ASPN ActiveState Programmer Network
ActiveState
/ Home / Perl / PHP / Python / Tcl / XSLT /
/ Safari / My ASPN /
Cookbooks | Documentation | Mailing Lists | Modules | News Feeds | Products | User Groups


Recent Messages
List Archives
About the List
List Leaders
Subscription Options

View Subscriptions
Help

View by Topic
ActiveState
.NET Framework
Open Source
Perl
PHP
Python
Tcl
Web Services
XML & XSLT

View by Category
Database
General
SOAP
System Administration
Tools
User Interfaces
Web Programming
XML Programming


MyASPN >> Mail Archive >> ruby-talk
ruby-talk
Re: Installing all gems
by Bill Guindon other posts by this author
May 2 2005 2:49PM messages near this date
Re: Installing all gems | Re: Debugger performance
On 4/30/05, caleb clausen <google@[...].net>  wrote:
>  On Fri, 29 Apr 2005 07:52:54 +0900, Bill Guindon wrote:
>  
>  > On 4/28/05, caleb clausen <google@[...].net> wrote:
>  >> I've tried a couple of times to install all the rubygems packages.
>  >> Inevitably, I run into a problem with a gem that's either broken or
>  >> incompatible with something else. Can anyone please give me some
>  >> instructions for installing all (or as many as possible rubygems?
>  >>
>  >
>  > Somebody's gotta ask the obvious question... Why would you want to?
>  >
>  > Really, I'm curious.
>  
>  My goal is not just to make trouble, nor do I just want to have more toys
>  than anybody else. (Tho the idea does appeal to me....)
>  
>  I'm trying to test RubyLexer. Part of RubyLexer's tests include a script
>  that can dertermine (not certainly, but with a high degree of accuracy)
>  whether RubyLexer is lexing a particular piece of source code in exactly
>  the same way as ruby itself. I've run this script (and fixed the problems
>  it found) against every bit of ruby source I could find on my system, and
>  now I'm looking for more.
>  
>  I want to make a parser as well, and I'll need to test that. I think
>  there's a variation of my current test strategy that will tell me, for an
>  arbitrary source file, whether it has been parsed correctly.
>  
>  Basically, rubygems will all be a sort of ad-hoc language test suite.
>  
>  This is the sort of thing that's very easy in Debian. That's how I got all
>  Debian packages that depend on ruby, which form part of my collection
>  already. I don't know the reason, maybe because there's more qc of
>  packages upfront, or because apt checks for problems before even
>  downloading, or because installation of one package doesn't abort the
>  whole process. But handing off a long list of packages, some of which are
>  bad or broken, to apt-get works very smoothly. Not so with rubygems. I
>  realize Debian is a much more mature system... I'm really just looking for
>  a workaround. If a package is too broken to install, I don't want it.
>  Maybe there's someone out there with lots of gems already installed who
>  can give me a list of non-broken packages?

Forgive the delayed reply, somehow gmail buried this one.

Ahh, I feel better knowing that there actually was a logical reason
for doing what seemed quite strange :)

Might pay to sign up for the gem developers mailing list [1].   It's
an interesting problem, and finding some solution to it would be
helpful (ie: some system for marking stability, etc. -- or gracefully
skipping failed installs)

[1] http://rubyforge.org/mailman/listinfo/rubygems-developers

-- 
Bill Guindon (aka aGorilla)
Thread:
Caleb Clausen
Bill Guindon

Privacy Policy | Email Opt-out | Feedback | Syndication
© ActiveState Software Inc. All rights reserved