Re: m4_quote [was: fix AT_SETUP's sh-escaping]

[Joel E. Denny]

On Wed, 25 Oct 2006, Eric Blake wrote:

> It exists for expanding relatively arbitrary parameters inside nested
> macros.  For example, in the upcoming M4 1.4.8 manual, I have the
> following example of a foreach macro (converted here to m4sugar syntax):
> 
> # foreachq(x, `item_1, item_2, ..., item_n', stmt)
> #   quoted list, simple version
> m4_define([foreachq], [m4_pushdef([$1])_foreachq($@)m4_popdef([$1])])
> m4_define([_arg1], [$1])
> m4_define([_foreachq], [m4_if(m4_quote($2), [], [],
>   [m4_define([$1], [_arg1($2)])$3[]$0([$1], [m4_shift($2)], [$3])])])
> 
> If you don't use m4_quote, then the unquoted expansion of $2 inside the
> first argument of m4_if will lead to syntax errors if $2 contains unquoted
> commas.

Thanks, that clears up my confusion.  Also, it really helps me to see why 
[$2] isn't equivalent to m4_quote($2).  The autoconf manual's example in 
section "8.3.3 Evaluation Macros" doesn't draw a distinction between the 
two:

  $ cat example.m4
  # Overquote, so that quotes are visible.
  m4_define([show], [$[]1 = [$1], $[]@ = address@hidden)
  m4_divert(0)dnl
  show(a, b)
  show(m4_quote(a, b))
  show(m4_dquote(a, b))
  $ autom4te -l m4sugar example.m4
  $1 = a, $@ = [a],[b]
  $1 = a,b, $@ = [a,b]
  $1 = [a],[b], $@ = [[a],[b]]

I just did this experiment, which for me is clearer than either above 
example:

  $ cat example.m4
  m4_divert(0)dnl
  m4_define([mkargs],[1,2,3])
  m4_define([echo_arg1],[[$1]])
  echo_arg1(mkargs)
  echo_arg1([mkargs])
  echo_arg1(m4_quote(mkargs))
  echo_arg1(m4_dquote(mkargs))
  $ autom4te -l m4sugar example.m4
  1
  mkargs
  1,2,3
  [1],[2],[3]

I wonder if this could be added to the autoconf manual.