Re: config.sub/config.guess using nonportable $(...) substitutions

[Paul Eggert]

On 3/8/21 3:00 PM, Dmitry V. Levin wrote:
> The only rationale provided by the previous maintainer so far is a short
> message in config-patches mailing list [1].

The config maintainer Ben Elliston has wanted to get rid of the 
old-fashioned accent graves for many years. In November 2017 he 
installed a change to do that, but I explained it wouldn't work well on 
Solaris 10 so he reverted. Two years ago he asked me about this again, 
and I responded that Solaris 10's end-of-life was scheduled to be 
January 2021 and so it'd be best to wait until then, and he agreed. Last 
April we emailed each other again about this, and I told him that Oracle 
had extended Solaris 10's end-of-life to January 2024 but he demurred, 
writing that he didn't want to wait the extra three years.

Except maybe for Solaris 10, shells that don't grok $(...) are museum 
pieces now. And anybody on Solaris 10 (which occasionally includes me, 
as my department still uses Solaris 10 on some machines) can easily run 
a better shell like /bin/ksh. It's a bit of a maintenance hassle for Ben 
to deal with `...` (it doesn't nest, and it has weird rules when 
combined with "...") and it's understandable that he would rather deal 
with actual config.guess problems than with completely-obsolete shells 
that don't support standard syntax.

I recall having a similar discussion back in the 1970s, when a shell 
script stopped working for me because its author put in a comment 
starting with "#", something the 7th Edition shell did not support. I 
wrote the author, who suggested I get a better shell, and life went on. 
And I'd expect a similar reaction today if someone asked us to remove 
the "#" comments from config.guess on the grounds that they don't work 
with Steve Bourne's original V7 shell.

At some point, failing to support $(...) is in the same ballpark as 
failing to support "#". I can see Ben's point of view that we've reached 
that point even if I would have waited another three years, so if Ben 
would rather use $(...) I'd rather not overrule him downstream.