astksh review

Igor Pechtchanski pechtcha@cs.nyu.edu
Thu May 22 02:48:00 GMT 2003


On Wed, 21 May 2003, Charles Wilson wrote:

> >>There is no problem with the regular configure/make (i.e., building
> >>the default configuration of the package).  However, if I wanted to,
> >>say, twiddle with the source and turn some features on/off, or
> >>experiment, or even, say, insert debug printouts for some bug that
> >>manifests itself when running shell scripts, it would be much easier
> >>to do so with pdksh -- I think I understand most of the source
> >>(haven't looked at it in a while).
>
> >>As an aside, it would be great if certain features/defines in the code
> >>were turned on/off with configure's "--enable-*"/"--disable-*" options
> >>(not sure if they are now, sounds like they aren't).
>
> People, let me introduce you to a "gift horse."  Please stop inspecting
> its teeth.
>
> Karsten and others have done a lot of work to accomodate the cygwin
> packaging standard.  It is *extremely* disrepectful for us (the cygwin
> community, as it were) to then insist on *internal* changes to the
> product itself.
>
> e.g. "Stop using nmake and this really confusing Makefile system.  Use
> autoconf"
>
> That is NOT helpful -- nor does it belong on this list.  We're talking
> about packaging ast-ksh so that it can be installed on cygwin using
> setup, and conform to our packaging rules.
>
> Those rules say nothing about requiring autoconf.  There's a tacit
> assumption that cygwin packages should be buildable on a cygwin system
> without additional, non-official cygwin packages -- but no requirement
> that any *particular* cygwin tool (like autoconf) be used.  (ast-ksh
> satisfies the build-on-stock-cygwin "rule", but not the
> thou-shalt-use-autoconf "rule")
>
> ast-ksh uses a Makefile-driven build system.  It is not autoconfed.  It
> never will be autoconfed.  And if it WERE to be autoconfed, that
> discussion would belong over on the ast-ksh mailing list, because you're
> talking about massive overall changes to the upstream package, not
> simply cygwin-specific porting tweaks.
>
> (FWIW, I vote "yes" on this package, as is)
> --Chuck

Chuck,

While I concede your point about the gift horse, I was NOT trying to get
Karsten to change any of the build procedures by my comment, or even
request that the configure script be augmented (a wistful comment was
made, and I apologize if I stepped on people's toes).

What I did adamantly vote AGAINST is deprecating the pdksh package.  As
long as both packages are allowed to coexist (and the user can select
which one he/she wants to use), I'm happy.
	Igor
-- 
				http://cs.nyu.edu/~pechtcha/
      |\      _,,,---,,_		pechtcha@cs.nyu.edu
ZZZzz /,`.-'`'    -.  ;-;;,_		igor@watson.ibm.com
     |,4-  ) )-,_. ,\ (  `'-'		Igor Pechtchanski
    '---''(_/--'  `-'\_) fL	a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton



More information about the Cygwin-apps mailing list