# Ruby Weekly News 5th - 11th June 2006

**URL:** <https://rubytalk.org/t/ruby-weekly-news-5th-11th-june-2006/28201>\
**Category:** ruby-talk\
**Created:** [14 June 2006 10:13 UTC](https://rubytalk.org/t/ruby-weekly-news-5th-11th-june-2006/28201 "2006-06-14T10:13:20Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Tim\_Sutherland1](https://avatars.discourse-cdn.com/v4/letter/t/8e7dd6/32.png) [@Tim\_Sutherland1](https://rubytalk.org/u/Tim_Sutherland1)\
**Post date:** [14 June 2006 10:13 UTC](https://rubytalk.org/t/ruby-weekly-news-5th-11th-june-2006/28201/1 "2006-06-14T10:13:20Z")

</div>

[http://www.rubyweeklynews.org/20060611.html](http://www.rubyweeklynews.org/20060611.html)

Use the link above for a nicer version of this newsletter that's formatted for  
the web, including links to threads and other sites. You can also subscribe  
to a newsletter-only mailing list, or to RSS/Atom feeds.

Ruby Weekly News 5th - 11th June 2006

> **···**
>
> =====================================
> 
> &nbsp;&nbsp;&nbsp;Ruby Weekly News is a summary of the week's activity for the ruby-talk  
> &nbsp;&nbsp;&nbsp;mailing list (and its mirror equivalents the ruby-talk google group and  
> &nbsp;&nbsp;&nbsp;the Ruby forum) and the comp.lang.ruby newsgroup.
> 
> &nbsp;&nbsp;&nbsp;This week's newsletter is brought to you by Tim Sutherland.
> 
> &nbsp;&nbsp;&nbsp;[Contribute to the next newsletter]
> 
> # Articles and Announcements
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Ruby Whitepapers (ruby-talk)  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;------------------------------
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Eric Armstrong wrote a series of articles Ruby Makes My Head Explode  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;in a one-week spree:
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Ruby Rocks!  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Rake Rocks!  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Getting Started in Ruby  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Installing Ruby (Linux info added by John Gabriele)  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Installing Ruby Gems  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Adding Ruby Extensions  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Which GUI Language to Use? (FxRuby vs. WxRuby)
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Feel like contributing? (comp.lang.ruby)  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Gregarican asks for help completing a Ruby telephony library he'd been  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;working on before running out of time.
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"It provides a working telephony library that communicates with PBX  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;systems for call detail reporting, call control, etc. It's maybe  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;50-75% complete and hasn't been updated by me since November 2005.  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;This \*isn't\* a VoIP project. It's CTI."
> 
> # User Group News
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Phoenix Ruby Users Group meeting this Monday, June 12 (ruby-talk)  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;-------------------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;The Phoenix Ruby Users Group had their June meeting on the 12th.
> 
> # Link of the Week
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\* Ruby Inside: Daily Ruby tips, news, code and frivolity  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;--------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Ruby Inside is a blog by Peter Cooper that's been running for a couple  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;of weeks. So far it's had multiple interesting posts every day,  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;ranging from original overviews of little-known Ruby libraries, to  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;announcements of new tools, libraries and articles.
> 
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Worth a look.
> 
> # Threads
> 
> &nbsp;&nbsp;Another Look at SELECTOR NAMESPACES (ruby-talk)  
> &nbsp;&nbsp;-----------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Trans posted some simple code adding partial support for selector  
> &nbsp;&nbsp;&nbsp;namespaces in Ruby. It works by making copies of top-level constants, and  
> &nbsp;&nbsp;&nbsp;is only for demonstration purposes.
> 
> &nbsp;&nbsp;XML Parsing Speed - ruby libxml & REXML (comp.lang.ruby)  
> &nbsp;&nbsp;--------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;subimage's code was taking too long to parse XML files (some of which were  
> &nbsp;&nbsp;&nbsp;400MB). He tried using libxml instead of REXML, and this helped, but it  
> &nbsp;&nbsp;&nbsp;was still too slow.
> 
> &nbsp;&nbsp;&nbsp;rcoder explained that he should use stream (event) parsing for large XML  
> &nbsp;&nbsp;&nbsp;documents, instead of parsing entire documents into memory at once (and to  
> &nbsp;&nbsp;&nbsp;make things worse, then using XPath to select nodes).
> 
> &nbsp;&nbsp;&nbsp;After switching to REXML's stream-parsing mode, subimage remarked "Ok so I  
> &nbsp;&nbsp;&nbsp;finally dug into the stream parser and this is lightning fast!"
> 
> &nbsp;&nbsp;&nbsp;Finally, Mathieu Blondel tested the expat bindings for Ruby a few months  
> &nbsp;&nbsp;&nbsp;ago and found its stream-parser to be up to 20 times faster than REXML's.
> 
> &nbsp;&nbsp;Can Ruby play with COM ? (ruby-talk)  
> &nbsp;&nbsp;------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Can Ruby talk to Windows software that has a COM interface? Absolutely,  
> &nbsp;&nbsp;&nbsp;with the standard library `win32ole'.
> 
> &nbsp;&nbsp;Runtime debugging tools needed?!! (comp.lang.ruby)  
> &nbsp;&nbsp;--------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;The web-testing tool Watir brought Jonathan Ni to Ruby, and he now has  
> &nbsp;&nbsp;&nbsp;some ideas for speeding up test development. His company has a large and  
> &nbsp;&nbsp;&nbsp;complex set of Watir tests for their web application, and need to  
> &nbsp;&nbsp;&nbsp;constantly modify the tests as the application they're developing changes.
> 
> &nbsp;&nbsp;&nbsp;They're finding the `run test suite', `fix error in test', `re-run' cycle  
> &nbsp;&nbsp;&nbsp;takes too long (partly due to it taking a long time to set up the test  
> &nbsp;&nbsp;&nbsp;data), so Jonathan would like to run the tests and have them pause  
> &nbsp;&nbsp;&nbsp;whenever an error occurs, allow him to edit the test code, and continue  
> &nbsp;&nbsp;&nbsp;without restarting.
> 
> &nbsp;&nbsp;&nbsp;Andrew McDonagh suggested he instead consider whether all his tests really  
> &nbsp;&nbsp;&nbsp;need to be "end-to-end" (i.e. act exactly like a normal user, going  
> &nbsp;&nbsp;&nbsp;through a web browser). "A more reliable, flexible, less maintenance  
> &nbsp;&nbsp;&nbsp;approach is to partition the tests so that very few actually go  
> &nbsp;&nbsp;&nbsp;end-to-end." He gave some examples of what he means by tests that aren't  
> &nbsp;&nbsp;&nbsp;end-to-end.
> 
> &nbsp;&nbsp;&nbsp;Jonathan could also look into making the tests more independent, so he can  
> &nbsp;&nbsp;&nbsp;fix and re-run each small test without having to wait for the whole lot to  
> &nbsp;&nbsp;&nbsp;complete.
> 
> &nbsp;&nbsp;&nbsp;The original debug/fix/continue idea is a good one though. It would be  
> &nbsp;&nbsp;&nbsp;worth looking into the various Ruby debuggers out there. Perhaps the  
> &nbsp;&nbsp;&nbsp;ruby-talk readers have some ideas? (It's worth raising awareness that a  
> &nbsp;&nbsp;&nbsp;lot of programmers are coming to Ruby through Watir.)
> 
> &nbsp;&nbsp;Ruby Weekly News 29th May - 4th June 2006 (ruby-talk)  
> &nbsp;&nbsp;-----------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;In the thread announcing last week's Ruby Weekly News, James Edward Gray  
> &nbsp;&nbsp;&nbsp;II said he was "90% sure" he'd found a new host for the ruby-talk mailing  
> &nbsp;&nbsp;&nbsp;list comp.lang.ruby newsgroup gateway. "Give me a couple of weeks to iron  
> &nbsp;&nbsp;&nbsp;out all the details...", then hopefully the list and newsgroup will be  
> &nbsp;&nbsp;&nbsp;mirrored again.
> 
> &nbsp;&nbsp;Ruby for Highschoolers? (ruby-talk)  
> &nbsp;&nbsp;-----------------------------------
> 
> &nbsp;&nbsp;&nbsp;This thread discussed the relative merits of Scheme and Ruby as languages  
> &nbsp;&nbsp;&nbsp;for a high-school introductory programming course.
> 
> &nbsp;&nbsp;&nbsp;Scheme is a smaller, simpler language, and there is a lot of teaching  
> &nbsp;&nbsp;&nbsp;material already prepared for it. While Ruby has many features (e.g.  
> &nbsp;&nbsp;&nbsp;plenty of syntactic sugar) that make it nice to develop in once you know  
> &nbsp;&nbsp;&nbsp;the language, these can be confusing to beginners.
> 
> &nbsp;&nbsp;&nbsp;There were also arguments that Ruby is an excellent language for  
> &nbsp;&nbsp;&nbsp;beginners. Curt Hibbs points out "Learn To Program" by Chris Pine,  
> &nbsp;&nbsp;&nbsp;a book (and online tutorial) that's geared towards those who have never  
> &nbsp;&nbsp;&nbsp;programmed before. It teaches programming using Ruby.
> 
> &nbsp;&nbsp;&nbsp;Ruby has an advantage over Scheme in that it has a rich collection of  
> &nbsp;&nbsp;&nbsp;libraries, so even new programmers can write useful, real (as well as fun)  
> &nbsp;&nbsp;&nbsp;programs - great for motivation.
> 
> &nbsp;&nbsp;iterative question (ruby-talk)  
> &nbsp;&nbsp;------------------------------
> 
> &nbsp;&nbsp;&nbsp;Joe wants to get an array of all Dates in a range.
> 
> &nbsp;&nbsp;&nbsp;Daniel Schierbeck said (start\_date..end\_date).to\_a would do the trick, and  
> &nbsp;&nbsp;&nbsp;Chris showed an alternative approach, using Date#upto to iterate over the  
> &nbsp;&nbsp;&nbsp;date range.
> 
> &nbsp;&nbsp;Ruby Quiz Wins Contest (ruby-talk)  
> &nbsp;&nbsp;----------------------------------
> 
> &nbsp;&nbsp;&nbsp;Last week's Ruby Quiz, "Hash to OpenStruct (#81)" by Hans Fugal had  
> &nbsp;&nbsp;&nbsp;enough entries to qualify for a signed copy of James Edward Gray II's  
> &nbsp;&nbsp;&nbsp;"Best of Ruby Quiz" book.
> 
> &nbsp;&nbsp;&nbsp;Cool.
> 
> &nbsp;&nbsp;Anyone aware of a Ruby-module for ICQ? (ruby-talk)  
> &nbsp;&nbsp;--------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Groleo Marius wanted to know if Ruby has any libraries for talking to ICQ.
> 
> &nbsp;&nbsp;&nbsp;Net::IM by Paul Vaillant will support the protocol in the future, and  
> &nbsp;&nbsp;&nbsp;Francis Cianfrocca has implemented Oscar/AOL (the AOL instant messaging  
> &nbsp;&nbsp;&nbsp;protocol, including ICQ) in Ruby, and has it running on several large  
> &nbsp;&nbsp;&nbsp;sites, but has not released the code.
> 
> &nbsp;&nbsp;&nbsp;In the end, Groleo decided to port the Perl library Net::Oscar to Ruby.
> 
> &nbsp;&nbsp;[ruby-forum.com](http://ruby-forum.com) (ruby-talk)  
> &nbsp;&nbsp;--------------------------
> 
> &nbsp;&nbsp;&nbsp;[ruby-forum.com](http://ruby-forum.com) is a collection of web forums that are gateways to  
> &nbsp;&nbsp;&nbsp;Ruby-related mailing lists. (Posts to the lists show up on the forums, and  
> &nbsp;&nbsp;&nbsp;posts from the forums show up on the lists.)
> 
> &nbsp;&nbsp;&nbsp;Unfortunately, many users of the Ruby web forum apparently don't realise  
> &nbsp;&nbsp;&nbsp;their messages are sent to the ruby-talk list, so they do things (like not  
> &nbsp;&nbsp;&nbsp;quoting messages) which make perfect sense on a web forum, but cause  
> &nbsp;&nbsp;&nbsp;inconvenience for everyone else. In addition, posters from the forum are  
> &nbsp;&nbsp;&nbsp;often newbies who ask questions that have been answered many times before.
> 
> &nbsp;&nbsp;&nbsp;Matthew Smillie suggests the forum have links to [ruby-doc.org](http://ruby-doc.org) and others,  
> &nbsp;&nbsp;&nbsp;to encourage ruby-forum users to check the documentation before posting  
> &nbsp;&nbsp;&nbsp;questions.
> 
> &nbsp;&nbsp;&nbsp;\> [...] I'm also of the view that the quality and tone of the questions  
> &nbsp;&nbsp;&nbsp;\> from ruby-forum often leave something to be desired.  
> &nbsp;&nbsp;&nbsp;\>  
> &nbsp;&nbsp;&nbsp;\> I have a hunch this is because google groups and mailing lists tend to  
> &nbsp;&nbsp;&nbsp;\> be used by people who already know what they're looking for, and already  
> &nbsp;&nbsp;&nbsp;\> know where the docs live. I'd bet that a number of the people posting  
> &nbsp;&nbsp;&nbsp;\> from the forum are probably not aware of all the resources available to  
> &nbsp;&nbsp;&nbsp;\> them.
> 
> &nbsp;&nbsp;&nbsp;The thread turned into a discussion of how to read ruby-talk, which now  
> &nbsp;&nbsp;&nbsp;receives over 4000 messages per month. (Eric Hodel linked to some code of  
> &nbsp;&nbsp;&nbsp;his for using statistical techniques for selecting "interesting" posts.)
> 
> &nbsp;&nbsp;&nbsp;Dave Burt: "Does anyone read all 4000 posts per month?" Austin Ziegler and  
> &nbsp;&nbsp;&nbsp;Hal Fulton say they try. Your Ruby Weekly News editor reads almost every  
> &nbsp;&nbsp;&nbsp;subject line 🙂
> 
> &nbsp;&nbsp;How can I pin a Ruby object in memory? (ruby-talk)  
> &nbsp;&nbsp;--------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;A C extension question: in .NET your C code can't simply store a pointer  
> &nbsp;&nbsp;&nbsp;to a managed object, since the .NET garbage collector can move objects  
> &nbsp;&nbsp;&nbsp;around in memory (the main implementation has a compacting garbage  
> &nbsp;&nbsp;&nbsp;collector), so you sometimes have to wrap objects in a GCHandle, which is  
> &nbsp;&nbsp;&nbsp;guaranteed not to move.
> 
> &nbsp;&nbsp;&nbsp;John Lam asked if the same is required in Ruby C extensions. Minkoo Seo  
> &nbsp;&nbsp;&nbsp;said no, since Ruby's collector is not a compacting one.
> 
> &nbsp;&nbsp;&nbsp;(C extensions are allowed to depend on this fact. At least, objects  
> &nbsp;&nbsp;&nbsp;referenced by VALUEs on the stack or in registers, or those referenced by  
> &nbsp;&nbsp;&nbsp;rb\_gc\_mark in C extensions, are guaranteed not to move. It would be  
> &nbsp;&nbsp;&nbsp;possible in the future to change Ruby to use a partially-collecting  
> &nbsp;&nbsp;&nbsp;garbage collector that could move objects not referenced by C extensions.)
> 
> &nbsp;&nbsp;&nbsp;Mauricio Fernandez also posted some information on navigating the source  
> &nbsp;&nbsp;&nbsp;code of the Ruby interpreter.
> 
> &nbsp;&nbsp;Serious YAML bug (comp.lang.ruby)  
> &nbsp;&nbsp;---------------------------------
> 
> &nbsp;&nbsp;&nbsp;Frantisek Fuka found a bug with YAML in Ruby 1.8.4 (2005-12-24) shipped  
> &nbsp;&nbsp;&nbsp;with his Dapper Drake Linux distribution (Ubuntu) where it is limited to  
> &nbsp;&nbsp;&nbsp;reading files \< 4096 characters long.
> 
> &nbsp;&nbsp;&nbsp;ts said the bug is fixed with more recent versions of 1.8.4 that you can  
> &nbsp;&nbsp;&nbsp;download from CVS.
> 
> &nbsp;&nbsp;&nbsp;Francis Hwang:  
> &nbsp;&nbsp;&nbsp;\> If you need to stick with YAML, I'd recommend getting Ruby from CVS  
> &nbsp;&nbsp;&nbsp;\> head; it's supposed to contain a newer version of Syck. If you can get  
> &nbsp;&nbsp;&nbsp;\> away with it, though, I'd suggest using Marshal instead. It's less cool  
> &nbsp;&nbsp;&nbsp;\> than YAML, but pretty solid otherwise.
> 
> &nbsp;&nbsp;Problems with rb\_struct\_define and 64 bit Ruby (ruby-talk)  
> &nbsp;&nbsp;----------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;ts: "Well, the architecture x86\_64 was created to make in sort that C  
> &nbsp;&nbsp;&nbsp;programmer understand the difference between (int)0 and (void \*)0".
> 
> &nbsp;&nbsp;&nbsp;Read this thread if you don't think there's a difference, or that it  
> &nbsp;&nbsp;&nbsp;doesn't matter.
> 
> &nbsp;&nbsp;Modules / Documentation / CPAN like repository?? (ruby-talk)  
> &nbsp;&nbsp;------------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Paul D. Kraus wondered if Ruby has an equivalent to perldoc - a  
> &nbsp;&nbsp;&nbsp;command-line tool for displaying documentation information on standard  
> &nbsp;&nbsp;&nbsp;libraries as well as those installed separately.
> 
> &nbsp;&nbsp;&nbsp;James Edward Gray II said that ri is the Ruby command, for example  
> &nbsp;&nbsp;&nbsp;"ri Find" or "ri String#split". He noted that documentation is not  
> &nbsp;&nbsp;&nbsp;always automatically installed for Ruby, or for third-party libraries.  
> &nbsp;&nbsp;&nbsp;"If you build from source, it's important to make install-all."
> 
> &nbsp;&nbsp;&nbsp;James Britt added that [ruby-doc.org](http://ruby-doc.org) has online documentation.
> 
> &nbsp;&nbsp;Calling a Class Method with a class name (comp.lang.ruby)  
> &nbsp;&nbsp;---------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;This thread (and a couple of others), asked how to get a class from its  
> &nbsp;&nbsp;&nbsp;name, without using eval(class\_name). The solution is  
> &nbsp;&nbsp;&nbsp;Object.const\_get(class\_name). If you're using Rails, you can also use  
> &nbsp;&nbsp;&nbsp;class\_name.constantize, which has the benefit of working even when  
> &nbsp;&nbsp;&nbsp;class\_name is e.g. "A:🅱:C".
> 
> &nbsp;&nbsp;Running Coach (#82) (ruby-talk)  
> &nbsp;&nbsp;-------------------------------
> 
> &nbsp;&nbsp;&nbsp;Ruby Quiz this week is inspired by Benjohn Barnes and his girlfriend's  
> &nbsp;&nbsp;&nbsp;jogging programme, which includes periods of jogging, brisk walking and  
> &nbsp;&nbsp;&nbsp;resting.
> 
> &nbsp;&nbsp;&nbsp;"I was explaining to a friend that it's incredibly difficult for me to  
> &nbsp;&nbsp;&nbsp;look at a stop watch and work out in my head if we're supposed to be  
> &nbsp;&nbsp;&nbsp;jogging or walking, how many more jogs we've got to do, and when I can  
> &nbsp;&nbsp;&nbsp;stop and rest. He suggested: `why not tape yourself giving prompts about  
> &nbsp;&nbsp;&nbsp;when to start and stop'."
> 
> &nbsp;&nbsp;&nbsp;Being programmers, we should of course automate this.
> 
> &nbsp;&nbsp;&nbsp;\> Write a program to create the tracks for each of the eight weeks. Make  
> &nbsp;&nbsp;&nbsp;\> it give helpful and enthusiastic advice like "you've got to run for  
> &nbsp;&nbsp;&nbsp;\> another minute / 30 seconds / 15 seconds ...", "walk now for two  
> &nbsp;&nbsp;&nbsp;\> minutes, you've got three jogs left", "you're on jog 2 of 6", or "well  
> &nbsp;&nbsp;&nbsp;\> done, that's your last jog. Don't forget to cool down and stretch!"
> 
> &nbsp;&nbsp;ruby qt (comp.lang.ruby)  
> &nbsp;&nbsp;------------------------
> 
> &nbsp;&nbsp;&nbsp;Shea Martin asked if the Ruby/Qt bindings were unmaintained, since all the  
> &nbsp;&nbsp;&nbsp;references to it are old, but Richard J. Dale explained that it's due to a  
> &nbsp;&nbsp;&nbsp;name change:
> 
> &nbsp;&nbsp;&nbsp;Bindings for Qt1 and Qt2 were called Ruby/Qt, while the Qt3 and Qt4  
> &nbsp;&nbsp;&nbsp;bindings are called QtRuby. The latter is complete and is actively  
> &nbsp;&nbsp;&nbsp;maintained.
> 
> &nbsp;&nbsp;&nbsp;The book "Rapid GUI Development with QtRuby" was also recommended.
> 
> &nbsp;&nbsp;Which encoding causes fewest problems in Ruby 1.8.2? (ruby-talk)  
> &nbsp;&nbsp;----------------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Jim Smith wants to know which string encoding works best with Ruby: UTF-8,  
> &nbsp;&nbsp;&nbsp;ISO-8859-1 or ISO-8859-15.
> 
> &nbsp;&nbsp;&nbsp;Matz said that all of them will work (with the appropriate -K? option to  
> &nbsp;&nbsp;&nbsp;set the encoding), but for UTF-8 you need to be careful when doing  
> &nbsp;&nbsp;&nbsp;character-based operations (length and indexing) - use Regexp-based  
> &nbsp;&nbsp;&nbsp;methods for these instead of String ones.
> 
> &nbsp;&nbsp;&nbsp;(For example, String#length may return an incorrect value for a UTF-8  
> &nbsp;&nbsp;&nbsp;string, but mystring.split(/./).length returns the correct value. Future  
> &nbsp;&nbsp;&nbsp;Ruby releases will have better support for different encodings.)
> 
> &nbsp;&nbsp;Set minimum for random number generator (comp.lang.ruby)  
> &nbsp;&nbsp;--------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;In Python, random.randint(min, max) returns a random integer between min  
> &nbsp;&nbsp;&nbsp;and max (inclusive). OrganicFreeStyle asked what the Ruby equivalent to  
> &nbsp;&nbsp;&nbsp;this is. "Rand allows you to set the max, but what about the min?"
> 
> &nbsp;&nbsp;&nbsp;ZeD replied: min + rand(max-min)
> 
> &nbsp;&nbsp;&nbsp;That's the exclusive form; you need min + rand(max-min+1) to include the  
> &nbsp;&nbsp;&nbsp;maximum in the possible range of values.
> 
> &nbsp;&nbsp;&nbsp;There have been discussions in the past about having a Range#rand method  
> &nbsp;&nbsp;&nbsp;that would behave sensibly for a range of integers. For example,  
> &nbsp;&nbsp;&nbsp;(1..10).rand would return an integer between 1 and 10 inclusve if the  
> &nbsp;&nbsp;&nbsp;following code was defined:
> 
> class Range  
> &nbsp;&nbsp;&nbsp;def rand  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;min + Kernel.rand(max-min+1)  
> &nbsp;&nbsp;&nbsp;end  
> end
> 
> &nbsp;&nbsp;&nbsp;(The above naïve version won't work with ranges of anything other than  
> &nbsp;&nbsp;&nbsp;integers - not even floats.)
> 
> &nbsp;&nbsp;Creating a future date (comp.lang.ruby)  
> &nbsp;&nbsp;---------------------------------------
> 
> &nbsp;&nbsp;&nbsp;How to create the date for a year from today? JimC posted a nice solution  
> &nbsp;&nbsp;&nbsp;that uses Date#\>\>(n), a method that shifts a date by n months.
> 
> require 'date'  
> next\_year = Date.today \>\> 12
> 
> &nbsp;&nbsp;Getting a list of Processes (ruby-talk)  
> &nbsp;&nbsp;---------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Ways of getting a list of running processes in Windows: use the  
> &nbsp;&nbsp;&nbsp;`sys/proctable' library, or use win32ole and `winmgmts'.
> 
> &nbsp;&nbsp;&nbsp;An example of the latter was given by Brad:
> 
> def get\_process\_info()  
> &nbsp;&nbsp;&nbsp;procs = WIN32OLE.connect("winmgmts:\\\\.")  
> &nbsp;&nbsp;&nbsp;procs.InstancesOf("win32\_process").each do |p|  
> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;puts p.name.to\_s.downcase  
> &nbsp;&nbsp;&nbsp;end  
> end
> 
> &nbsp;&nbsp;I love Ruby - But how bright is Ruby's Future? (ruby-talk)  
> &nbsp;&nbsp;----------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;A poster asks about Ruby's future - are lots of people using it, will it  
> &nbsp;&nbsp;&nbsp;be maintained, has he made the right decision in starting to learn Ruby.
> 
> &nbsp;&nbsp;&nbsp;Jim Freeze said "stay tuned, you have made a good choice", and asked if  
> &nbsp;&nbsp;&nbsp;he'd noticed all the press lately (particularly around Rails). Linux  
> &nbsp;&nbsp;&nbsp;Journal (cover), Dr. Dobbs (cover), Business Week, Wired, Newsweek, and so  
> &nbsp;&nbsp;&nbsp;on.
> 
> &nbsp;&nbsp;&nbsp;"The press hype may die down, but the infrastructure that is being built  
> &nbsp;&nbsp;&nbsp;(and the fact that people like the language) pretty much secure Ruby's  
> &nbsp;&nbsp;&nbsp;existence for some time to come."
> 
> &nbsp;&nbsp;&nbsp;Many posts in the thread say that it's worthwhile learning new languages,  
> &nbsp;&nbsp;&nbsp;regardless of their popularity, and others discuss the corporate  
> &nbsp;&nbsp;&nbsp;acceptance (and acceptability) of Ruby compared with Python. (Coverage in  
> &nbsp;&nbsp;&nbsp;business magazines certainly helps.)
> 
> &nbsp;&nbsp;reading data from excel (ruby-talk)  
> &nbsp;&nbsp;-----------------------------------
> 
> &nbsp;&nbsp;&nbsp;Parvinder Ghotra wants to read data from Excel files using Ruby, and was  
> &nbsp;&nbsp;&nbsp;recommended to either use Win32OLE to get Excel itself to read the files,  
> &nbsp;&nbsp;&nbsp;or to use the `parseexcel' library.
> 
> &nbsp;&nbsp;EURUKO 2006 (ruby-talk)  
> &nbsp;&nbsp;-----------------------
> 
> &nbsp;&nbsp;&nbsp;Stephan Kämper pondered the location for EURUKO 2006 (European Ruby  
> &nbsp;&nbsp;&nbsp;Conference), saying he would quite like Hamburg or Copenhagen, but  
> &nbsp;&nbsp;&nbsp;wouldn't mind Munich again.
> 
> &nbsp;&nbsp;&nbsp;Others "voted" on these locations, with Tim Becker adding Köln (Cologne)  
> &nbsp;&nbsp;&nbsp;as an option, and Laurent Sansonetti suggesting Brussels, noting that  
> &nbsp;&nbsp;&nbsp;FOSDEM (Free and Open Source Developers European Meeting) is successfully  
> &nbsp;&nbsp;&nbsp;held there every year.
> 
> &nbsp;&nbsp;&nbsp;Robert Klemme said the location will ultimately be up to whichever person  
> &nbsp;&nbsp;&nbsp;or organisation makes the effort to organise it.
> 
> # New Releases
> 
> Releases this week include ...
> 
> &nbsp;&nbsp;DRMAA for Ruby (ruby-talk)  
> &nbsp;&nbsp;--------------------------
> 
> &nbsp;&nbsp;&nbsp;"DRMAA is an interface standard for job submission and control to  
> &nbsp;&nbsp;&nbsp;Distributed Resource Manager (DRM) systems such as Condor,  
> &nbsp;&nbsp;&nbsp;PBS/OpenPBS/Torque, Platform LSF, Sun's N1 Grid Engine and others."
> 
> &nbsp;&nbsp;&nbsp;Andreas Haas created DRMAA bindings for Ruby (with Ruby/DL).
> 
> &nbsp;&nbsp;Ruby-GetText-Package-1.6.0 (ruby-talk)  
> &nbsp;&nbsp;--------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Masao Mutoh released a new Ruby-GetText-Package (a localisation library  
> &nbsp;&nbsp;&nbsp;and tool similar to GNU gettext - it assists developers in creating  
> &nbsp;&nbsp;&nbsp;programs that can output messages in different languages).
> 
> &nbsp;&nbsp;&nbsp;Chinese (zh\_CN) and Italian (it) translations are now included, and the  
> &nbsp;&nbsp;&nbsp;management of textdomains is improved.
> 
> &nbsp;&nbsp;RbYAML 0.2: Pure Ruby YAML 1.1 parser and emitter (ruby-talk)  
> &nbsp;&nbsp;-------------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;The pure-Ruby YAML parser and emitter, RbYAML, had a major release, with  
> &nbsp;&nbsp;&nbsp;performance improvements, better test cases, bugfixes and more.
> 
> &nbsp;&nbsp;Mongrel Pre-Release 0.3.13 -- Katana Suicide Concurrency (ruby-talk)  
> &nbsp;&nbsp;--------------------------------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;A Mongrel pre-release (Mongrel is a web server for Rails, Camping and  
> &nbsp;&nbsp;&nbsp;Nitro web applications) was announced by Zed Shaw. The official 0.3.13  
> &nbsp;&nbsp;&nbsp;will be out soon, and will be followed by 0.4 (aka "Enterprisey Edition  
> &nbsp;&nbsp;&nbsp;1.2").
> 
> &nbsp;&nbsp;Ruby Reports 0.4.5 (ruby-talk)  
> &nbsp;&nbsp;------------------------------
> 
> &nbsp;&nbsp;&nbsp;Ruby Reports 0.4.4 and 0.4.5, the "Hey, I've got a brand new formatting  
> &nbsp;&nbsp;&nbsp;engine" and "Arbitrary Additions to DataSet" editions were released by  
> &nbsp;&nbsp;&nbsp;Gregory Brown.
> 
> &nbsp;&nbsp;&nbsp;Ruby Reports is a 2006 Google Summer of Code project, and Gregory is  
> &nbsp;&nbsp;&nbsp;aiming to have a release every 10 days for the rest of (Northern  
> &nbsp;&nbsp;&nbsp;Hemisphere) summer.
> 
> &nbsp;&nbsp;&nbsp;Acknowledgements: "Dudley Flanders for his excellent work on DataSets in  
> &nbsp;&nbsp;&nbsp;this release."
> 
> &nbsp;&nbsp;Madeleine 0.7.3 (ruby-talk)  
> &nbsp;&nbsp;---------------------------
> 
> &nbsp;&nbsp;&nbsp;"Madeleine is a Ruby implementation of Object Prevalence: Transparent  
> &nbsp;&nbsp;&nbsp;persistence of business objects using command logging and snapshots."
> 
> &nbsp;&nbsp;&nbsp;This release fixes a win32 bug introduced in the last. "The other news is  
> &nbsp;&nbsp;&nbsp;that we're moving to Rich Kilmer's basement, i.e. RubyForge, so update  
> &nbsp;&nbsp;&nbsp;your links!"
> 
> &nbsp;&nbsp;rubyrss-1.0 (ruby-talk)  
> &nbsp;&nbsp;-----------------------
> 
> &nbsp;&nbsp;&nbsp;RubyRSS is a new tool for parsing and generating RSS feeds. Kouhei Sutou  
> &nbsp;&nbsp;&nbsp;noted that it is similar to RSS Maker, which is part of Ruby's standard  
> &nbsp;&nbsp;&nbsp;library.
> 
> &nbsp;&nbsp;FreeRIDE 0.9.6 - The Free Ruby IDE (ruby-talk)  
> &nbsp;&nbsp;----------------------------------------------
> 
> &nbsp;&nbsp;&nbsp;Laurent Julliard announced a bugfix release of FreeRIDE.
> 
> &nbsp;&nbsp;&nbsp;"FreeRIDE aims to be a full-featured, first-class IDE on a par with those  
> &nbsp;&nbsp;&nbsp;available for other languages, with all the best-of-breed features that  
> &nbsp;&nbsp;&nbsp;you would expect in a high-end IDE."
> 
> &nbsp;&nbsp;ruby-prof for windows (ruby-talk)  
> &nbsp;&nbsp;---------------------------------
> 
> &nbsp;&nbsp;&nbsp;Charlie Savage ported Shugo Maeda's fast Ruby profiler "ruby-prof" to  
> &nbsp;&nbsp;&nbsp;Windows, including a binary for easy installation.
> 
> &nbsp;&nbsp;&nbsp;"ruby-prof is a code profiler like the built-in ruby profiler but is  
> &nbsp;&nbsp;&nbsp;orders of magnitude faster."
> 
> &nbsp;&nbsp;slave-0.0.1 (ruby-talk)  
> &nbsp;&nbsp;-----------------------
> 
> &nbsp;&nbsp;&nbsp;Ara.T.Howard issued the first slave release, a library for creating  
> &nbsp;&nbsp;&nbsp;master-child processes that communicate via DRb, with heartbeat  
> &nbsp;&nbsp;&nbsp;monitoring.
> 
> &nbsp;&nbsp;&nbsp;"The purpose of Slaves is to be able to easily set up a collection of  
> &nbsp;&nbsp;&nbsp;objects communicating via drb protocols instead of having to use IPC."
> 
> &nbsp;&nbsp;rq-2.3.3 (ruby-talk)  
> &nbsp;&nbsp;--------------------
> 
> &nbsp;&nbsp;&nbsp;"ruby queue (rq) is a zero-admin zero-configuration tool used to create  
> &nbsp;&nbsp;&nbsp;instant unix clusters."
> 
> &nbsp;&nbsp;&nbsp;Ara.T.Howard fixed some bugs, added the ability to dump  
> &nbsp;&nbsp;&nbsp;stdin/stderr/stdout and made `rotate' more robust.

---

<div class="post-metadata">

**Author:** ![Florian\_Gross2](https://avatars.discourse-cdn.com/v4/letter/f/b4bc9f/32.png) [@Florian\_Gross2](https://rubytalk.org/u/Florian_Gross2)\
**Post date:** [14 June 2006 14:16 UTC](https://rubytalk.org/t/ruby-weekly-news-5th-11th-june-2006/28201/2 "2006-06-14T14:16:04Z")

</div>

Tim Sutherland wrote:

> &nbsp;&nbsp;&nbsp;The original debug/fix/continue idea is a good one though. It would be  
> &nbsp;&nbsp;&nbsp;worth looking into the various Ruby debuggers out there. Perhaps the  
> &nbsp;&nbsp;&nbsp;ruby-talk readers have some ideas? (It's worth raising awareness that a  
> &nbsp;&nbsp;&nbsp;lot of programmers are coming to Ruby through Watir.)

I'd really like to do this with ruby-breakpoint. Having it automatically issue a breakpoint when a Test::Unit test fails. If things go as planned you will then be able to use the GUI client to fix things. I'm just not yet sure if editing the code (and saving it back to the disk) and continuing executing the code just after the breakpoint will be possible. I think it might be possible with redo and a trace\_func that skips over code, but I'm not yet 100% sure.

> **···**
>
> --  
> [http://flgr.0x42.net/](http://flgr.0x42.net/)
