreon(void);
+static void             indirectsym(void);
 static void             keywordedit(const char *);
+static const char      *matchsym(const char *, const char *);
 static void             nest(void);
+static Linetype         parseline(void);
 static void             process(void);
+static void             processinout(const char *, const char *);
+static const char      *skipargs(const char *);
 static const char      *skipcomment(const char *);
+static const char      *skiphash(void);
+static const char      *skipline(const char *);
 static const char      *skipsym(const char *);
 static void             state(Ifstate);
-static int              strlcmp(const char *, const char *, size_t);
 static void             unnest(void);
 static void             usage(void);
+static void             version(void);
+static const char      *xstrdup(const char *, const char *);
 
-#define endsym(c) (!isalpha((unsigned char)c) && !isdigit((unsigned char)c) && c != '_')
+#define endsym(c) (!isalnum((unsigned char)c) && c != '_')
 
 /*
  * The main program.
@@ -236,7 +254,7 @@ main(int argc, char *argv[])
 {
 	int opt;
 
-	while ((opt = getopt(argc, argv, "i:D:U:I:cdeklnst")) != -1)
+	while ((opt = getopt(argc, argv, "i:D:U:f:I:M:o:x:bBcdehKklmnsStV")) != -1)
 		switch (opt) {
 		case 'i': /* treat stuff controlled by these symbols as text */
 			/*
@@ -246,20 +264,26 @@ main(int argc, char *argv[])
 			 */
 			opt = *optarg++;
 			if (opt == 'D')
-				addsym(true, true, optarg);
+				addsym1(true, true, optarg);
 			else if (opt == 'U')
-				addsym(true, false, optarg);
+				addsym1(true, false, optarg);
 			else
 				usage();
 			break;
 		case 'D': /* define a symbol */
-			addsym(false, true, optarg);
+			addsym1(false, true, optarg);
 			break;
 		case 'U': /* undef a symbol */
-			addsym(false, false, optarg);
+			addsym1(false, false, optarg);
 			break;
-		case 'I':
-			/* no-op for compatibility with cpp */
+		case 'I': /* no-op for compatibility with cpp */
+			break;
+		case 'b': /* blank deleted lines instead of omitting them */
+		case 'l': /* backwards compatibility */
+			lnblank = true;
+			break;
+		case 'B': /* compress blank lines around removed section */
+			compblank = true;
 			break;
 		case 'c': /* treat -D as -U and vice versa */
 			complement = true;
@@ -270,50 +294,210 @@ main(int argc, char *argv[])
 		case 'e': /* fewer errors from dodgy lines */
 			iocccok = true;
 			break;
+		case 'f': /* definitions file */
+			defundefile(optarg);
+			break;
+		case 'h':
+			help();
+			break;
+		case 'K': /* keep ambiguous #ifs */
+			strictlogic = true;
+			break;
 		case 'k': /* process constant #ifs */
 			killconsts = true;
 			break;
-		case 'l': /* blank deleted lines instead of omitting them */
-			lnblank = true;
+		case 'm': /* modify in place */
+			inplace = true;
+			break;
+		case 'M': /* modify in place and keep backup */
+			inplace = true;
+			backext = optarg;
 			break;
 		case 'n': /* add #line directive after deleted lines */
 			lnnum = true;
 			break;
+		case 'o': /* output to a file */
+			ofilename = optarg;
+			break;
 		case 's': /* only output list of symbols that control #ifs */
 			symlist = true;
 			break;
+		case 'S': /* list symbols with their nesting depth */
+			symlist = symdepth = true;
+			break;
 		case 't': /* don't parse C comments */
 			text = true;
 			break;
+		case 'V':
+			version();
+			break;
+		case 'x':
+			exitmode = atoi(optarg);
+			if(exitmode < 0 || exitmode > 2)
+				usage();
+			break;
 		default:
 			usage();
 		}
 	argc -= optind;
 	argv += optind;
-	if (argc > 1) {
-		errx(2, "can only do one file");
-	} else if (argc == 1 && strcmp(*argv, "-") != 0) {
-		filename = *argv;
-		input = fopen(filename, "r");
-		if (input == NULL)
-			err(2, "can't open %s", filename);
-	} else {
+	if (compblank && lnblank)
+		errx(2, "-B and -b are mutually exclusive");
+	if (symlist && (ofilename != NULL || inplace || argc > 1))
+		errx(2, "-s only works with one input file");
+	if (argc > 1 && ofilename != NULL)
+		errx(2, "-o cannot be used with multiple input files");
+	if (argc > 1 && !inplace)
+		errx(2, "multiple input files require -m or -M");
+	if (argc == 0)
+		argc = 1;
+	if (argc == 1 && !inplace && ofilename == NULL)
+		ofilename = "-";
+	indirectsym();
+
+	atexit(cleantemp);
+	if (ofilename != NULL)
+		processinout(*argv, ofilename);
+	else while (argc-- > 0) {
+		processinout(*argv, *argv);
+		argv++;
+	}
+	switch(exitmode) {
+	case(0): exit(exitstat);
+	case(1): exit(!exitstat);
+	case(2): exit(0);
+	default: abort(); /* bug */
+	}
+}
+
+/*
+ * File logistics.
+ */
+static void
+processinout(const char *ifn, const char *ofn)
+{
+	struct stat st;
+
+	if (ifn == NULL || strcmp(ifn, "-") == 0) {
 		filename = "[stdin]";
-		input = stdin;
+		linefile = NULL;
+		input = fbinmode(stdin);
+	} else {
+		filename = ifn;
+		linefile = ifn;
+		input = fopen(ifn, "rb");
+		if (input == NULL)
+			err(2, "can't open %s", ifn);
+	}
+	if (strcmp(ofn, "-") == 0) {
+		output = fbinmode(stdout);
+		process();
+		return;
+	}
+	if (stat(ofn, &st) < 0) {
+		output = fopen(ofn, "wb");
+		if (output == NULL)
+			err(2, "can't create %s", ofn);
+		process();
+		return;
 	}
+
+	tempname = astrcat(ofn, ".XXXXXX");
+	output = mktempmode(tempname, st.st_mode);
+	if (output == NULL)
+		err(2, "can't create %s", tempname);
+
 	process();
-	debug("bug at line %d", __LINE__);
-	abort(); /* bug */
+
+	if (backext != NULL) {
+		char *backname = astrcat(ofn, backext);
+		if (rename(ofn, backname) < 0)
+			err(2, "can't rename \"%s\" to \"%s\"", ofn, backname);
+		free(backname);
+	}
+	if (replace(tempname, ofn) < 0)
+		err(2, "can't rename \"%s\" to \"%s\"", tempname, ofn);
+	free(tempname);
+	tempname = NULL;
+}
+
+/*
+ * For cleaning up if there is an error.
+ */
+static void
+cleantemp(void)
+{
+	if (tempname != NULL)
+		remove(tempname);
+}
+
+/*
+ * Self-identification functions.
+ */
+
+static void
+version(void)
+{
+	const char *c = copyright;
+	for (;;) {
+		while (*++c != '$')
+			if (*c == '\0')
+				exit(0);
+		while (*++c != '$')
+			putc(*c, stderr);
+		putc('\n', stderr);
+	}
+}
+
+static void
+synopsis(FILE *fp)
+{
+	fprintf(fp,
+	    "usage:	unifdef [-bBcdehKkmnsStV] [-x{012}] [-Mext] [-opath] \\\n"
+	    "		[-[i]Dsym[=val]] [-[i]Usym] [-fpath] ... [file] ...\n");
 }
 
 static void
 usage(void)
 {
-	fprintf(stderr, "usage: unifdef [-cdeklnst] [-Ipath]"
-	    " [-Dsym[=val]] [-Usym] [-iDsym[=val]] [-iUsym] ... [file]\n");
+	synopsis(stderr);
 	exit(2);
 }
 
+static void
+help(void)
+{
+	synopsis(stdout);
+	printf(
+	    "	-Dsym=val  define preprocessor symbol with given value\n"
+	    "	-Dsym      define preprocessor symbol with value 1\n"
+	    "	-Usym	   preprocessor symbol is undefined\n"
+	    "	-iDsym=val \\  ignore C strings and comments\n"
+	    "	-iDsym      ) in sections controlled by these\n"
+	    "	-iUsym	   /  preprocessor symbols\n"
+	    "	-fpath	file containing #define and #undef directives\n"
+	    "	-b	blank lines instead of deleting them\n"
+	    "	-B	compress blank lines around deleted section\n"
+	    "	-c	complement (invert) keep vs. delete\n"
+	    "	-d	debugging mode\n"
+	    "	-e	ignore multiline preprocessor directives\n"
+	    "	-h	print help\n"
+	    "	-Ipath	extra include file path (ignored)\n"
+	    "	-K	disable && and || short-circuiting\n"
+	    "	-k	process constant #if expressions\n"
+	    "	-Mext	modify in place and keep backups\n"
+	    "	-m	modify input files in place\n"
+	    "	-n	add #line directives to output\n"
+	    "	-opath	output file name\n"
+	    "	-S	list #if control symbols with nesting\n"
+	    "	-s	list #if control symbols\n"
+	    "	-t	ignore C strings and comments\n"
+	    "	-V	print version\n"
+	    "	-x{012}	exit status mode\n"
+	);
+	exit(0);
+}
+
 /*
  * A state transition function alters the global #if processing state
  * in a particular way. The table below is indexed by the current
@@ -327,7 +511,8 @@ usage(void)
  * When we have processed a group that starts off with a known-false
  * #if/#elif sequence (which has therefore been deleted) followed by a
  * #elif that we don't understand and therefore must keep, we edit the
- * latter into a #if to keep the nesting correct.
+ * latter into a #if to keep the nesting correct. We use memcpy() to
+ * overwrite the 4 byte token "elif" with "if  " without a '\0' byte.
  *
  * When we find a true #elif in a group, the following block will
  * always be kept and the rest of the sequence after the next #elif or
@@ -380,75 +565,66 @@ static void Oelif (void) { if (!iocccok) Eioccc(); Pelif(); }
 static void Idrop (void) { Fdrop();  ignoreon(); }
 static void Itrue (void) { Ftrue();  ignoreon(); }
 static void Ifalse(void) { Ffalse(); ignoreon(); }
-/* edit this line */
-static void Mpass (void) { strncpy(keyword, "if  ", 4); Pelif(); }
-static void Mtrue (void) { keywordedit("else\n");  state(IS_TRUE_MIDDLE); }
-static void Melif (void) { keywordedit("endif\n"); state(IS_FALSE_TRAILER); }
-static void Melse (void) { keywordedit("endif\n"); state(IS_FALSE_ELSE); }
+/* modify this line */
+static void Mpass (void) { memcpy(keyword, "if  ", 4); Pelif(); }
+static void Mtrue (void) { keywordedit("else");  state(IS_TRUE_MIDDLE); }
+static void Melif (void) { keywordedit("endif"); state(IS_FALSE_TRAILER); }
+static void Melse (void) { keywordedit("endif"); state(IS_FALSE_ELSE); }
 
 static state_fn * const trans_table[IS_COUNT][LT_COUNT] = {
 /* IS_OUTSIDE */
 { Itrue, Ifalse,Fpass, Ftrue, Ffalse,Eelif, Eelif, Eelif, Eelse, Eendif,
   Oiffy, Oiffy, Fpass, Oif,   Oif,   Eelif, Eelif, Eelif, Eelse, Eendif,
-  print, done },
+  print, done,  abort },
 /* IS_FALSE_PREFIX */
 { Idrop, Idrop, Fdrop, Fdrop, Fdrop, Mpass, Strue, Sfalse,Selse, Dendif,
   Idrop, Idrop, Fdrop, Fdrop, Fdrop, Mpass, Eioccc,Eioccc,Eioccc,Eioccc,
-  drop,  Eeof },
+  drop,  Eeof,  abort },
 /* IS_TRUE_PREFIX */
 { Itrue, Ifalse,Fpass, Ftrue, Ffalse,Dfalse,Dfalse,Dfalse,Delse, Dendif,
   Oiffy, Oiffy, Fpass, Oif,   Oif,   Eioccc,Eioccc,Eioccc,Eioccc,Eioccc,
-  print, Eeof },
+  print, Eeof,  abort },
 /* IS_PASS_MIDDLE */
 { Itrue, Ifalse,Fpass, Ftrue, Ffalse,Pelif, Mtrue, Delif, Pelse, Pendif,
   Oiffy, Oiffy, Fpass, Oif,   Oif,   Pelif, Oelif, Oelif, Pelse, Pendif,
-  print, Eeof },
+  print, Eeof,  abort },
 /* IS_FALSE_MIDDLE */
 { Idrop, Idrop, Fdrop, Fdrop, Fdrop, Pelif, Mtrue, Delif, Pelse, Pendif,
   Idrop, Idrop, Fdrop, Fdrop, Fdrop, Eioccc,Eioccc,Eioccc,Eioccc,Eioccc,
-  drop,  Eeof },
+  drop,  Eeof,  abort },
 /* IS_TRUE_MIDDLE */
 { Itrue, Ifalse,Fpass, Ftrue, Ffalse,Melif, Melif, Melif, Melse, Pendif,
   Oiffy, Oiffy, Fpass, Oif,   Oif,   Eioccc,Eioccc,Eioccc,Eioccc,Pendif,
-  print, Eeof },
+  print, Eeof,  abort },
 /* IS_PASS_ELSE */
 { Itrue, Ifalse,Fpass, Ftrue, Ffalse,Eelif, Eelif, Eelif, Eelse, Pendif,
   Oiffy, Oiffy, Fpass, Oif,   Oif,   Eelif, Eelif, Eelif, Eelse, Pendif,
-  print, Eeof },
+  print, Eeof,  abort },
 /* IS_FALSE_ELSE */
 { Idrop, Idrop, Fdrop, Fdrop, Fdrop, Eelif, Eelif, Eelif, Eelse, Dendif,
   Idrop, Idrop, Fdrop, Fdrop, Fdrop, Eelif, Eelif, Eelif, Eelse, Eioccc,
-  drop,  Eeof },
+  drop,  Eeof,  abort },
 /* IS_TRUE_ELSE */
 { Itrue, Ifalse,Fpass, Ftrue, Ffalse,Eelif, Eelif, Eelif, Eelse, Dendif,
   Oiffy, Oiffy, Fpass, Oif,   Oif,   Eelif, Eelif, Eelif, Eelse, Eioccc,
-  print, Eeof },
+  print, Eeof,  abort },
 /* IS_FALSE_TRAILER */
 { Idrop, Idrop, Fdrop, Fdrop, Fdrop, Dfalse,Dfalse,Dfalse,Delse, Dendif,
   Idrop, Idrop, Fdrop, Fdrop, Fdrop, Dfalse,Dfalse,Dfalse,Delse, Eioccc,
-  drop,  Eeof }
+  drop,  Eeof,  abort }
 /*TRUEI  FALSEI IF     TRUE   FALSE  ELIF   ELTRUE ELFALSE ELSE  ENDIF
   TRUEI  FALSEI IF     TRUE   FALSE  ELIF   ELTRUE ELFALSE ELSE  ENDIF (DODGY)
-  PLAIN  EOF */
+  PLAIN  EOF    ERROR */
 };
 
 /*
  * State machine utility functions
  */
 static void
-done(void)
-{
-	if (incomment)
-		error("EOF in comment");
-	exit(exitstat);
-}
-static void
 ignoreoff(void)
 {
-	if (depth == 0) {
-		debug("bug at line %d", __LINE__);
+	if (depth == 0)
 		abort(); /* bug */
-	}
 	ignoring[depth] = ignoring[depth-1];
 }
 static void
@@ -459,31 +635,25 @@ ignoreon(void)
 static void
 keywordedit(const char *replacement)
 {
-	size_t size = tline + sizeof(tline) - keyword;
-	char *dst = keyword;
-	const char *src = replacement;
-	if (size != 0) {
-		while ((--size != 0) && (*src != '\0'))
-			*dst++ = *src++;
-		*dst = '\0';
-	}
+	snprintf(keyword, tline + sizeof(tline) - keyword,
+	    "%s%s", replacement, newline);
 	print();
 }
 static void
 nest(void)
 {
-	depth += 1;
-	if (depth >= MAXDEPTH)
+	if (depth > MAXDEPTH-1)
+		abort(); /* bug */
+	if (depth == MAXDEPTH-1)
 		error("Too many levels of nesting");
+	depth += 1;
 	stifline[depth] = linenum;
 }
 static void
 unnest(void)
 {
-	if (depth == 0) {
-		debug("bug at line %d", __LINE__);
+	if (depth == 0)
 		abort(); /* bug */
-	}
 	depth -= 1;
 }
 static void
@@ -493,7 +663,20 @@ state(Ifstate is)
 }
 
 /*
+ * The last state transition function. When this is called,
+ * lineval == LT_EOF, so the process() loop will terminate.
+ */
+static void
+done(void)
+{
+	if (incomment)
+		error("EOF in comment");
+	closeio();
+}
+
+/*
  * Write a line to the output or not, according to command line options.
+ * If writing fails, closeio() will print the error and exit.
  */
 static void
 flushline(bool keep)
@@ -501,16 +684,58 @@ flushline(bool keep)
 	if (symlist)
 		return;
 	if (keep ^ complement) {
-		if (lnnum && delcount > 0)
-			printf("#line %d\n", linenum);
-		fputs(tline, stdout);
-		delcount = 0;
+		bool blankline = tline[strspn(tline, " \t\r\n")] == '\0';
+		if (blankline && compblank && blankcount != blankmax) {
+			delcount += 1;
+			blankcount += 1;
+		} else {
+			if (lnnum && delcount > 0)
+				hashline();
+			if (fputs(tline, output) == EOF)
+				closeio();
+			delcount = 0;
+			blankmax = blankcount = blankline ? blankcount + 1 : 0;
+		}
 	} else {
-		if (lnblank)
-			putc('\n', stdout);
+		if (lnblank && fputs(newline, output) == EOF)
+			closeio();
 		exitstat = 1;
 		delcount += 1;
+		blankcount = 0;
 	}
+	if (debugging && fflush(output) == EOF)
+		closeio();
+}
+
+/*
+ * Format of #line directives depends on whether we know the input filename.
+ */
+static void
+hashline(void)
+{
+	int e;
+
+	if (linefile == NULL)
+		e = fprintf(output, "#line %d%s", linenum, newline);
+	else
+		e = fprintf(output, "#line %d \"%s\"%s",
+		    linenum, linefile, newline);
+	if (e < 0)
+		closeio();
+}
+
+/*
+ * Flush the output and handle errors.
+ */
+static void
+closeio(void)
+{
+	/* Tidy up after findsym(). */
+	if (symdepth && !zerosyms)
+		printf("\n");
+	if (output != NULL && (ferror(output) || fclose(output) == EOF))
+			err(2, "%s: can't write to output", filename);
+	fclose(input);
 }
 
 /*
@@ -519,14 +744,18 @@ flushline(bool keep)
 static void
 process(void)
 {
-	Linetype lineval;
-
-	for (;;) {
-		linenum++;
-		lineval = get_line();
+	Linetype lineval = LT_PLAIN;
+	/* When compressing blank lines, act as if the file
+	   is preceded by a large number of blank lines. */
+	blankmax = blankcount = 1000;
+	zerosyms = true;
+	newline = NULL;
+	linenum = 0;
+	while (lineval != LT_EOF) {
+		lineval = parseline();
 		trans_table[ifstate[depth]][lineval]();
-		debug("process %s -> %s depth %d",
-		    linetype_name[lineval],
+		debug("process line %d %s -> %s depth %d",
+		    linenum, linetype_name[lineval],
 		    ifstate_name[ifstate[depth]], depth);
 	}
 }
@@ -537,104 +766,131 @@ process(void)
  * help from skipcomment().
  */
 static Linetype
-get_line(void)
+parseline(void)
 {
 	const char *cp;
 	int cursym;
-	int kwlen;
 	Linetype retval;
 	Comment_state wascomment;
 
-	if (fgets(tline, MAXLINE, input) == NULL)
-		return (LT_EOF);
-	retval = LT_PLAIN;
 	wascomment = incomment;
-	cp = skipcomment(tline);
-	if (linestate == LS_START) {
-		if (*cp == '#') {
-			linestate = LS_HASH;
-			cp = skipcomment(cp + 1);
-		} else if (*cp != '\0')
-			linestate = LS_DIRTY;
+	cp = skiphash();
+	if (cp == NULL)
+		return (LT_EOF);
+	if (newline == NULL) {
+		if (strrchr(tline, '\n') == strrchr(tline, '\r') + 1)
+			newline = newline_crlf;
+		else
+			newline = newline_unix;
 	}
-	if (!incomment && linestate == LS_HASH) {
-		keyword = tline + (cp - tline);
-		cp = skipsym(cp);
-		kwlen = cp - keyword;
-		/* no way can we deal with a continuation inside a keyword */
-		if (strncmp(cp, "\\\n", 2) == 0)
-			Eioccc();
-		if (strlcmp("ifdef", keyword, kwlen) == 0 ||
-		    strlcmp("ifndef", keyword, kwlen) == 0) {
-			cp = skipcomment(cp);
-			if ((cursym = findsym(cp)) < 0)
-				retval = LT_IF;
-			else {
-				retval = (keyword[2] == 'n')
-				    ? LT_FALSE : LT_TRUE;
-				if (value[cursym] == NULL)
-					retval = (retval == LT_TRUE)
-					    ? LT_FALSE : LT_TRUE;
-				if (ignore[cursym])
-					retval = (retval == LT_TRUE)
-					    ? LT_TRUEI : LT_FALSEI;
-			}
-			cp = skipsym(cp);
-		} else if (strlcmp("if", keyword, kwlen) == 0)
-			retval = ifeval(&cp);
-		else if (strlcmp("elif", keyword, kwlen) == 0)
-			retval = ifeval(&cp) - LT_IF + LT_ELIF;
-		else if (strlcmp("else", keyword, kwlen) == 0)
-			retval = LT_ELSE;
-		else if (strlcmp("endif", keyword, kwlen) == 0)
-			retval = LT_ENDIF;
+	if (*cp == '\0') {
+		retval = LT_PLAIN;
+		goto done;
+	}
+	keyword = tline + (cp - tline);
+	if ((cp = matchsym("ifdef", keyword)) != NULL ||
+	    (cp = matchsym("ifndef", keyword)) != NULL) {
+		cp = skipcomment(cp);
+		if ((cursym = findsym(&cp)) < 0)
+			retval = LT_IF;
 		else {
-			linestate = LS_DIRTY;
-			retval = LT_PLAIN;
+			retval = (keyword[2] == 'n')
+			    ? LT_FALSE : LT_TRUE;
+			if (value[cursym] == NULL)
+				retval = (retval == LT_TRUE)
+				    ? LT_FALSE : LT_TRUE;
+			if (ignore[cursym])
+				retval = (retval == LT_TRUE)
+				    ? LT_TRUEI : LT_FALSEI;
 		}
-		cp = skipcomment(cp);
-		if (*cp != '\0') {
+	} else if ((cp = matchsym("if", keyword)) != NULL)
+		retval = ifeval(&cp);
+	else if ((cp = matchsym("elif", keyword)) != NULL)
+		retval = linetype_if2elif(ifeval(&cp));
+	else if ((cp = matchsym("else", keyword)) != NULL)
+		retval = LT_ELSE;
+	else if ((cp = matchsym("endif", keyword)) != NULL)
+		retval = LT_ENDIF;
+	else {
+		cp = skipsym(keyword);
+		/* no way can we deal with a continuation inside a keyword */
+		if (strncmp(cp, "\\\r\n", 3) == 0 ||
+		    strncmp(cp, "\\\n", 2) == 0)
+			Eioccc();
+		cp = skipline(cp);
+		retval = LT_PLAIN;
+		goto done;
+	}
+	cp = skipcomment(cp);
+	if (*cp != '\0') {
+		cp = skipline(cp);
+		if (retval == LT_TRUE || retval == LT_FALSE ||
+		    retval == LT_TRUEI || retval == LT_FALSEI)
+			retval = LT_IF;
+		if (retval == LT_ELTRUE || retval == LT_ELFALSE)
+			retval = LT_ELIF;
+	}
+	/* the following can happen if the last line of the file lacks a
+	   newline or if there is too much whitespace in a directive */
+	if (linestate == LS_HASH) {
+		long len = cp - tline;
+		if (fgets(tline + len, MAXLINE - len, input) == NULL) {
+			if (ferror(input))
+				err(2, "can't read %s", filename);
+			/* append the missing newline at eof */
+			strcpy(tline + len, newline);
+			cp += strlen(newline);
+			linestate = LS_START;
+		} else {
 			linestate = LS_DIRTY;
-			if (retval == LT_TRUE || retval == LT_FALSE ||
-			    retval == LT_TRUEI || retval == LT_FALSEI)
-				retval = LT_IF;
-			if (retval == LT_ELTRUE || retval == LT_ELFALSE)
-				retval = LT_ELIF;
-		}
-		if (retval != LT_PLAIN && (wascomment || incomment)) {
-			retval += LT_DODGY;
-			if (incomment)
-				linestate = LS_DIRTY;
 		}
-		/* skipcomment should have changed the state */
-// Hmm hppens sometimes on valid files
-//		if (linestate == LS_HASH) {
-//			debug("bug at line %d", __LINE__);
-//			abort(); /* bug */
-//		}
 	}
-	if (linestate == LS_DIRTY) {
-		while (*cp != '\0')
-			cp = skipcomment(cp + 1);
+	if (retval != LT_PLAIN && (wascomment || linestate != LS_START)) {
+		retval = linetype_2dodgy(retval);
+		linestate = LS_DIRTY;
 	}
-	debug("parser %s comment %s line",
+done:
+	debug("parser line %d state %s comment %s line", linenum,
 	    comment_name[incomment], linestate_name[linestate]);
 	return (retval);
 }
 
 /*
  * These are the binary operators that are supported by the expression
- * evaluator. Note that if support for division is added then we also
- * need short-circuiting booleans because of divide-by-zero.
+ * evaluator.
  */
-static int op_lt(int a, int b) { return (a < b); }
-static int op_gt(int a, int b) { return (a > b); }
-static int op_le(int a, int b) { return (a <= b); }
-static int op_ge(int a, int b) { return (a >= b); }
-static int op_eq(int a, int b) { return (a == b); }
-static int op_ne(int a, int b) { return (a != b); }
-static int op_or(int a, int b) { return (a || b); }
-static int op_and(int a, int b) { return (a && b); }
+static Linetype op_strict(long *p, long v, Linetype at, Linetype bt) {
+	if(at == LT_IF || bt == LT_IF) return (LT_IF);
+	return (*p = v, v ? LT_TRUE : LT_FALSE);
+}
+static Linetype op_lt(long *p, Linetype at, long a, Linetype bt, long b) {
+	return op_strict(p, a < b, at, bt);
+}
+static Linetype op_gt(long *p, Linetype at, long a, Linetype bt, long b) {
+	return op_strict(p, a > b, at, bt);
+}
+static Linetype op_le(long *p, Linetype at, long a, Linetype bt, long b) {
+	return op_strict(p, a <= b, at, bt);
+}
+static Linetype op_ge(long *p, Linetype at, long a, Linetype bt, long b) {
+	return op_strict(p, a >= b, at, bt);
+}
+static Linetype op_eq(long *p, Linetype at, long a, Linetype bt, long b) {
+	return op_strict(p, a == b, at, bt);
+}
+static Linetype op_ne(long *p, Linetype at, long a, Linetype bt, long b) {
+	return op_strict(p, a != b, at, bt);
+}
+static Linetype op_or(long *p, Linetype at, long a, Linetype bt, long b) {
+	if (!strictlogic && (at == LT_TRUE || bt == LT_TRUE))
+		return (*p = 1, LT_TRUE);
+	return op_strict(p, a || b, at, bt);
+}
+static Linetype op_and(long *p, Linetype at, long a, Linetype bt, long b) {
+	if (!strictlogic && (at == LT_FALSE || bt == LT_FALSE))
+		return (*p = 0, LT_FALSE);
+	return op_strict(p, a && b, at, bt);
+}
 
 /*
  * An evaluation function takes three arguments, as follows: (1) a pointer to
@@ -643,12 +899,12 @@ static int op_and(int a, int b) { return (a && b); }
  * value of the expression; and (3) a pointer to a char* that points to the
  * expression to be evaluated and that is updated to the end of the expression
  * when evaluation is complete. The function returns LT_FALSE if the value of
- * the expression is zero, LT_TRUE if it is non-zero, or LT_IF if the
- * expression could not be evaluated.
+ * the expression is zero, LT_TRUE if it is non-zero, LT_IF if the expression
+ * depends on an unknown symbol, or LT_ERROR if there is a parse failure.
  */
 struct ops;
 
-typedef Linetype eval_fn(const struct ops *, int *, const char **);
+typedef Linetype eval_fn(const struct ops *, long *, const char **);
 
 static eval_fn eval_table, eval_unary;
 
@@ -659,139 +915,139 @@ static eval_fn eval_table, eval_unary;
  * element of the table. Innermost expressions have special non-table-driven
  * handling.
  */
-static const struct ops {
+struct op {
+	const char *str;
+	Linetype (*fn)(long *, Linetype, long, Linetype, long);
+};
+struct ops {
 	eval_fn *inner;
-	struct op {
-		const char *str;
-		int short_circuit_val;
-		int (*fn)(int, int);
-	} op[5];
-} eval_ops[] = {
-	{ eval_table, { { "||", 1, op_or } } },
-	{ eval_table, { { "&&", 0, op_and } } },
-	{ eval_table, { { "==", -1, op_eq },
-			{ "!=", -1, op_ne } } },
-	{ eval_unary, { { "<=", -1, op_le },
-			{ ">=", -1, op_ge },
-			{ "<", -1, op_lt },
-			{ ">", -1, op_gt } } }
+	struct op op[5];
 };
+static const struct ops eval_ops[] = {
+	{ eval_table, { { "||", op_or } } },
+	{ eval_table, { { "&&", op_and } } },
+	{ eval_table, { { "==", op_eq },
+			{ "!=", op_ne } } },
+	{ eval_unary, { { "<=", op_le },
+			{ ">=", op_ge },
+			{ "<", op_lt },
+			{ ">", op_gt } } }
+};
+
+/* Current operator precedence level */
+static long prec(const struct ops *ops)
+{
+	return (ops - eval_ops);
+}
 
 /*
- * Function for evaluating the innermost parts of expressions, viz.
- * "!expr", "(expr)", "defined(symbol)", "defined symbol", "symbol", "number".
- * We reset the keepthis flag when we find a non-constant subexpression.
+ * Function for evaluating the innermost parts of expressions,
+ * viz. !expr (expr) number defined(symbol) symbol
+ * We reset the constexpr flag in the last two cases.
  */
-// TODO: we use LT_IF both as "I don't know whether it's false or true"
-// (example: "#if defined FOO") and when we see syntax error
-// (example: "#if (1 || 2" - no closing paren!), but this is wrong.
-// Binary && and || need to distinguish these cases in order to handle this:
-// "#if defined KNOWN_UNDEFINED && FOO" - discard
-// "#if defined KNOWN_UNDEFINED && (syntax_error_here" - do not discard!
 static Linetype
-eval_unary(const struct ops *ops, int *valp, const char **cpp)
+eval_unary(const struct ops *ops, long *valp, const char **cpp)
 {
 	const char *cp;
 	char *ep;
 	int sym;
+	bool defparen;
+	Linetype lt;
 
 	cp = skipcomment(*cpp);
 	if (*cp == '!') {
-		debug("eval%d !", ops - eval_ops);
+		debug("eval%d !", prec(ops));
 		cp++;
-		if (eval_unary(ops, valp, &cp) == LT_IF) {
-			*cpp = cp;
-			return (LT_IF);
+		lt = eval_unary(ops, valp, &cp);
+		if (lt == LT_ERROR)
+			return (LT_ERROR);
+		if (lt != LT_IF) {
+			*valp = !*valp;
+			lt = *valp ? LT_TRUE : LT_FALSE;
 		}
-		*valp = !*valp;
-
 	} else if (*cp == '(') {
-		Linetype expr_res;
-
 		cp++;
-		debug("eval%d (%s", ops - eval_ops, cp);
-		expr_res = eval_table(eval_ops, valp, &cp);
+		debug("eval%d (", prec(ops));
+		lt = eval_table(eval_ops, valp, &cp);
+		if (lt == LT_ERROR)
+			return (LT_ERROR);
 		cp = skipcomment(cp);
-		*cpp = cp;
 		if (*cp++ != ')')
-			return (LT_IF);
-		*cpp = cp;
-		if (expr_res == LT_IF)
-			return (LT_IF);
-
+			return (LT_ERROR);
 	} else if (isdigit((unsigned char)*cp)) {
-		debug("eval%d number", ops - eval_ops);
+		debug("eval%d number", prec(ops));
 		*valp = strtol(cp, &ep, 0);
-		cp = skipsym(cp);
-
-	} else if (strncmp(cp, "defined", 7) == 0 && endsym(cp[7])) {
-		bool parens;
-
+		if (ep == cp)
+			return (LT_ERROR);
+		lt = *valp ? LT_TRUE : LT_FALSE;
+		cp = ep;
+	} else if (matchsym("defined", cp) != NULL) {
 		cp = skipcomment(cp+7);
-		debug("eval%d defined '%s'", ops - eval_ops, cp);
-		parens = (*cp == '(');
-		if (parens)
+		if (*cp == '(') {
 			cp = skipcomment(cp+1);
-		sym = findsym(cp);
-		cp = skipsym(cp);
+			defparen = true;
+		} else {
+			defparen = false;
+		}
+		sym = findsym(&cp);
 		cp = skipcomment(cp);
-		if (parens) {
-			if (*cp != ')')
-				return (LT_IF);
-			cp = skipcomment(cp+1);
+		if (defparen && *cp++ != ')') {
+			debug("eval%d defined missing ')'", prec(ops));
+			return (LT_ERROR);
 		}
-		*cpp = cp;
 		if (sym < 0) {
-			debug("sym not found, returning LT_IF");
-			return (LT_IF);
+			debug("eval%d defined unknown", prec(ops));
+			lt = LT_IF;
+		} else {
+			debug("eval%d defined %s", prec(ops), symname[sym]);
+			*valp = (value[sym] != NULL);
+			lt = *valp ? LT_TRUE : LT_FALSE;
 		}
-		*valp = (value[sym] != NULL);
-		keepthis = false;
-
+		constexpr = false;
 	} else if (!endsym(*cp)) {
-		debug("eval%d symbol", ops - eval_ops);
-		sym = findsym(cp);
-		cp = skipsym(cp);
-		*cpp = cp;
-		if (sym < 0)
-			return (LT_IF);
-		if (value[sym] == NULL)
+		debug("eval%d symbol", prec(ops));
+		sym = findsym(&cp);
+		if (sym < 0) {
+			lt = LT_IF;
+			cp = skipargs(cp);
+		} else if (value[sym] == NULL) {
 			*valp = 0;
-		else {
+			lt = LT_FALSE;
+		} else {
 			*valp = strtol(value[sym], &ep, 0);
 			if (*ep != '\0' || ep == value[sym])
-				return (LT_IF);
+				return (LT_ERROR);
+			lt = *valp ? LT_TRUE : LT_FALSE;
+			cp = skipargs(cp);
 		}
-		keepthis = false;
-
+		constexpr = false;
 	} else {
-		debug("eval%d bad expr", ops - eval_ops);
-		return (LT_IF);
+		debug("eval%d bad expr", prec(ops));
+		return (LT_ERROR);
 	}
 
 	*cpp = cp;
-	debug("eval%d = %d", ops - eval_ops, *valp);
-	return (*valp ? LT_TRUE : LT_FALSE);
+	debug("eval%d = %d", prec(ops), *valp);
+	return (lt);
 }
 
 /*
  * Table-driven evaluation of binary operators.
  */
 static Linetype
-eval_table(const struct ops *ops, int *valp, const char **cpp)
+eval_table(const struct ops *ops, long *valp, const char **cpp)
 {
-	Linetype left_side;
 	const struct op *op;
 	const char *cp;
-	int val;
+	long val;
+	Linetype lt, rt;
 
-	debug("eval%d '%s'", ops - eval_ops, *cpp);
-	left_side = ops->inner(ops+1, valp, cpp);
+	debug("eval%d", prec(ops));
 	cp = *cpp;
-
+	lt = ops->inner(ops+1, valp, &cp);
+	if (lt == LT_ERROR)
+		return (LT_ERROR);
 	for (;;) {
-		Linetype right_side;
-
 		cp = skipcomment(cp);
 		for (op = ops->op; op->str != NULL; op++)
 			if (strncmp(cp, op->str, strlen(op->str)) == 0)
@@ -799,38 +1055,17 @@ eval_table(const struct ops *ops, int *valp, const char **cpp)
 		if (op->str == NULL)
 			break;
 		cp += strlen(op->str);
-		debug("eval%d '%s'", ops - eval_ops, op->str);
-		right_side = ops->inner(ops+1, &val, &cp);
-		*cpp = cp;
-
-		/* If short_circuit_val is 0 or 1, we can ignore
-		 * right side if left size is known, and its value
-		 * (i.e., *valp) is 0 or !0, respectively */
-		if (left_side != LT_IF && op->short_circuit_val == !!*valp) {
-			debug("op->short_circuit_val:%d *valp:%d cp:'%s'",
-					op->short_circuit_val, *valp, cp);
-			*valp = !!*valp;
-			break;
-		}
-		/* Same for the right side */
-		if (right_side != LT_IF && op->short_circuit_val == !!val) {
-			debug("op->short_circuit_val:%d val:%d cp:'%s'",
-					op->short_circuit_val, val, cp);
-			left_side = right_side;
-			*valp = !!val;
-			break;
-		}
-
-		if (left_side == LT_IF || right_side == LT_IF)
-			return (LT_IF);
-		*valp = op->fn(*valp, val);
-		left_side = right_side;
+		debug("eval%d %s", prec(ops), op->str);
+		rt = ops->inner(ops+1, &val, &cp);
+		if (rt == LT_ERROR)
+			return (LT_ERROR);
+		lt = op->fn(valp, lt, *valp, rt, val);
 	}
 
-	debug("eval%d = %d LT_IF:%d", ops - eval_ops, *valp, (left_side == LT_IF));
-	if (left_side == LT_IF)
-		return (LT_IF);
-	return (*valp ? LT_TRUE : LT_FALSE);
+	*cpp = cp;
+	debug("eval%d = %d", prec(ops), *valp);
+	debug("eval%d lt = %s", prec(ops), linetype_name[lt]);
+	return (lt);
 }
 
 /*
@@ -841,14 +1076,56 @@ eval_table(const struct ops *ops, int *valp, const char **cpp)
 static Linetype
 ifeval(const char **cpp)
 {
-	int ret;
-	int val;
+	Linetype ret;
+	long val = 0;
 
 	debug("eval %s", *cpp);
-	keepthis = killconsts ? false : true;
+	constexpr = killconsts ? false : true;
 	ret = eval_table(eval_ops, &val, cpp);
-	debug("val:%d ret:%d keepthis:%d", val, ret, keepthis);
-	return (keepthis ? LT_IF : ret);
+	debug("eval = %d", val);
+	return (constexpr ? LT_IF : ret == LT_ERROR ? LT_IF : ret);
+}
+
+/*
+ * Read a line and examine its initial part to determine if it is a
+ * preprocessor directive. Returns NULL on EOF, or a pointer to a
+ * preprocessor directive name, or a pointer to the zero byte at the
+ * end of the line.
+ */
+static const char *
+skiphash(void)
+{
+	const char *cp;
+
+	linenum++;
+	if (fgets(tline, MAXLINE, input) == NULL) {
+		if (ferror(input))
+			err(2, "can't read %s", filename);
+		else
+			return (NULL);
+	}
+	cp = skipcomment(tline);
+	if (linestate == LS_START && *cp == '#') {
+		linestate = LS_HASH;
+		return (skipcomment(cp + 1));
+	} else if (*cp == '\0') {
+		return (cp);
+	} else {
+		return (skipline(cp));
+	}
+}
+
+/*
+ * Mark a line dirty and consume the rest of it, keeping track of the
+ * lexical state.
+ */
+static const char *
+skipline(const char *cp)
+{
+	linestate = LS_DIRTY;
+	while (*cp != '\0')
+		cp = skipcomment(cp + 1);
+	return (cp);
 }
 
 /*
@@ -869,11 +1146,16 @@ skipcomment(const char *cp)
 	}
 	while (*cp != '\0')
 		/* don't reset to LS_START after a line continuation */
-		if (strncmp(cp, "\\\n", 2) == 0)
+		if (strncmp(cp, "\\\r\n", 3) == 0)
+			cp += 3;
+		else if (strncmp(cp, "\\\n", 2) == 0)
 			cp += 2;
 		else switch (incomment) {
 		case NO_COMMENT:
-			if (strncmp(cp, "/\\\n", 3) == 0) {
+			if (strncmp(cp, "/\\\r\n", 4) == 0) {
+				incomment = STARTING_COMMENT;
+				cp += 4;
+			} else if (strncmp(cp, "/\\\n", 3) == 0) {
 				incomment = STARTING_COMMENT;
 				cp += 3;
 			} else if (strncmp(cp, "/*", 2) == 0) {
@@ -893,7 +1175,7 @@ skipcomment(const char *cp)
 			} else if (strncmp(cp, "\n", 1) == 0) {
 				linestate = LS_START;
 				cp += 1;
-			} else if (strchr(" \t", *cp) != NULL) {
+			} else if (strchr(" \r\t", *cp) != NULL) {
 				cp += 1;
 			} else
 				return (cp);
@@ -925,7 +1207,10 @@ skipcomment(const char *cp)
 				cp += 1;
 			continue;
 		case C_COMMENT:
-			if (strncmp(cp, "*\\\n", 3) == 0) {
+			if (strncmp(cp, "*\\\r\n", 4) == 0) {
+				incomment = FINISHING_COMMENT;
+				cp += 4;
+			} else if (strncmp(cp, "*\\\n", 3) == 0) {
 				incomment = FINISHING_COMMENT;
 				cp += 3;
 			} else if (strncmp(cp, "*/", 2) == 0) {
@@ -954,13 +1239,37 @@ skipcomment(const char *cp)
 				incomment = C_COMMENT;
 			continue;
 		default:
-			debug("bug at line %d", __LINE__);
 			abort(); /* bug */
 		}
 	return (cp);
 }
 
 /*
+ * Skip macro arguments.
+ */
+static const char *
+skipargs(const char *cp)
+{
+	const char *ocp = cp;
+	int level = 0;
+	cp = skipcomment(cp);
+	if (*cp != '(')
+		return (cp);
+	do {
+		if (*cp == '(')
+			level++;
+		if (*cp == ')')
+			level--;
+		cp = skipcomment(cp+1);
+	} while (level != 0 && *cp != '\0');
+	if (level == 0)
+		return (cp);
+	else
+	/* Rewind and re-detect the syntax error later. */
+		return (ocp);
+}
+
+/*
  * Skip over an identifier.
  */
 static const char *
@@ -972,27 +1281,69 @@ skipsym(const char *cp)
 }
 
 /*
- * Look for the symbol in the symbol table. If is is found, we return
+ * Skip whitespace and take a copy of any following identifier.
+ */
+static const char *
+getsym(const char **cpp)
+{
+	const char *cp = *cpp, *sym;
+
+	cp = skipcomment(cp);
+	cp = skipsym(sym = cp);
+	if (cp == sym)
+		return NULL;
+	*cpp = cp;
+	return (xstrdup(sym, cp));
+}
+
+/*
+ * Check that s (a symbol) matches the start of t, and that the
+ * following character in t is not a symbol character. Returns a
+ * pointer to the following character in t if there is a match,
+ * otherwise NULL.
+ */
+static const char *
+matchsym(const char *s, const char *t)
+{
+	while (*s != '\0' && *t != '\0')
+		if (*s != *t)
+			return (NULL);
+		else
+			++s, ++t;
+	if (*s == '\0' && endsym(*t))
+		return(t);
+	else
+		return(NULL);
+}
+
+/*
+ * Look for the symbol in the symbol table. If it is found, we return
  * the symbol table index, else we return -1.
  */
 static int
-findsym(const char *str)
+findsym(const char **strp)
 {
-	const char *cp;
+	const char *str;
 	int symind;
 
-	cp = skipsym(str);
-	if (cp == str)
-		return (-1);
+	str = *strp;
+	*strp = skipsym(str);
 	if (symlist) {
-		printf("%.*s\n", (int)(cp-str), str);
+		if (*strp == str)
+			return (-1);
+		if (symdepth && firstsym)
+			printf("%s%3d", zerosyms ? "" : "\n", depth);
+		firstsym = zerosyms = false;
+		printf("%s%.*s%s",
+		       symdepth ? " " : "",
+		       (int)(*strp-str), str,
+		       symdepth ? "" : "\n");
 		/* we don't care about the value of the symbol */
 		return (0);
 	}
 	for (symind = 0; symind < nsyms; ++symind) {
-		if (strlcmp(symname[symind], str, cp-str) == 0) {
-			debug("findsym %s %s", symname[symind],
-			    value[symind] ? value[symind] : "");
+		if (matchsym(symname[symind], str) != NULL) {
+			debugsym("findsym", symind);
 			return (symind);
 		}
 	}
@@ -1000,51 +1351,196 @@ findsym(const char *str)
 }
 
 /*
+ * Resolve indirect symbol values to their final definitions.
+ */
+static void
+indirectsym(void)
+{
+	const char *cp;
+	int changed, sym, ind;
+
+	do {
+		changed = 0;
+		for (sym = 0; sym < nsyms; ++sym) {
+			if (value[sym] == NULL)
+				continue;
+			cp = value[sym];
+			ind = findsym(&cp);
+			if (ind == -1 || ind == sym ||
+			    *cp != '\0' ||
+			    value[ind] == NULL ||
+			    value[ind] == value[sym])
+				continue;
+			debugsym("indir...", sym);
+			value[sym] = value[ind];
+			debugsym("...ectsym", sym);
+			changed++;
+		}
+	} while (changed);
+}
+
+/*
+ * Add a symbol to the symbol table, specified with the format sym=val
+ */
+static void
+addsym1(bool ignorethis, bool definethis, char *symval)
+{
+	const char *sym, *val;
+
+	sym = symval;
+	val = skipsym(sym);
+	if (definethis && *val == '=') {
+		symval[val - sym] = '\0';
+		val = val + 1;
+	} else if (*val == '\0') {
+		val = definethis ? "1" : NULL;
+	} else {
+		usage();
+	}
+	addsym2(ignorethis, sym, val);
+}
+
+/*
  * Add a symbol to the symbol table.
  */
 static void
-addsym(bool ignorethis, bool definethis, char *sym)
+addsym2(bool ignorethis, const char *sym, const char *val)
 {
+	const char *cp = sym;
 	int symind;
-	char *val;
 
-	symind = findsym(sym);
+	symind = findsym(&cp);
 	if (symind < 0) {
 		if (nsyms >= MAXSYMS)
 			errx(2, "too many symbols");
 		symind = nsyms++;
 	}
-	symname[symind] = sym;
 	ignore[symind] = ignorethis;
-	val = sym + (skipsym(sym) - sym);
-	if (definethis) {
-		if (*val == '=') {
-			value[symind] = val+1;
-			*val = '\0';
-		} else if (*val == '\0')
-			value[symind] = "";
-		else
-			usage();
+	symname[symind] = sym;
+	value[symind] = val;
+	debugsym("addsym", symind);
+}
+
+static void
+debugsym(const char *why, int symind)
+{
+	debug("%s %s%c%s", why, symname[symind],
+	    value[symind] ? '=' : ' ',
+	    value[symind] ? value[symind] : "undef");
+}
+
+/*
+ * Add symbols to the symbol table from a file containing
+ * #define and #undef preprocessor directives.
+ */
+static void
+defundefile(const char *fn)
+{
+	filename = fn;
+	input = fopen(fn, "rb");
+	if (input == NULL)
+		err(2, "can't open %s", fn);
+	linenum = 0;
+	while (defundef())
+		;
+	if (ferror(input))
+		err(2, "can't read %s", filename);
+	else
+		fclose(input);
+	if (incomment)
+		error("EOF in comment");
+}
+
+/*
+ * Read and process one #define or #undef directive
+ */
+static bool
+defundef(void)
+{
+	const char *cp, *kw, *sym, *val, *end;
+	Comment_state wascomment;
+
+	wascomment = incomment;
+	cp = skiphash();
+	if (cp == NULL)
+		return (false);
+	if (*cp == '\0')
+		goto done;
+	/* strip trailing whitespace, and do a fairly rough check to
+	   avoid unsupported multi-line preprocessor directives */
+	end = cp + strlen(cp);
+	while (end > tline && strchr(" \t\n\r", end[-1]) != NULL)
+		--end;
+	if (end > tline && end[-1] == '\\')
+		Eioccc();
+
+	kw = cp;
+	if ((cp = matchsym("define", kw)) != NULL) {
+		sym = getsym(&cp);
+		if (sym == NULL)
+			error("missing macro name in #define");
+		if (*cp == '(') {
+			val = "1";
+		} else {
+			cp = skipcomment(cp);
+			val = (cp < end) ? xstrdup(cp, end) : "";
+		}
+		debug("#define");
+		addsym2(false, sym, val);
+	} else if ((cp = matchsym("undef", kw)) != NULL) {
+		sym = getsym(&cp);
+		if (sym == NULL)
+			error("missing macro name in #undef");
+		cp = skipcomment(cp);
+		debug("#undef");
+		addsym2(false, sym, NULL);
 	} else {
-		if (*val != '\0')
-			usage();
-		value[symind] = NULL;
+		error("unrecognized preprocessor directive");
 	}
+	skipline(cp);
+done:
+	debug("parser line %d state %s comment %s line", linenum,
+	    comment_name[incomment], linestate_name[linestate]);
+	return (true);
 }
 
 /*
- * Compare s with n characters of t.
- * The same as strncmp() except that it checks that s[n] == '\0'.
+ * Concatenate two strings into new memory, checking for failure.
  */
-static int
-strlcmp(const char *s, const char *t, size_t n)
+static char *
+astrcat(const char *s1, const char *s2)
 {
-	while (n-- && *t != '\0')
-		if (*s != *t)
-			return ((unsigned char)*s - (unsigned char)*t);
-		else
-			++s, ++t;
-	return ((unsigned char)*s);
+	char *s;
+	int len;
+	size_t size;
+
+	len = snprintf(NULL, 0, "%s%s", s1, s2);
+	if (len < 0)
+		err(2, "snprintf");
+	size = (size_t)len + 1;
+	s = (char *)malloc(size);
+	if (s == NULL)
+		err(2, "malloc");
+	snprintf(s, size, "%s%s", s1, s2);
+	return (s);
+}
+
+/*
+ * Duplicate a segment of a string, checking for failure.
+ */
+static const char *
+xstrdup(const char *start, const char *end)
+{
+	size_t n;
+	char *s;
+
+	if (end < start) abort(); /* bug */
+	n = (size_t)(end - start) + 1;
+	s = malloc(n);
+	if (s == NULL)
+		err(2, "malloc");
+	snprintf(s, n, "%s", start);
+	return (s);
 }
 
 /*
@@ -1070,5 +1566,6 @@ error(const char *msg)
 	else
 		warnx("%s: %d: %s (#if line %d depth %d)",
 		    filename, linenum, msg, stifline[depth], depth);
+	closeio();
 	errx(2, "output may be truncated");
 }
diff --git a/extra/scripts/unifdef.h b/extra/scripts/unifdef.h
new file mode 100644
index 000000000..d57a8660b
--- /dev/null
+++ b/extra/scripts/unifdef.h
@@ -0,0 +1,56 @@
+/*
+ * Copyright (c) 2012 - 2013 Tony Finch <dot@dotat.at>
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <sys/stat.h>
+
+#include <ctype.h>
+#include <errno.h>
+#include <stdarg.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+/* Avoid err.h since this are non-standard BSD extensions */
+#define vwarnx(fmt, args)   ({ fprintf(stderr, "unifdef: "); vfprintf(stderr, fmt, args); fprintf(stderr, "\n"); })
+#define warnx(fmt, args...) fprintf(stderr, "unifdef: " fmt "\n", ## args)
+#define errx(exit_code, fmt, args...) ({ warnx(fmt, ## args); exit(exit_code); })
+#define err(exit_code, fmt, args...)  errx(exit_code, fmt ": %s", ## args, strerror(errno))
+
+/* portability stubs */
+
+#define fbinmode(fp) (fp)
+
+#define replace(old,new) rename(old,new)
+
+static FILE *
+mktempmode(char *tmp, int mode)
+{
+	int fd = mkstemp(tmp);
+	if (fd < 0) return (NULL);
+	fchmod(fd, mode & (S_IRWXU|S_IRWXG|S_IRWXO));
+	return (fdopen(fd, "wb"));
+}
diff --git a/extra/scripts/version.h b/extra/scripts/version.h
new file mode 100644
index 000000000..ed3e7c0fd
--- /dev/null
+++ b/extra/scripts/version.h
@@ -0,0 +1,2 @@
+"@(#) $Version: unifdef-2.9.5.55501a6 $\n"
+"@(#) $Date: 2013-06-12 15:50:39 +0100 $\n"