Re: texi2dvi: A more pleasant way to compile

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: texi2dvi: A more pleasant way to compile

Alexandre Duret-Lutz
>>> "Karl" == Karl Berry <[hidden email]> writes:

 >> So the new work is merely to add *.t2d in .cvsignore (or whatever), and
 >> CLEANFILES = *.t2d in Makefile.am, until Automake is updated?

Can we work on an interface that will not require any change in
Automake the next time texi2dvi invents new files or
directories?

Having an operation mode where texi2dvi cleans itself the
auxiliary files that it would otherwise create sounds right to
me.  In other words, I'd like `make mostlyclean' to call
`texi2dvi --mostlyclean input.texi' for all `input.texi',
instead of having to know how texi2dvi works internally.

(I think that's an old idea from Akim.)
--
Alexandre Duret-Lutz



Reply | Threaded
Open this post in threaded view
|

Re: texi2dvi: A more pleasant way to compile

Akim Demaille
>>> "adl" == Alexandre Duret-Lutz <[hidden email]> writes:

>>> "Karl" == Karl Berry <[hidden email]> writes:
 >>> So the new work is merely to add *.t2d in .cvsignore (or whatever), and
 >>> CLEANFILES = *.t2d in Makefile.am, until Automake is updated?

 > Can we work on an interface that will not require any change in
 > Automake the next time texi2dvi invents new files or
 > directories?

That's the whole point: now there is no reason for additional
files/directories to be created.

 > Having an operation mode where texi2dvi cleans itself the
 > auxiliary files that it would otherwise create sounds right to
 > me.  In other words, I'd like `make mostlyclean' to call
 > `texi2dvi --mostlyclean input.texi' for all `input.texi',
 > instead of having to know how texi2dvi works internally.

 > (I think that's an old idea from Akim.)

That's right :) But that's already too late for older texi2dvi, so I
think you will most probably never use this feature in Automake (just
as you will most probably not get rid of the silly list of hard coded
extensions that Texinfo files might create)

Sure, you can affort to

      -texi2dvi --mostlyclean
      rm -rf ...

but is there really any point?  I'm fine with implementing this, I'm
merely asking whether it will be used.



Reply | Threaded
Open this post in threaded view
|

Re: texi2dvi: A more pleasant way to compile

Alexandre Duret-Lutz-2
>>> "Akim" == Akim Demaille <[hidden email]> writes:

 Akim> That's the whole point: now there is no reason for additional
 Akim> files/directories to be created.

How would we know?  Maybe someone will ask for that directory to
be named differently.  Maybe someone will discover that some
other files are needed in the current directory.  Maybe we
simply can't imagine what will needed.

 Akim> is there really any point?

It's simple to implement and it's cleaner to use.

It also shows a good example of an interface we would both like
to see in many other tools.

 Akim> you will most probably not get rid of the silly list of
 Akim> hard coded extensions

I would like to.  Not now, of course.
--
Alexandre Duret-Lutz



Reply | Threaded
Open this post in threaded view
|

Re: texi2dvi: A more pleasant way to compile

Akim Demaille
>>> "adl" == Alexandre Duret-Lutz <[hidden email]> writes:

 Akim> is there really any point?

 > It's simple to implement and it's cleaner to use.  It also shows a
 > good example of an interface we would both like to see in many
 > other tools.

Of course you're right.  Thanks for keeping me away from the dark side.