[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: fmttest(1) Suffers Segmentation Violation.
From: |
Ralph Corderoy |
Subject: |
Re: fmttest(1) Suffers Segmentation Violation. |
Date: |
Tue, 04 May 2021 09:16:24 +0100 |
Hi Ken,
> > This dumps core for me.
> >
> > fmttest --x y -forma ''
> >
> > I've had a look at the code and I suspect it's an easy fix related
> > to spotting no messages have been provided before attempting to loop
> > from 0 to 0 inclusive and indexing 0-1 but thought it better to
> > leave the fix to someone more confident in what's meant to happen.
> > :-)
>
> Oh, strangely enough ... it's because it's trying to call
> seq_setprev() with no messages set.
Not for me as I don't have Previous-Sequence set. git-grep(1) suggests
our test suite doesn't set it either?
> So, I guess "-message" is the default.
Oh, it definitely is. :-)
> And that suggests to me if you don't give a message on the command
> line, it should default to 'cur' like other nmh programs, unless
> people think it makes sense to fail in that case.
fmttest doesn't alter the message so I think ‘cur’ is the correct
default value of ‘msgs’.
> (And I realize from reading the man page I never documented what
> -file/-nofile means).
fmttest(1) here says
‘If the -file switch is given, the arguments are interpreted as
filenames instead of message numbers, but otherwise the behavior is
the same (except that the %(msg), %(cur), and %(unseen) function
escapes will not provide any useful information).’
--
Cheers, Ralph.