From: matz@... (Yukihiro Matsumoto)
Date: 2001-06-01T15:44:41+09:00
Subject: [ruby-talk:16121] Re: RCR: coexistent ruby versions installed by default

In message "[ruby-talk:16107] RCR: coexistent ruby versions installed by default"
    on 01/06/01, "Guy N. Hurst" <gnhurst@hurstlinks.com> writes:

|As for me, I accept the apology. I did spend considerable time
|trying to figure out this particular quirk. Time isn't free.

I'm sorry.  I confess I've ever fallen into this particular quirk
several times.  That's why I spend much time to seek the way to fix
it.

By the way, the another pitfall I've fallen in is

  p ("xx"*2).to_i

it's parsed as

  (p("xx"*2)).to_i

which is not my intention most of the cases.  I hacked the parser last
weekend (in 1.7.1 in the CVS), hopefully I succeeded to fix this one.

|On the other hand, I agree to allow old code to break...

Well, well, it depends on HOW MUCH code it would break.

|RCR?
|----
|I think it would do everyone a favor to change the default installation
|of Ruby to allow concurrent multiple versions of Ruby to exist.

OK.  You mean --program-suffix="$(MAJOR)$(MINOR)$(TEENY)" by default?

							matz.