Debian Bug report logs - #268744
nmh: pick(1) Has Trouble with Multi-Line Subjects.

Reply or subscribe to this bug.

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox):

From: Ralph Corderoy <ralph@inputplus.co.uk>
To: Debian Bug Tracking System <submit@bugs.debian.org>
Subject: nmh: pick(1) Has Trouble with Multi-Line Subjects.
Date: Sat, 28 Aug 2004 22:43:06 +0100

Package: nmh
Version: 1.0.4+dev-20010317-1
Severity: normal
File: /usr/bin/mh/pick

pick(1) says

    Pattern matching is performed on a per-line basis.  Within the
    header of the message, each component is treated as one long line,
    but in the body, each line is separate. Lower-case letters in the
    search pattern will match either lower or upper case in the message,
    while upper case will match only upper case.

yet I can't match the split line in a long subject.

    $ scan -format '%{subject}\n'  
    foo bar
    $ scan -format '%{subject}\n' | od -c
    0000000   f   o   o       b   a   r  \n
    0000010
    $ grep -1 '^Subject' `mhpath 1`
    To: ralph@inputplus.co.uk
    Subject: foo
     bar
    $ pick -sub foo  
    1
    $ pick -sub bar
    1
    $ pick -sub 'foo bar'
    pick: no messages match specification
    $ pick -sub 'foo.bar'
    pick: no messages match specification
    $ pick -sub 'foo^Mbar'
    pick: no messages match specification
    $ pick -sub 'foo
    bar'
    pick: no messages match specification
    $ pick -sub 'foo
     bar'
    pick: no messages match specification

I can find `foo' or `bar' so the subject is being treated as one long
line, as the man page says, but what are the lines being joined with?
The search clearly works across both lines since these two both work.

    $ pick -sub 'f.*o'
    1
    $ pick -sub 'f.*r'
    1

I'd expect it to be a space unless the man page says otherwise.

-- System Information
Debian Release: 3.0
Architecture: i386
Kernel: Linux dayna 2.4.18-dayna.2.6 #1 Sun Jul 4 11:06:01 BST 2004 i686
Locale: LANG=C, LC_CTYPE=

Versions of packages nmh depends on:
ii  base-files                  3.0.2        Debian base system miscellaneous f
ii  libc6                       2.3.2.ds1-13 GNU C Library: Shared libraries an
ii  liblockfile1                1.03         NFS-safe locking library, includes
ii  libncurses5                 5.4-4        Shared libraries for terminal hand
ii  libsasl7                    1.5.27-3     Authentication abstraction library
ii  netbase                     4.07         Basic TCP/IP networking system

Message #10 received at 268744@bugs.debian.org (full text, mbox):

From: Harald Geyer <Harald.Geyer@gmx.at>
To: 268744@bugs.debian.org, ralph@inputplus.co.uk, control@bugs.debian.org
Subject: nmh: pick: Handling of Multi-Line Subjects
Date: Thu, 23 Sep 2004 12:06:59 +0200

severity 268744 minor
stop

Hi!

I tried pick on the following file:

---
To: ralph@inputplus.co.uk
Subject: foo
 bar

Test.
---

$ pick -sub 'foo..bar'
1
$ pick -sub 'foo  bar'
1

Then adding a second space in front of "bar":

$ pick -sub 'foo..bar'
pick: no messages match specification
$ pick -sub 'foo  bar'
pick: no messages match specification
$ pick -sub 'foo...bar'
1
$ pick -sub 'foo   bar'
1

Obviously lines are joined by one space, but all other whitespace
is preserved. This is bad, because the indentation can be any number
of spaces and/or tabs IIRC.

However almost always exactly one character is used. Also it is easy to
write some general regexp, if needed. I set the severity of this bug
to minor, as it is really easy to workaround.

Harald

Send a report that this bug log contains spam.