From 651c5d428c2ef103ee8c5b1a310d6f29f0304744 Mon Sep 17 00:00:00 2001

From: Karel Zak <kzak@redhat.com>

Date: Tue, 27 Mar 2018 10:40:13 +0200

Subject: [PATCH] column: fix leading space characters bug

The bug has been introduced during column(1) rewrite. The function

read_input() need to skip leading space only temporary to detect empty

lines, but the rest of the code has to use the original buffer (line).

I've tried to fix one of the symptoms by 5c7b67fbbf41c973ca8d49b1e8bdba22dbb917aa

(alter), but this solution is unnecessary and too complex.

Changes:

* don't ignore leading space

* remove unnecessary stuff introduced by 5c7b67fbbf41c973ca8d49b1e8bdba22dbb917aa

* fix regression test with incorrect separator

Addresses: https://github.com/karelzak/util-linux/issues/575

Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=1560283

Signed-off-by: Karel Zak <kzak@redhat.com>

---

 tests/expected/column/table-input-separator-space |  2 +-

 tests/ts/column/table                             |  2 +-

 text-utils/column.c                               | 36 ++---------------------

 4 files changed, 5 insertions(+), 38 deletions(-)

diff --git a/tests/expected/column/table-input-separator-space b/tests/expected/column/table-input-separator-space

index 8a6513c11..25d9b5ab0 100644

--- a/tests/expected/column/table-input-separator-space

+++ b/tests/expected/column/table-input-separator-space

@@ -1,5 +1,5 @@

 AAA	BBBB	C	DDDD

-BBB	CCCC	DDD

+	BBB	CCCC	DDD

 AA	BB		DD

 AAAA	B	CC	D

 AA		CC	DD

diff --git a/tests/ts/column/table b/tests/ts/column/table

index 27b52e7c8..5c89d5eaf 100755

--- a/tests/ts/column/table

+++ b/tests/ts/column/table

@@ -37,7 +37,7 @@ $TS_CMD_COLUMN --separator ',' --table $TS_SELF/files/table-sep >> $TS_OUTPUT 2>

 ts_finalize_subtest

 ts_init_subtest "input-separator-space"

-$TS_CMD_COLUMN --separator ',' --table $TS_SELF/files/table-sep-space >> $TS_OUTPUT 2>&1

+$TS_CMD_COLUMN --separator "$(echo -e '\t')" --table $TS_SELF/files/table-sep-space >> $TS_OUTPUT 2>&1

 ts_finalize_subtest

 ts_init_subtest "long"

diff --git a/text-utils/column.c b/text-utils/column.c

index 89d46d280..195814328 100644

--- a/text-utils/column.c

+++ b/text-utils/column.c

@@ -86,7 +86,6 @@ struct column_control {

 	const char *tree_parent;

 	wchar_t *input_separator;

-	char *input_separator_raw;

 	const char *output_separator;

 	wchar_t	**ents;		/* input entries */

@@ -96,7 +95,6 @@ struct column_control {

 	unsigned int greedy :1,

 		     json :1,

 		     header_repeat :1,

-		     input_sep_space : 1,	/* input separator contains space chars */

 		     tab_noheadings :1;

 };

@@ -470,19 +468,7 @@ static int read_input(struct column_control *ctl, FILE *fp)

 	char *buf = NULL;

 	size_t bufsz = 0;

 	size_t maxents = 0;

-	int rc = 0, is_space_sep = 0;

-

-	/* Check if columns separator contains spaces chars */

-	if (ctl->mode == COLUMN_MODE_TABLE && ctl->input_separator_raw) {

-		char *p;

-

-		for (p = ctl->input_separator_raw; *p; p++) {

-			if (isspace(*p)) {

-				is_space_sep = 1;

-				break;

-			}

-		}

-	}

+	int rc = 0;

 	/* Read input */

 	do {

@@ -496,19 +482,6 @@ static int read_input(struct column_control *ctl, FILE *fp)

 			err(EXIT_FAILURE, _("read failed"));

 		}

 		str = (char *) skip_space(buf);

-

-		/* The table columns separator could be a space. In this case

-		 * don't skip the separator if at begin of the line. For example:

-		 *

-		 * echo -e "\tcol1\tcol2\nrow\t1\t2" \

-		 *	| column -t -s "$(echo -e '\t')" --table-columns A,B,C

-		 */

-		if (is_space_sep && str > buf) {

-			char *x = strpbrk(buf, ctl->input_separator_raw);

-			if (x && x < str)

-				str = x;

-		}

-

 		if (str) {

 			p = strchr(str, '\n');

 			if (p)

@@ -517,13 +490,13 @@ static int read_input(struct column_control *ctl, FILE *fp)

 		if (!str || !*str)

 			continue;

-		wcs = mbs_to_wcs(str);

+		wcs = mbs_to_wcs(buf);

 		if (!wcs) {

 			/*

 			 * Convert broken sequences to \x<hex> and continue.

 			 */

 			size_t tmpsz = 0;

-			char *tmp = mbs_invalid_encode(str, &tmpsz);

+			char *tmp = mbs_invalid_encode(buf, &tmpsz);

 			if (!tmp)

 				err(EXIT_FAILURE, _("read failed"));

@@ -720,7 +693,6 @@ int main(int argc, char **argv)

 	ctl.output_separator = "  ";

 	ctl.input_separator = mbs_to_wcs("\t ");

-	ctl.input_separator_raw = xstrdup("\t ");

 	while ((c = getopt_long(argc, argv, "c:dE:eH:hi:JN:n:O:o:p:R:r:s:T:tVW:x", longopts, NULL)) != -1) {

@@ -775,9 +747,7 @@ int main(int argc, char **argv)

 			break;

 		case 's':

 			free(ctl.input_separator);

-			free(ctl.input_separator_raw);

 			ctl.input_separator = mbs_to_wcs(optarg);

-			ctl.input_separator_raw = xstrdup(optarg);

 			ctl.greedy = 0;

 			break;

 		case 'T':

-- 

2.14.3