Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by ROFISH on Friday January 30, 2009 @04:14PM
(#458699)
I know the Rails team has worked hard towards getting 1.9 compatibility, but what are your opinions about using this on production servers? (I'm kinda edgey myself, it's why I ask.)
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by sho on Friday January 30, 2009 @10:10PM
(#459169)
I think Rails itself works, the problem is the servers. Mongrel doesn't work on 1.9, and the others have problems too. Also the DB libraries.
Plenty of other gems don't work either, basic ones like JSON (works only in pure Ruby mode). Hopefully the stable 1.9 release spurs maintainers on a bit, but I imagine it'll take a while.
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by nickb on Friday January 30, 2009 @06:16PM
(#458896)
I wonder what will happen to 1.9 branch. It's not supported well by anything. 1.9 is suppose to be a transitional branch anyway... a place where Matz can experiment a bit. 2.0 is suppose to be the next big thing (like Python 3000) but I have no idea if that will ever see the light of the day.
I'm placing my bets on JRuby, Rubinius and MagLev in the meantime.
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by acangiano on Friday January 30, 2009 @06:45PM
(#458928)
Actually nickb, 1.9.1 is no longer thought of as a transitional/development release. It's a stable release we are all expected to migrate to sooner or later.
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by carbon8 on Friday January 30, 2009 @08:30PM
(#459066)
This change in what version numbers mean definitely isn't known by the entire Ruby community. The O'Reilly books' statements about 1.9 versions being development versions and the fact that 1.9 has been in development until today has made this a little confusing.
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by catch23 on Friday January 30, 2009 @08:06PM
(#459041)
I think we were all just used to seeing even numbers as stable release versions. I don't think there's anything wrong with having odd number production release versions.
Also, 1.9 wasn't really an experimental place for Matz, it was initially YARV then Matz conceded it to be the next official ruby branch. None of the future VMs will/have been done by Matz. He actually joked about that during RubyConf 2007... the reason ruby VMs became fast was because Matz was taken out of the picture.
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by glomek on Friday January 30, 2009 @08:06PM
(#459043)
I heard years ago that Ruby 1.9 was going to have tail call optimization, but I can't find it anywhere in the news file. Does anyone know if this made it in or not?
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by catch23 on Friday January 30, 2009 @08:28PM
(#459065)
I think it's been implemented in 1.9 for a long time, but it's not enabled by default. You have to set a #define to configure it, however I don't know why it's not enabled by default. Maybe someone in the "know" can tell us why.
Re:Ruby 1.9.1 Released: First Stable Ruby 1.9 Release
by Freaky on Friday January 30, 2009 @09:25PM
(#459122)
"In fact, I don't check this option. So some bugs may hide on this option. And backtrace (what you see on exception) is changed. It's convention issue." -- http://www.ruby-forum.com/topic/164263
Plenty of other gems don't work either, basic ones like JSON (works only in pure Ruby mode). Hopefully the stable 1.9 release spurs maintainers on a bit, but I imagine it'll take a while.
I'm placing my bets on JRuby, Rubinius and MagLev in the meantime.
Also, 1.9 wasn't really an experimental place for Matz, it was initially YARV then Matz conceded it to be the next official ruby branch. None of the future VMs will/have been done by Matz. He actually joked about that during RubyConf 2007... the reason ruby VMs became fast was because Matz was taken out of the picture.