[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#2624: textutils needs to break fmt-related files out separately



david@elo.ods.com (David Engel)

> > I don't know about its current status, but the textutils fmt program
> > used to be broken in various ways where the elvis fmt program wasn't
> > broken.  I don't know if the textutils fmt program has been fixed
> > or not.
>
> If the textutils fmt hasn't been fixed yet, it should be.  If it can't
> easily be fixed, then it should be removed completely.

Last night, I downloaded the current textutils package, and the
current dpkg package which is needed in order to install it.  A
few quick tests didn't turn up any obvious bugs.  It looks like
the earlier broken version of textutils fmt has been fix.  I'm now
leaning towards either withdrawing the elv-fmt package from the
distribution, and thereby making our lives a bit less complicated.
I'm pushed for time just now, but will probably post a request tonight
that this be done.

[...]

> > > simply use dpkg-divert to avoid conflicting with the one in textutils?
> >
> > dpkg-divert is a new one on me.  What's that?
> > Do you perhaps mean update-alternatives?
>
> No, I don't mean update-alternatives, which I personally don't care
> for but I won't get into that.  Dpkg diversions are a feature Ian J.
> added a while back where one package can selectively override files
> from other packages by diverting them to other names.  Dpkg remembers
> the diversions and automatically does the right thing when either
> package is upgraded or removed.  Please read
> /usr/doc/dpkg/diversions.text for details.

I've now looked quickly at the /usr/doc/dpkg docs in my newly-installed
dpkg-1.1.3.  I somehow missed the discussion on dpkg-divert, and perhaps
some of the other material as well.  After a quick reading of the
diversions.txt file I think I'll probably still withdraw elv-fmt in
favor of textutils fmt, but dpkg-divert may be useful in some of the
other messy inter-package situations I have.  Thanks for the pointer
to this -- I'll read it more thoroughly tonight.



Reply to: