Add tablist support and a mandoc-manpage to expand(1)

and mark it as finished in the README.

This is another example showing how broken the GNU coreutils are:

$ echo -e "äää\tüüü\tööö" | gnu-expand -t "5,10,20"
äää    üüü    ööö
$ echo -e "äää\tüüü\tööö" | sbase-expand -t "5,10,20"
äää  üüü  ööö

This is due to the fact that they are still not UTF8-aware and
actually see "ä" as two single characters, expanding the "äää" with
4 spaces to a tab of length 10.
The correct way however is to expand the "äää" with 2 spaces to a
tab of length 5.
One can only imagine how this silently breaks a lot of code around
the world.
WHAT WERE THEY THINKING?

README

@@ -28,7 +28,7 @@ The following tools are implemented ('*' == finished, '#' == UTF-8 support,
 =  du              no                              -H, -L, (-x)
 =* echo            yes                             none
 =* env             yes                             none
-#  expand          yes                             none
+#* expand          yes                             none
    expr            yes                             none
 =* false           yes                             none
    fold            yes                             none

expand.1

@@ -1,25 +1,50 @@
-.TH EXPAND 1 sbase\-VERSION
+.Dd January 25, 2015
-.SH NAME
+.Dt EXPAND 1 sbase\-VERSION
-expand \- expand tabs to spaces
+.Sh NAME
-.SH SYNOPSIS
+.Nm expand
-.B expand
+.Nd expand tabs to spaces
-.RB [ \-t
+.Sh SYNOPSIS
-.IR n ]
+.Nm expand
-.RI [ file ...]
+.Op Fl i
-.SH DESCRIPTION
+.Op Fl t Ar tablist
-expand processes the named files or the standard input, writing the
+.Op Ar file ...
-standard output with tabs changed into spaces.  Backspace characters
+.Sh DESCRIPTION
-are preserved into the output and decrement the column count for tab
+.Nm
-calculations.
+converts tabs to spaces in each
-.SH OPTIONS
+.Ar file
-.TP
+as specified in
-.BI \-i
+.Ar tablist .
-Only change tabs to spaces at the start of lines.
+If no file is given,
-.TP
+.Nm
-.BI \-t " n"
+reads from stdin.
-Expand tabs to
+.Pp
-.I n
+Backspace characters are preserved and decrement the column count
-spaces.  We currently support only a single numerical argument.
+for tab calculations.
-.SH SEE ALSO
+.Sh OPTIONS
-.IR unexpand (1),
+.Bl -tag -width Ds
-.IR fold (1)
+.It Fl i
+Only expand tabs at the beginning of lines, i.e. expand each
+line until a character different from '\et' and ' ' is reached.
+.It Fl t Ar tablist
+Specify tab size or tabstops.
+.Ar tablist
+is a list of one (in the former case) or multiple (in the latter case)
+strictly positive integers separated by ' ' or ','.
+.Pp
+The default
+.Ar tablist
+is "8".
+.El
+.Sh SEE ALSO
+.Xr unexpand 1 ,
+.Xr fold 1
+.Sh STANDARDS
+The
+.Nm
+utility is compliant with the
+.St -p1003.1-2008
+specification.
+.Pp
+The
+.Op Fl i
+flag is an extension to that specification

expand.c

@@ -1,89 +1,86 @@
 /* See LICENSE file for copyright and license details. */
 #include <stdio.h>
 #include <stdlib.h>
+#include <string.h>
 
 #include "utf.h"
 #include "util.h"
 
-static int expand(const char *, FILE *, int);
+static int     iflag      = 0;
+static size_t *tablist    = NULL;
+static size_t  tablistlen = 0;
 
-static int iflag = 0;
+static size_t
-
+parselist(const char *s, size_t slen)
-static void
-usage(void)
 {
-	eprintf("usage: %s [-i] [-t n] [file...]\n", argv0);
+	size_t i, m, len;
-}
+	char *sep;
 
-int
+	if (s[0] == ',' || s[0] == ' ')
-main(int argc, char *argv[])
+		eprintf("expand: tablist can't begin with a ',' or ' '.\n");
-{
+	if (s[slen - 1] == ',' || s[slen - 1] == ' ')
-	FILE *fp;
+		eprintf("expand: tablist can't end with a ',' or ' '.\n");
-	int tabstop = 8;
-	int ret = 0;
 
-	ARGBEGIN {
+	len = 1;
-	case 'i':
+	for (i = 0; i < slen; i++) {
-		iflag = 1;
+		if (s[i] == ',' || s[i] == ' ') {
-		break;
+			if (i > 0 && (s[i - 1] == ',' || s[i - 1] == ' '))
-	case 't':
+				eprintf("expand: empty field in tablist.\n");
-		tabstop = estrtol(EARGF(usage()), 0);
+			len++;
-		if (!tabstop)
-			eprintf("tab size cannot be zero\n");
-		break;
-	default:
-		usage();
-	} ARGEND;
-
-	if (argc == 0) {
-		expand("<stdin>", stdin, tabstop);
-	} else {
-		for (; argc > 0; argc--, argv++) {
-			if (!(fp = fopen(argv[0], "r"))) {
-				weprintf("fopen %s:", argv[0]);
-				ret = 1;
-				continue;
-			}
-			expand(argv[0], fp, tabstop);
-			fclose(fp);
 		}
 	}
-	return ret;
+	tablist = emalloc((len + 1) * sizeof(size_t));
+
+	m = 0;
+	for (i = 0; i < slen; i += sep - (s + i) + 1) {
+		tablist[m++] = strtol(s + i, &sep, 0);
+		if (tablist[m - 1] == 0)
+			eprintf("expand: tab size can't be zero.\n");
+		if (*sep && *sep != ',' && *sep != ' ')
+			eprintf("expand: invalid number in tablist.\n");
+		if (m > 1 && tablist[m - 1] < tablist[m - 2])
+			eprintf("expand: tablist must be ascending.\n");
+	}
+
+	/* tab length = 1 for the overflowing case later in the matcher */
+	tablist[len] = 1;
+	return len;
 }
 
 static int
-expand(const char *file, FILE *fp, int tabstop)
+expand(const char *file, FILE *fp)
 {
-	int col = 0;
+	size_t bol = 1, col = 0, i;
 	Rune r;
-	int bol = 1;
-
-	for (;;) {
-		if (!readrune(file, fp, &r))
-			break;
 
+	while (readrune(file, fp, &r)) {
 		switch (r) {
 		case '\t':
+			if (tablistlen == 1)
+				i = 0;
+			else for (i = 0; i < tablistlen; i++)
+				if (col < tablist[i])
+					break;
 			if (bol || !iflag) {
 				do {
 					col++;
 					putchar(' ');
-				} while (col % tabstop);
+				} while (col % tablist[i]);
 			} else {
 				putchar('\t');
-				col += tabstop - col % tabstop;
+				col = tablist[i];
 			}
 			break;
 		case '\b':
+			bol = 0;
 			if (col)
 				col--;
-			bol = 0;
+			putchar('\b');
-			writerune("<stdout>", stdout, &r);
 			break;
 		case '\n':
-			col = 0;
 			bol = 1;
-			writerune("<stdout>", stdout, &r);
+			col = 0;
+			putchar('\n');
 			break;
 		default:
 			col++;
@@ -96,3 +93,47 @@ expand(const char *file, FILE *fp, int tabstop)
 
 	return 0;
 }
+
+static void
+usage(void)
+{
+	eprintf("usage: %s [-i] [-t tablist] [file ...]\n", argv0);
+}
+
+int
+main(int argc, char *argv[])
+{
+	FILE *fp;
+	char *tl = "8";
+	int   ret = 0;
+
+	ARGBEGIN {
+	case 'i':
+		iflag = 1;
+		break;
+	case 't':
+		tl = EARGF(usage());
+		if (!*tl)
+			eprintf("expand: tablist cannot be empty.\n");
+		break;
+	default:
+		usage();
+	} ARGEND;
+
+	tablistlen = parselist(tl, strlen(tl));
+
+	if (argc == 0)
+		expand("<stdin>", stdin);
+	else {
+		for (; argc > 0; argc--, argv++) {
+			if (!(fp = fopen(argv[0], "r"))) {
+				weprintf("fopen %s:", argv[0]);
+				ret = 1;
+				continue;
+			}
+			expand(argv[0], fp);
+			fclose(fp);
+		}
+	}
+	return ret;
+}