bug#12903: "cut -b 0-" is supposed to be invalid

[Bernhard Voelker]

On 11/16/2012 11:23 PM, Jim Meyering wrote:
> Bernhard Voelker wrote:
> 
>> Hi Marcel,
>>
>> On 11/16/2012 07:35 AM, Marcel Böhme wrote:
>>>    Hi,
>>>    The command "echo 12345 | cut -b 0-" prints an empty line while it
>>>    should fail with "fields and positions are numbered from 1" according
>>>    to this specification "cut: diagnose a range starting with 0 (-f 0-2)
>>>    as invalid".
>>>    Can you confirm this is an incomplete bugfix made on 2007-05-22?
>>
>> I'd say yes.
>> Here's a proposed fix.
> ...

> Thank you!
> That change looks fine, at first glance,
> but please amend the log to mention the bug number URL.
> Also, technically this is a bug, so please mention the
> fix in NEWS.

Thanks for the review.

I amended with updated NEWS and also improved the test to include
both the field option -f and the position options -b and -c,
plus mentioned the bug URL in the commit message.

Have a nice day,
Berny


>From 10fc244ac8e4f24eb330eae110fadc0b7d78cf8d Mon Sep 17 00:00:00 2001
From: Bernhard Voelker <address@hidden>
Date: Sun, 18 Nov 2012 22:20:16 +0100
Subject: [PATCH] cut: do not accept the invalid range 0-
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The command "echo 12345 | cut -b 0-" prints an empty line while
it should fail with "fields and positions are numbered from 1".

* src/cut.c (set_fields): Add a diagnostic for the invalid open
range which starts with Zero, i.e., the range 0-.
* tests/misc/cut.pl: Add tests to ensure the range 0- fails for
fields (-f) and for positions (-b, -c).
* NEWS: Mention the fix.

Reported by Marcel Böhme in <http://bugs.gnu.org/12903>.
---
 NEWS              |    4 ++++
 src/cut.c         |    3 +++
 tests/misc/cut.pl |    5 +++++
 3 files changed, 12 insertions(+), 0 deletions(-)

diff --git a/NEWS b/NEWS
index db6a207..e4a284c 100644
--- a/NEWS
+++ b/NEWS
@@ -10,6 +10,10 @@ GNU coreutils NEWS                                    -*- 
outline -*-

 ** Bug fixes

+  cut no longer accepts the invalid range 0-, which made it print empty lines.
+  Instead, cut now fails and emits an appropriate diagnostic.
+  [This bug was present in "the beginning".]
+
   pr -n no longer crashes when passed values >= 32.  Also line numbers are
   consistently padded with spaces, rather than with zeros for certain widths.
   [bug introduced in TEXTUTILS-1_22i]
diff --git a/src/cut.c b/src/cut.c
index 87380ac..2a57148 100644
--- a/src/cut.c
+++ b/src/cut.c
@@ -369,6 +369,9 @@ set_fields (const char *fieldstr)
           dash_found = true;
           fieldstr++;

+          if (lhs_specified && !value)
+            FATAL_ERROR (_("fields and positions are numbered from 1"));
+
           initial = (lhs_specified ? value : 1);
           value = 0;
         }
diff --git a/tests/misc/cut.pl b/tests/misc/cut.pl
index 0ce051a..cd56555 100755
--- a/tests/misc/cut.pl
+++ b/tests/misc/cut.pl
@@ -46,6 +46,11 @@ my @Tests =
   # It was treated just like "-2".
   ['zero-2', '-f0-2', {ERR=>$from_1}, {EXIT => 1} ],

+  # Up to coreutils-8.20, specifying a range of 0- was not an error.
+  ['zero-3b', '-b0-', {ERR=>$from_1}, {EXIT => 1} ],
+  ['zero-3c', '-c0-', {ERR=>$from_1}, {EXIT => 1} ],
+  ['zero-3f', '-f0-', {ERR=>$from_1}, {EXIT => 1} ],
+
   ['1', '-d:', '-f1,3-', {IN=>"a:b:c\n"}, {OUT=>"a:c\n"}],
   ['2', '-d:', '-f1,3-', {IN=>"a:b:c\n"}, {OUT=>"a:c\n"}],
   ['3', qw(-d: -f2-), {IN=>"a:b:c\n"}, {OUT=>"b:c\n"}],
-- 
1.7.7