[bug #27275] Wrong behaviour with GREP_USE_DFA=0

[anonymous]

URL:
  <http://savannah.gnu.org/bugs/?27275>

                 Summary: Wrong behaviour with GREP_USE_DFA=0
                 Project: grep
            Submitted by: None
            Submitted on: Tue 18 Aug 2009 09:03:34 AM UTC
                Category: None
                Severity: 3 - Normal
              Item Group: None
                  Status: None
                 Privacy: Public
             Assigned to: None
             Open/Closed: Open
         Discussion Lock: Any

    _______________________________________________________

Details:

>From Debian bug #538737. I confirmed it with version 2.5.4

http://bugs.debian.org/538737 :

"Setting the (un-documented) environment variable GREP_USE_DFA to 0 causes
grep to generate incorrect results.

Start with a three line text file:
$ cat file.txt
 foo <= space
        foo <= tab
sfoo

Now use grep with GREP_USE_DFA set to 1.  Notice correct match:
$ GREP_USE_DFA=1 grep '^\s*foo' file.txt
sfoo

Now try with GREP_USE_DFA set to 0.
$ GREP_USE_DFA=0 grep '^\s*foo' file.txt
 foo <= space
        foo <= tab

The program now matches the other two lines, but not the one that
should match.

This is a serious issue since disabling DFA is the default for
multi-byte encodings such as UTF-8.

You can get the same (wrong) results by setting LANG:

$ LANG=C grep '^\s*foo' file.txt
sfoo
$ LANG=en_US.UTF-8 grep '^\s*foo' file.txt
 foo <= space
        foo <= tab
"





    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?27275>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/