July 2009 Archives
Jul 22 07:14 PM
Corehackers wiki policy change
As of Jul 21 2009, the corehackers wiki requires users
to create an account before editing or creating content. We made
this policy change for a lot of reasons. I'll focus on one of the
positive ones. I want to be able to give credit where credit is
due. If a random IP address uploads a fantastic new FAQ or howto, I
can't provide attribution. Having usernames gives us a way to
provide attribution and give credit where it is due.
Hmm... Speaking of attribution, we really should put a CC license on the wiki content. The question is which one. I lean towards CC-BY. If you'd like to weigh in, please toss me an email.
Hmm... Speaking of attribution, we really should put a CC license on the wiki content. The question is which one. I lean towards CC-BY. If you'd like to weigh in, please toss me an email.
Jul 12 08:46 PM
sungo vs unicode
I'm finally following p5p again.
Yesterday, it uncovered a yak that demanded to be shaved. Someone
posted an email with charset utf8 and some dreaded non-ascii
characters. How dare they! More importantly, something in the
Eterm-ssh-screen-mutt pipeline managed to corrupt the terminal. So,
I set out to get at least utf8 support in my commonly used
applications.
Firefox already supports unicode so I didn't have to worry about that. Next, E17 needed to support unicode in the titlebars. Easy peasy. I changed the standard font to Bitstream Vera Sans which supports utf8 and, I suspect, the full unicode set. Then came the difficult part. How to get my full terminal pipeline to handle those funny characters properly....
I'll save you the gnashing of teeth and give you the step by step:
It's important that the various environment variables be set everywhere, including the remote servers you're accessing.
mutt will automatically detect your environment and support the new charsets. For irssi, you need to run /set term_charset UTF-8.
From here, the only potential problem is the font you're using. If the font doesn't support unicode, none of the above matters. The font that I use for my terminals, 'fixed', has unicode support. As I mentioned above, the Bitstream family does too. If you're a fan of the font 'Anonymous', grab Anonymous Pro. Lots of other font families support unicode as well.
Hopefully, this will save someone some drama. :)
Firefox already supports unicode so I didn't have to worry about that. Next, E17 needed to support unicode in the titlebars. Easy peasy. I changed the standard font to Bitstream Vera Sans which supports utf8 and, I suspect, the full unicode set. Then came the difficult part. How to get my full terminal pipeline to handle those funny characters properly....
I'll save you the gnashing of teeth and give you the step by step:
- Switch to urxvt. (I've used Eterm forever but it doesn't support unicode and probably never will).
- export LC_CTYPE="en_US.utf8"; export LC_MESSAGES="en_US.utf8"; (That last var is important for applications like mutt.)
- export TERM=rxvt-unicode (Better, launch urxvt as urxvt -tn rxvt-unicode.)
- In ~/.screenrc, add defutf8 on. Alternately, launch screen as screen -U
It's important that the various environment variables be set everywhere, including the remote servers you're accessing.
mutt will automatically detect your environment and support the new charsets. For irssi, you need to run /set term_charset UTF-8.
From here, the only potential problem is the font you're using. If the font doesn't support unicode, none of the above matters. The font that I use for my terminals, 'fixed', has unicode support. As I mentioned above, the Bitstream family does too. If you're a fan of the font 'Anonymous', grab Anonymous Pro. Lots of other font families support unicode as well.
Hopefully, this will save someone some drama. :)
Jul 10 07:02 PM
Corehackers: Progress Report
I've been a bad project lead. I've
not contributed much this week to The Grand Movement. That's a
crazy day job for you. However, I'm very pleased to report that
lots of contributions have been made without me. The wiki is growing daily with all
sorts of good ideas and even a
magical perl5 api glossary (still in its infancy). chromatic opined on
the importance of corehackers and followed it up with strictperl,
a 'strict-by-default' patch to the perl5 core, based on the
corehackers codebase. To round things out, Rafael decided to cherry pick our early patches into
perl5-blead.
I'm ecstatic to see so many good things happening without any shepherding from me.
I'm ecstatic to see so many good things happening without any shepherding from me.
Jul 04 12:35 PM
Corehackers wiki
While I announced it on irc, I forgot
to announce it here.
The corehackers site/wiki is now live. Great thanks to ShadowCat Systems for hosting it.
The corehackers site/wiki is now live. Great thanks to ShadowCat Systems for hosting it.
Jul 04 12:10 PM
Actions are greater than words
DarkPAN, Modern Perl, perl5i,
corehackers.... ironman, epo, #corehackers, p5p/#p5p...
We as a community have been jibbering a lot about improving perl5. How we want more releases of perl5; how we want our favorite ideas in perl5; how various segments of the community may or may not matter to our immediate universe. God knows that I've contributed to the jibbering, for better or for worse.
It's time for something to happen. Rafael said it in a well written post. In case you missed it, I'm going to say this as simply as I know how.
Code/Docs or STFU
There are lots of places to help. If you need help, epo and corehackers are more than willing the point you in a direction, ranging from introductory to super-advanced. If you don't need help, subscribe to p5p, join #p5p on irc.perl.org, look at the perl5 rt queue and get to work.
I'm tired of talking. I'm tired of reading blog posts with no actions behind them. This is your perl and it's not going to improve on its own. p5p are hard at work on the core and, unless you are too, your "contributions" are just noise and static that are distracting our core team from writing code and doing things that matter.
We as a community have been jibbering a lot about improving perl5. How we want more releases of perl5; how we want our favorite ideas in perl5; how various segments of the community may or may not matter to our immediate universe. God knows that I've contributed to the jibbering, for better or for worse.
It's time for something to happen. Rafael said it in a well written post. In case you missed it, I'm going to say this as simply as I know how.
Code/Docs or STFU
There are lots of places to help. If you need help, epo and corehackers are more than willing the point you in a direction, ranging from introductory to super-advanced. If you don't need help, subscribe to p5p, join #p5p on irc.perl.org, look at the perl5 rt queue and get to work.
I'm tired of talking. I'm tired of reading blog posts with no actions behind them. This is your perl and it's not going to improve on its own. p5p are hard at work on the core and, unless you are too, your "contributions" are just noise and static that are distracting our core team from writing code and doing things that matter.
