bug-gnulib
[Top][All Lists]
Advanced

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

Re: [PATCH] maint.mk: don't maintain a second build-aux variable.


From: Jim Meyering
Subject: Re: [PATCH] maint.mk: don't maintain a second build-aux variable.
Date: Fri, 28 Oct 2011 09:50:31 +0200

Gary V. Vaughan wrote:
> Hi Jim,
...
>>> -# Override this in cfg.mk if you use a non-standard build-aux directory.
>>> -build_aux ?= $(srcdir)/build-aux
>>> +ifneq ($(build_aux),)
>>> + $(error '*** set $$(_build-aux) relative to $$(srcdir) instead of
>>> $$(build_aux)')
>>
>> That line is longer than 80.
>> Please shorten or split it, and
>
> Done.
>
>> use a prefix of "$(ME): " as
>> in most of the other diagnostics in this file.  Or at least
>> include $(ME) somewhere, so people know where it's coming from.
>
> $ fgrep '$(error ' ~/Devo/gnulib/top/*
> top/maint.mk: $(error '*** set $$(_build-aux) relative to $$(srcdir) instead 
> of $$(build_aux)')
> top/maint.mk:  $(error '*** you need to 
> s/_prohibit_regexp/_sc_search_regexp/, and adapt')
>
> I changed the one error I added as requested, but I guess the other use which
> I copied from needs to have $(ME) added too (and be wrapped to avoid breaking
> the 80 column limit)?

You're welcome to change it if you'd like, but like your addition,
that $(error ...) provoker is solely to alert people to a problem
arising from a variable name change, and it's due to be removed in
a few months:

  # _sc_search_regexp used to be named _prohibit_regexp.  However,
  # upgrading to the new definition and leaving the old name undefined
  # would usually convert each custom rule using $(_prohibit_regexp)
  # (usually defined in cfg.mk) into a no-op.  This definition ensures
  # that people know right away if they're still using the old name.
  # FIXME: remove in 2012.
  _prohibit_regexp = \
    $(error '*** you need to s/_prohibit_regexp/_sc_search_regexp/, and adapt')

Please add a similar FIXME comment (but with say "2013"), so we
do eventually remove it.

> Okay to push?

Yes.  Thanks.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]