Re: m4_defn overhead

[Eric Blake]

Ralf Wildenhues <Ralf.Wildenhues <at> gmx.de> writes:

> > m4_define([_LT_DECL],
> > [lt_if_append_uniq([lt_decl_varnames], [$2], [[, ]],...
> > 
> > where lt_if_append_uniq defers to m4_append.  Notice that the separator is 
the 
> > overquoted string `[, ]';
> 
> Hmm.  The Quotation Rule of Thumb in the manual recommends double
> quoting of risky literals, which certainly includes comma.  Maybe
> it would help to note that SEPARATOR shouldn't be double-quoted?

Probably, along with a note that STRING also need not be double-quoted.

A lot of the m4 text macros (m4_split, m4_normalize, m4_flatten, m4_strip) 
intentionally operate on quoted text and return quoted output; the m4_* rules 
are thus slightly different than the AC_* rules.

> 
> For a counter example, the first argument to AS_HELP_STRING needs to be
> doubly quoted if it contains a comma (the node `Autoconf Language' in
> the manual hints the alert reader that this may be needed, but it's a
> rather obscure hint).

Not for long.  I still have one of my pending patches [1] that allows single-
quoted commas in AS_HELP_STRING without eating the trailing whitespace, via my 
new macro m4_expand (or whatever better name we can come up with).

[1] http://lists.gnu.org/archive/html/autoconf-patches/2007-10/msg00071.html

[Note to self - I'm now up to 10 open topic branches on my local git 
repository, from all the patches I've been working in parallel; I need to start 
flushing them upstream...]


> 
> > which prior to my m4sugar patch was being expanded by 
> > m4_append to strip a layer of quoting, but now it is being used as-is.  
> > Furthermore, this line in libtool.m4 ONLY appears on libtool HEAD, which 
has 
> > not been released yet, so I hope very few others are being bitten by this 
bug.
> 
> Well, Libtool HEAD is used quite a bit out there, but I agree that we
> should fix things.

Yes - M4 HEAD for starters (don't you love the bootstrapping effect this change 
has?)   :)

Still, anyone brave enough to use unreleased software must be aware of the 
perils.


> 
> > So, I would like to patch autoconf to add a testsuite to m4sugar.at, 
testing 
> > the NEW semantics; add a NEWS item that describes the change in semantics, 
but 
> > not alter m4sugar.m4.  I will follow up with a patch to the libtool list 
that 
> > fixes the libtool usage, regardless of autoconf version it is run with.
> 
> I'm fine with the patch; you may want to add some tests that actually
> use comma in the SEPARATOR argument, to ensure Autoconf doesn't regress
> in its semantics there.

You got it.  I'll add those tests before committing the descendent of this 
patch, along with some verbiage in autoconf.texi.

> > +2007-10-11  Eric Blake  <ebb9 <at> byu.net>
> > +
> > +   Document effect of recent m4_append change on libtool HEAD.
> > +   * tests/m4sugar.at (m4@&t <at> _append): New test.
> > +   * NEWS: Document semantics change.

-- 
Eric Blake