hob x 0.2 .. blogink for ample abilities .. come into the arms of..

[why the lucky stiff]

ok, ruby-talk. enough.

the 1st public officially sealed Hobix has arrived. i've been coding
this thing since april. polishing and polysheeng.

Hobix is a complete blogging system, designed to be managed on the
file system and accessed through a command-line application.
Hobix is centrally powered by Textile, YAML and Ruby.

The command-line application is powered by a complete Ruby library
which
is designed to be fully scriptable and extensible.

Hobix comes with complete documentation in ri. Type: ri Hobix
You'll see a complete overview of classes and some examples.

For help with the commandline: hobix help

here's where you do this:

ruby -ropen-uri -e 'eval(open("http://go.hobix.com/").read)'

ride the automated web gondola. i love you slimy people.

hey, and i say join the Hobix Is The Way mailing list for daily tips and
discussion:
http://rubyforge.org/mailman/listinfo/hobix-is-the-way

i love all slimy people.

_why

 

[Lyle Johnson]

ride the automated web gondola. i love you slimy people.

Well said. I'm looking forward to trying this out. Would I be correct
to assume that your site (http://whytheluckystiff.net) is also powered
by Hobix?

 

[why the lucky stiff]

Well said. I'm looking forward to trying this out. Would I be correct
to assume that your site (http://whytheluckystiff.net) is also powered
by Hobix?

Since May, yes.
(http://whytheluckystiff.net/clog/meep/stealingMyBabies.html)

_why

 

[Vincent Isambart]

I just tried to install hobix and I got the following error when
creating a weblog named 'test' in '~/test-hobix', during the end of the
installation started with ruby -ropen-uri -e
'eval(open("http://go.hobix.com/").read)':
-e:1: /sw/lib/ruby/1.9/fileutils.rb:467:in `mkdir': No such file or
directory - /Users/vincent/test-hobix/entries (Errno::ENOENT)
from /sw/lib/ruby/1.9/fileutils.rb:467:in `_copy_entry'
from /sw/lib/ruby/1.9/fileutils.rb:448:in `copy_entry'
from /sw/lib/ruby/1.9/fileutils.rb:447:in `preserve'
from /sw/lib/ruby/1.9/fileutils.rb:447:in `copy_entry'
from /sw/lib/ruby/1.9/fileutils.rb:372:in `cp_r'
from /sw/lib/ruby/1.9/fileutils.rb:370:in `fu_traverse'
from /sw/lib/ruby/1.9/fileutils.rb:370:in `cp_r'
from /sw/lib/ruby/1.9/fileutils.rb:368:in `fu_each_src_dest'
... 8 levels...
from /sw/lib/ruby/site_ruby/1.9/hobix/commandline.rb:326:in
`setup_blogs'
from (eval):183
from (eval):124:in `each'
from (eval):124
After some checking, I got a directory '~' created where I started the
program P
It contains just an empty directory called test-hobix. I guess it did
not like directory names with '~' in them... I do not know if you can
correct it, but I may not be the only one to enter a directory name
staring with '~'.
When I created a blog with the following command, there was no problem:
hobix create test ~/test-hobix

An other comment: Does it work with webrick? The installation mentions
apache, but does it require it?

Vincent Isambart

 

[trans. (T. Onoma)]

ruby -ropen-uri -e 'eval(open("http://go.hobix.com/").read)'

Scary powerful is the Hobix'.

--
( o _
// trans.
/ \ (e-mail address removed)

I don't give a damn for a man that can only spell a word one way.
-Mark Twain

 

[why the lucky stiff]

After some checking, I got a directory '~' created where I started the
program P
It contains just an empty directory called test-hobix. I guess it did
not like directory names with '~' in them... I do not know if you can
correct it, but I may not be the only one to enter a directory name
staring with '~'.
When I created a blog with the following command, there was no
problem: hobix create test ~/test-hobix

Indeed, thankyou, this was a problem. The web installer has been
updated to reflect this fix. Update again with:

ruby -ropen-uri -e 'eval(open("http://go.hobix.com/").read)'

Don't worry. It'll keep your existing configuration.

Also, I've just added `hobix upgrade' which does the above.

An other comment: Does it work with webrick? The installation mentions
apache, but does it require it?

Hobix generates static HTML by default, so it'll work with any web server.

For example, here's a WEBrick server which will scan your ~/.hobixrc and
setup a subdirectory for each one.

#!/usr/local/bin/ruby
require 'webrick'
include WEBrick

s = HTTPServer.new(
ort => 2000,
ocumentRoot => Dir:wd + "/htdocs"
)

## mount subdirectories
require 'hobix/config'
require 'hobix/weblog'
config = File.open( File.expand_path( "~/.hobixrc" ) ) { |f|
YAML::load( f ) }
config['weblogs'].each do |name, path|
weblog = Hobix::Weblog.load( path )
s.mount("/#{ name }", HTTPServlet::FileHandler, weblog.output_path)
end

trap("INT"){ s.shutdown }
s.start

So, if you have a blog which you call 'test', it will appear at
http://localhost:2000/test/. In order for this to work right, that URL
needs to be added as the 'link' in that blog's hobix.yaml file.

_why

 

[Marcel Molina Jr.]

Scary powerful is the Hobix'.

that is, indeed, a *very* nifty installer...

marcel

 

[Mauricio Fernández]

ok, ruby-talk. enough.

the 1st public officially sealed Hobix has arrived. i've been coding
this thing since april. polishing and polysheeng.

Hobix is a complete blogging system, designed to be managed on the
file system and accessed through a command-line application.
Hobix is centrally powered by Textile, YAML and Ruby.

Just RPAfied it, so it's now in the preliminary Ruby Production Archive
(http://rpa-base.rubyforge.org)

The rpafied version solves a minor installation issue: the upstream
installer allows you to select the destination dirs for bin/ and lib/,
but the data files are installed unconditionally under ruby's $prefix.

batsman@tux-chan:~/src/rpa/ports$ rpa update --no-proxy
Getting port info from http://rpa-base.rubyforge.org/ports/ports.info.
100% [========================================] 80450 bytes

Ports added since the last update (in the last 14 days):
hobix 0.2a-1 Flexible generation of static blogs.

batsman@tux-chan:~/src/rpa/ports$ rpa query -x hobix
Matching available ports:
name: hobix
version: 0.2a-1
classification: Top.Application
requires: redcloth
description: Flexible generation of static blogs.

Hobix is a complete blogging system, designed to be managed on the
file system and accessed through a command-line application. It brings
together Textile, YAML and erb to create a powerful system for static
generation of blog sites.

batsman@tux-chan:~/src/rpa/ports$ rpa install hobix
Installing ports
Getting port hobix from http://rpa-base.rubyforge.org/ports/hobix_0.2a-1.rps.
100% [========================================] 30720 bytes
Building hobix (0.2a-1).
Building dependencies redcloth.
Generating RI data files.
Generating RDoc HTML documentation.
Fixed shebang in rpa/tmp/bin/hobix.
Calculating MD5 digests.
Building package in hobix_0.2a-1_i686-pc-linux-gnu.rpa.

===============================================================================
You might want to run
hobix setup_blogs
now to setup your hobix configuration now.

The documentation is available as
ri-rpa -p hobix Hobix
or in HTML format under $prefix/share/doc/rpa0.0/hobix/rdoc
===============================================================================

Installing hobix
Reusing cached package /home/batsman/usr/lib/ruby/rpa0.0/packages/hobix_0.2a-1_i686-pc-linux-gnu.rpa.
Installing dependencies redcloth.
Starting lightweight (metadata only) transaction for hobix
Reusing cached package /home/batsman/usr/lib/ruby/rpa0.0/packages/redcloth_2.0.11-2_i686-pc-linux-gnu.rpa.
Package redcloth unchanged.
Checking for file conflicts in hobix.
Starting transaction for hobix
Package /home/batsman/usr/lib/ruby/rpa0.0/packages/hobix_0.2a-1_i686-pc-linux-gnu.rpa unpacked.
Finished transaction for hobix
Starting lightweight (metadata only) transaction for hobix
Finished lightweight (metadata only) transaction for hobix
Finished lightweight (metadata only) transaction for hobix
Committed changes

 

[Gavin Sinclair]

that is, indeed, a *very* nifty installer...

Let's hope _why's site doesn't get hacked. Wouldn't want to eval
something nasty!

Gavin

 

[Stefan Schmiedl]

Indeed, thankyou, this was a problem. The web installer has been
updated to reflect this fix. Update again with:

ruby -ropen-uri -e 'eval(open("http://go.hobix.com/").read)'

Maybe I deserve this for using a CVS version, but I can't create
a new blog here:

$ ruby -v
ruby 1.9.0 (2004-09-10) [i686-linux]

.....
The default blog is available in the following modes:
apache1-ssi, apache2-ssi

Mode: [Enter for none]
/usr/local/lib/ruby/1.9/rexml/parsers/treeparser.rb:1:in `require': No
such file to load -- rexml/validation/validationexception (LoadError)
from /usr/local/lib/ruby/1.9/rexml/parsers/treeparser.rb:1
from /usr/local/lib/ruby/1.9/rexml/document.rb:12:in `require'
from /usr/local/lib/ruby/1.9/rexml/document.rb:12
from /usr/local/lib/ruby/site_ruby/1.9/hobix/out/rss.rb:18:in
`require'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/out/rss.rb:18
from /usr/local/lib/ruby/site_ruby/1.9/hobix/base.rb:38:in
`require'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/base.rb:38:in
`start'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:280:in
`start'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:279:in
`each'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:279:in
`start'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:287:in
`load'
from
/usr/local/lib/ruby/site_ruby/1.9/hobix/commandline.rb:161:in
`join_as_author'
from
/usr/local/lib/ruby/site_ruby/1.9/hobix/commandline.rb:137:in
`create_weblog'
from /usr/local/bin/hobix:41:in `call'
from /usr/local/bin/hobix:41

 

[Kent Sibilev]

Hm,

I have hobix v0.2a installed.

$ ruby -v
ruby 1.9.0 (2004-09-11) [powerpc-darwin7.5.0]

$ hobix create blog
/usr/local/bin/hobix:41:in `create_weblog': wrong number of arguments
(1 for 2) (ArgumentError)
from /usr/local/bin/hobix:41:in `call'
from /usr/local/bin/hobix:41

$ hobix create blog blog
*** Creation of weblog `blog' will add the following directory"
structure to directory /Users/ksibilev/tmp/blog"

/Users/ksibilev/tmp/blog
hobix.yaml <- configuration

entries/ <- edit and organize
your news items,
articles and so on.

skel/ <- contains your
templates

htdocs/ <- html is created here,
store all your images here,
this is your viewable
websyht

lib/ <- extra hobix libraries
(plugins) go here

Create this structure? [y/N]: y
The default blog is available in the following modes:
apache1-ssi, apache2-ssi

Mode: [Enter for none]
*** Joining blog `a bokononist asks', adding you as author.

$ hobix regen blog
[Building yearly pages]
## Page: /2004/index.html, updated Sat Sep 11 10:19:18 EDT 2004
/usr/local/lib/ruby/site_ruby/1.9/hobix/out/quick.rb:62:in `load':
Error `undefined method `content' for nil:NilClass' in erb
/Users/ksibilev/tmp/blog/skel/yearly.html.quick-archive.
(Hobix::Out::QuickError)
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:412:in
`retouch'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:380:in
`build_pages'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:301:in
`skel_yearly'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:504:in
`each_with_neighbors'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/base.rb:133:in
`times'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/base.rb:133:in
`each_with_neighbors'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:504:in
`skel_yearly'
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:301:in
`call'
... 8 levels...
from /usr/local/lib/ruby/site_ruby/1.9/hobix/weblog.rb:363:in
`regenerate'
from
/usr/local/lib/ruby/site_ruby/1.9/hobix/commandline.rb:231:in
`regen_action'
from /usr/local/bin/hobix:55:in `call'
from /usr/local/bin/hobix:55

Cheers,
Kent.

 

[why the lucky stiff]

/usr/local/lib/ruby/1.9/rexml/parsers/treeparser.rb:1:in `require': No
such file to load -- rexml/validation/validationexception (LoadError)

It looks like your REXML installation is screwed up. Go to your Ruby
CVS checkout and use `cvs update -d'. This should pull down the
`rexml/validation' directories.

Recompile 1.9.0 and take another shot.

_why

 

[why the lucky stiff]

$ hobix create blog
/usr/local/bin/hobix:41:in `create_weblog': wrong number of arguments (1
for 2) (ArgumentError)
from /usr/local/bin/hobix:41:in `call'
from /usr/local/bin/hobix:41

This is better now. Thankyou:

bash-2.05b$ hobix blogs test
*** wrong number of arguments (1 for 0)
*** use syntax: `hobix blogs'

bash-2.05b$ hobix list why too many args
*** wrong number of arguments (4 for 2)
*** use syntax: `hobix list weblog-name search/path'

$ hobix regen blog
[Building yearly pages]
## Page: /2004/index.html, updated Sat Sep 11 10:19:18 EDT 2004
/usr/local/lib/ruby/site_ruby/1.9/hobix/out/quick.rb:62:in `load': Error
`undefined method `content' for nil:NilClass' in erb
/Users/ksibilev/tmp/blog/skel/yearly.html.quick-archive.
(Hobix::Out::QuickError)

This is fixed as well. You'll need to ditch that blog and create a new one.

If you want to salvage the current blog, you'll need to add a `linklist'
to that blog's hobix.yaml. The linklist is a sequence of link titles
and link URLs (like a blogroll I guess). The syntax is:

linklist:
- hobix: http://hobix.com/
- del.icio.us: http://del.icio.us/

The default templates use the linklist and 2.0a wasn't dumping it for
you. But now it's fixed thanks to your ample abilities. Slime out.

_why

 

[why the lucky stiff]

Just RPAfied it, so it's now in the preliminary Ruby Production Archive
(http://rpa-base.rubyforge.org)

RPA is awesome. You are so _on_, bats.

The rpafied version solves a minor installation issue: the upstream
installer allows you to select the destination dirs for bin/ and lib/,
but the data files are installed unconditionally under ruby's $prefix.

Thankya, this is fixed at go.hobix.com as well.

_why

 

[Martin DeMello]

RPA is awesome. You are so _on_, bats.

+1!

martin

 

[Joel VanderWerf]

Let's hope _why's site doesn't get hacked. Wouldn't want to eval
something nasty!

Dunno. Is this any worse than manually downloading a .tgz or .gem from a
site that may have been hacked?

 

[Zach Dennis]

I am having difficultly getting a sample configuration to work. I am
trying to use the default.cfg file that comes in the Arrow-0.1.0.tar.gz
download. Everything appears to be working between Arrow and Apache, but
I have a feeling I have an improper config file for arrow.

Here is my setup:

- Ruby 1.8.2
- Apache 1.3.31
- Mod Ruby 1.2.2
- I have moved my default Arrow-0.1.0 directory to /etc/arrow
- I have /var/www/basketballminds.com/ directories and no directories
underneath of it

Here's what I /var/log/apache/error.log file:

File Edit Options Buffers Tools Help
[Sat Sep 11 14:31:09 2004] [notice] SIGHUP received. Attempting to restart
[Sat Sep 11 14:31:09 2004] [notice] Apache/1.3.31 (Debian GNU/Linux)
mod_ruby/1.2.2 Ruby/1.8.2(2004-07-29) configured -- re\suming normal
operations
[Sat Sep 11 14:31:09 2004] [notice] Accept mutex: sysvsem (Default: sysvsem)
[Sat Sep 11 14:31:13 2004] [notice] (global) : Arrow config file is
"/etc/arrow/default.cfg"
[Sat Sep 11 14:31:13 2004] [notice] Arrow:ispatcher : Configuring a
dispatcher for 'new.basketballminds.com': child serve\r 27256
[Sat Sep 11 14:31:13 2004] [notice] Arrow:ispatcher : Monitor skipped
by configuration
[Sat Sep 11 14:31:13 2004] [notice] Arrow:ispatcher : Setting global
log level to :info
[Sat Sep 11 14:31:13 2004] [notice] Arrow:ispatcher : Configuring the
Session class with #<Arrow::Config:0x40722a94
@load\er=#<Arrow::Config::YamlLoader:0x40728bb0>,
@name="/etc/arrow/default.cfg", @createTime=Sat Sep 11 14:31:13 UTC
2004, @stru\ct=#<Arrow::Config::ConfigStruct:0x407206e0 @modified=false,
@hash={:templates=>{:loader=>"Arrow::Template", ath=>["templ\ates",
"/www/templates"], :cacheConfig=>{:maxSize=>2621440, :expiration=>36,
:maxNum=>20, :maxObjSize=>131072}, :cache=>tru\e}, :startMonitor=>false,
:session=>{:storeType=>"file:/tmp", :expires=>"+48h",
:idName=>"arrow-session", :rewriteUrls=>tru\e, :idType=>"md5:.",
:lockType=>"recommended"}, :applets=>{:missingApplet=>"/missing",
ath=>["applets", "/www/applets", "\/etc/arrow/applets"],
ollInterval=>5, attern=>"*.rb", :errorApplet=>"/error",
:defaultApplet=>"/status", :config=>{}, :\layout=>{}},
:logLevel=>"info", :templateLogLevel=>"notice"}>>
[Sat Sep 11 14:31:13 2004] [notice] Arrow:ispatcher : Creating request
broker
[Sat Sep 11 14:31:13 2004] [error] Arrow::Broker : Configured
MissingApplet handler (/missing) doesn't exist
[Sat Sep 11 14:31:13 2004] [notice] Arrow::Broker : Using builtin
missing-applet handler.
[Sat Sep 11 14:31:13 2004] [error] Arrow::Broker : Applet returned false
value. Setting status to DECLINED
[Sat Sep 11 14:31:13 2004] [notice] Arrow:ispatcher : Transaction has
non-OK status: -1
[Sat Sep 11 14:31:13 2004] [error] [client 127.0.0.1] File does not
exist: /var/www/basketballminds.com/arrow/

Any help would be greatly appreciated! Thanks,

Zach

 

[James Britt]

Gavin Sinclair wrote:
...


Dunno. Is this any worse than manually downloading a .tgz or .gem from a
site that may have been hacked?

Somewhat, in that you can first examine the source before running it.

I suspect that, in actual practice, most people simply trust the source.

But having a tarball also means you can re-install an earlier version if
a newer one causes problems (assuming you keep the old one around.)

James

 

[Stefan Schmiedl]

It looks like your REXML installation is screwed up. Go to your Ruby
CVS checkout and use `cvs update -d'. This should pull down the
`rexml/validation' directories.

scary how some folks now what's happening on my computer ...
and don't ask how much tcl/tk stuff was missing here

Thanks for the quick fix.

s.

 

[Stefan Schmiedl]

scary how some folks now what's happening on my computer ...
and don't ask how much tcl/tk stuff was missing here

Thanks for the quick fix.

However, I have a second computer ... ;->

There's a Win2k on it, and ruby 1.8.2 (2004-07-29) (typical windows
installation via installer). Downloading and installation worked,
but when I created the "blahhg" blog (as per tutorial),
"hobix regen blahhg" did not create the files on the windows box,
as it did on the linux box.

The reason is that somehow some of the @..._path variables
of the weblog contain "c:/blogs/blahhg/c:/blogs/blahhg/...",
but currently I don't have time to dig further into this.

s.