Re: appears CVS build is broken (tty-supports-face-attributes-p)?

[Luc Teirlinck <teirllm@dms.auburn.edu>]

Miles Bader wrote:

   BTW, usually it's sufficient to do `make -C lisp recompile', and of
   course it's much faster.  So a generally good strategy after updating is
   to do that followed by a normal make, and if it fails in a suspicious
   way, then try make maintainer-clean + make bootstrap.

In practice, people are going to do what we _ask_ then to do in
INSTALL.CVS:

    Normally, it is not necessary to use "make bootstrap" after every CVS
    update.  Unless there are problems, we suggest the following
    procedure:

      $ ./configure
      $ make
      $ cd lisp
      $ make recompile EMACS=../src/emacs
      $ cd ..
      $ make

    (If you want to install the Emacs binary, type "make install" instead
    of "make" in the last command.)

If this is not the best strategy, INSTALL.CVS should be changed.

(Note however that for people with fast machines, running `make
maintainer-clean' and `make bootstrap' is so fast that it is not even
worth the trouble trying anything else.)

In case of trouble we recommend first to update loaddefs.el and if
this does not solve the problem, to run `make bootstrap'.  That used
to be sufficient.  After a relatively recent change it no longer is.
One has to first do `make maintainer-clean', then `make bootstrap'.
This is not pointed out in INSTALL.CVS.

There are two solutions.  The first is to point it out in INSTALL.CVS.
One problem with that is that people already familiar with
INSTALL.CVS, do not reread it every few days.

A more radical solution is to revert the recent change to `make
bootstrap'.  That change made `make bootstrap' run faster, but at a
cost.  We get reports on trouble during bootstrapping that are only
due to failure to run `make maintainer-clean' on a regular basis.
Failing to run `make maintainer-clean' does not necessarily cause
bootstrapping to fail.  Instead, it can cause bugs in the bootstrapped
Emacs.  So every time somebody reports a bug that seems
irreproducible, there is always the possibility that it was due
failure to run `make maintainer-clean'.

Has anybody benefited sufficiently from the change in `make bootstrap'
to compensate for all the time that many people are losing as a result
of it?  I am inclined to believe that the change to `make bootstrap'
should be reverted.

Sincerely,

Luc.