Nick commented out here that Ryan Tomayko post is correct.

Well, kind of, sorta… not entirely correct.

Mandatory warning

Take the following statements with a grain of objectivity without pondering how much bits and bytes a packaging solution like RubyGems adds to your scenario.

Now, let me present this with a story (I love stories)

Creating my application or library

I had an excellent idea for a new library we needed here at the office. This library will provide a client-side OOP access to a remote webservice we use daily.

After investigation of my remote webservice API and structure and exploration of existing HTTP/REST like interfaces, I decided that would be best to invest my time in leverage the low-level HTTP complexity to an existing library published by another developer.

Filling the gaps and start working

Started creating my library (called Foo) and based the internals calls on what was published for Bar at that time (let’s say was version 1)

So far, everything is great. Everything works great on my computer.

Time to share

It has been 2 weeks since I created this library and we internally wanted to use in another project.

This time, my colleague check out my library repository and try to use the library.

He realizes that Bar is required and is not present in his environment, which leads him to start a gem installation process.

He succeed installing Bar version 2, as is the latest available version by the developer of it, but my library no longer works with it, since there are several API changes.

Spotting that issue can be simple, but now think the following: I’ve used Rip instead of Rubygems, and Rip uses repositories directly, not packaged versions.

So now, Foo is not only version specific, but commit specific since the developer of Bar could have changed API between commits to finalize or stabilize it.

Even worse could be that I used setup.rb that was part of Bar and installed inside my Ruby installation, which then will make complicated to figure out which version I used to implement Foo

But what if…

• Usage of Git submodules for my dependencies… if I’m using Git at all.
• Gemify my library, which then will lead to negative comments from others about me imposing a packaging system on them.
• Do not release or share my library at all. Problem solved.

Add to that, that even if I released my library as a gem or not, there is no way to determine the usage of my Foo from Joe’s Foo variation. Something I raised about GitHub gem forks and commented out here

So, try all the approaches, get your own conclusions, but don’t say that the way I implement something is flawed. That is plain subjective.

Yeah, Windows doesn’t do ANSI coloring like any POSIX terminal, and it has it’s own API with extended control of the console functionality.

UPDATE: Please see this post for the beta1 release of new win32console.

So, I took over the project long ago, since required a couple of fixed. You could find the code here at GitHub

I must say that my backyard and house is getting crowded. Lot of small Open Source projects that require attention and some of those that require some special care, like this case.

What do I mean by this?

win32console do more than just fancy colors, it also translates the current console page back and forth using some clever but not portable character extrapolation…

Now, Ruby 1.9 doesn’t like you do that, not at all.

So wonder, what if I just get rid of that? How much of win32console features are being used beyond the ANSI coloring?

I just verified that tweak work with cucumber in 1.8 and 1.9, but dunno other user cases.

Want to know all that, since I’m creating a much simpler solution that uses FFI and will provide colors for both MRI and JRuby, no extension compilation and will deprecate the need of win32console.

What do you think? Please let me know your thoughts in the comments.

Thank you

To give some background, for the fresh ones just arriving to the blog. rake-compiler provides, besides C/Java extension standardization structure, the ability to cross-compile versions of Ruby to allow developers on Linux/OSX build and release gems for users running Windows.

Seems that latest patchlevels of Ruby 1.8.7 (172) and 1.9.1 (376) cannot be cross compiled.

I’ve reported that to Ruby’s bug tracker under issues #2464 and #2484

From what you can read there, there is a clear change in the rules about what is the minimum Ruby version that can be used to cross compile (known as BASERUBY)

Now, being pointed to the Developer’s Howto page, and checking the history and time, it is clear the note about the limitation is really fresh

The funny thing is that I still can cross compile 1.9.2dev using 1.8.6! Which is contrary to what the recently edited howto states!

Don’t get me wrong, I have no personal issue besides being annoyed, and that will require put some lock in cross-ruby compilation to avoid bug reports about something that is beyond my control.

So, with the goal of making this public and have a link to include in my canned response to people reporting bugs about it, the following is the list of versions and patchlevels of Ruby that I know works and aim to support in relation to bug fixes of rake-compiler itself:

Tricky list, even switching the version of Ruby to match the one desired to be installed, it doesn’t work (reported as #2485, I’m very persistent).

So, apologies to anyone having issues with this: I’m having the same ones here.

Update: Added 1.8.7-p299 to the list of no compatible

Update 2: Added 1.8.6-p398 to the list of compatibles and 1.8.7 still is not cross-compilable, sorry. 1.9.1-p378 is not cross-compilable either.

I have so cool news to everybody:

MySQL/Ruby support

As I announced 2 months ago, new releases of MySQL gem are going to be build from my repository at GitHub

Please note: only report bugs related to the gem itself, more likely the binary issues. Other type of bugs will be rejected, since this gem is a wrapper around MySQL/Ruby code from Tomita Mashahiro (tmtm)

SQLite3/Ruby bindings

I took the role of maintainer of SQLite3/Ruby project, but I really lack the skills to dig into all the inner details of it.

So, I’m opening the door to anyone with new features, enhancements or bug fixes to fork my clone of the project (here), do your changes and send a pull request.

Both RubyForge Tracker and GitHub pull requests will be honored.

Pure-Ruby Readline (rb-readline)

Released version 0.2.0 a couple of weeks ago, but still is far from perfect.

We are still missing IO redirect and be able to run test_readline.rb test directly.

Roger Pack worked on that, but he also needs some help

Please, fork the project, do your changes and send your pull requests!

Windows Console (win32console) new release soon.

A few patches to improve 1.9 compatibility are under way.

rake-compiler get more audience!

That’s right folks!, rake-compiler now is being used to drive the compile and release process of native gems for DataMapper, FFI and even Thin !

Not too bad for 10 months old tool, ah?

Thanks to all the developers involved in these projects for caring about their products and adding the compatibility for Windows.

More projects listed here … Keep them coming!

Ruby Installer for Windows

Preview and DevKit releases has been moved to RubyForge

While our preview concept is quite stable, we are working on getting it better

Got my certificates to deliver signed installers, how cool is that?

Still alive an kicking

If you think I’m quiet, please join us at RubyInstaller Google Group, will find out that I talk more there than here

This version is mainly a bug fixing release:

• Accept ‘cygwin’ as valid terminal [rogerdpack]
• Accept ‘dumb’ as valid terminal. Closes GH-7
• Fix crash when performing reverse look search. Closes GH-6

Install instructions

1. Download rb-readline-0.2.0.zip from GitHub downloads
2. Extract to a folder and open a command prompt there (or cd into it)
3. Execute ruby setup.rb

That should update the installed rb-readline. Just to verify, execute the following in the command prompt:

ruby -rreadline -e "puts Readline::RB_READLINE_VERSION"

The output should say 0.2.0

More to come

Stay tuned for upcoming announcements, we are cooking RC for RubyInstaller! And at the same time, updated website, more documentation at the Wiki and of course, tutorials full of success!

Please join us at RubyInstaller Google Group and share your questions or issues!

February 24, 2009 is the date that Jamis Buck marked as the end of several of his open source projects, including SQLite3/Ruby. (you can read the post here)

Previously, he asked for help updating SQLite3/Ruby to make it work on Windows.

That sad news left a lot of us with a bad taste, and very unhappy, not because we no longer will have someone to complain at, but because he no longer enjoyed working on those projects.

Over the past months I’ve been improving rake-compiler to be able to catch most of the building issues of several projects, including my own (I love to scratch my own itch).

As you have noticed on my “getting started with Rails and SQLite3″: http://blog.mmediasys.com/2009/07/06/getting-started-with-rails-and-sqlite3/ post, I successfully built, installed and used a SQLite3/Ruby gem on Windows.

The next question was, what to do? The work to get all those lovely gems was there, initial 1.9 was there, but was not official

While SQLite3/Ruby being the de-facto for getting started with Rails, we couldn’t let it die.

Enough words, sent some emails to Jamis and now I can publish those gems to RubyForge.

What all that babbling means?

This means:

• My fork at GitHub is the new mainstream for the releases
• I’m going to go over the open bugs and tickets and asses validity and relevancy based on work that was already done in my fork.
• The release cycle has been improved and almost automated. It can be performed from Windows, Linux or OSX, even using latter to create Windows native gems.

Now, what happens with new features:

• Pull requests with patches and bug fixes are going to be accepted.
• New features will be evaluated as long they don’t impose structural changes and carry with them tests cases.
• Patches that improve Ruby 1.9 compatibility are highly appreciated.

I don’t have strong knowledge of all the internals of this tool, so don’t expect earth breaking changes from me, except ensuring stability.

I hope this is good news to everybody. Now I’m going to stalk MySQL binding author and get permission to push those lovely gems

Cheers everybody!

You can review all the designs submitted here, and also read the comments from each designer.

In no special order:

• Thuva Tharma
• Ben Alpert
• Silviu Postavaru
• Francesco Agnoletto
• Pavel Macek

Thanks again the designers who submitted entries!

How to vote?

Good question! You can click this link, share http://poll.fm/111u6 on Twitter or use the embedded poll widget below! Good luck, please vote (and do not cheat!)

Updated: Voting ends July 23, 2009, 11:59pm GMT-3

Updated: Voting has officially ended

Please go here to read the results.

Check those over there:

http://wiki.github.com/oneclick/rubyinstaller/website-contest

This is excellent news. Thank you guys for participating!

Deadline

I’m aiming to close submissions June 19, 2009. So if you have a design or idea, mock it up and add to the wiki page, point to a blog post or something and let the public know you’re participating!

UPDATE: Deadline extended until Monday, June 22, 2009 at 8pm Argentine Timezone (GMT-3)

Logo

As Pavel mentioned over the wiki page:

I like it, but I was not sure about the shape of ruby gem.

This is something weird since Ruby users are used to see Ruby logo with the shape of a diamond, which is incorrect.

Let me explain it better: the mineral ruby can be cut in diamond shape, but is really uncommon.

More common shapes are facetted balls or emerald cuts. Take a look here for more examples.

Almost every package, tool or library associated with Ruby tends to use a diamond cut ruby as logo.

When analyzed this detail with Rodolfo Budeguer from Estudio Domo, we found that sticking to that shape would not only fall into the repetition (and be a copycat), but also move away from the goal of it.

If you pay close attention, will noticed that the emerald cut of RubyInstaller shows the 4 colors that are also present in… the Windows logo.

Why is that? Well, because is Ruby Installer for Windows

Voting and Deliverables

Once again, would like to mention public voting would define the selected design. No new design will be accepted after the deadline above mentioned. So make room to do the mockups soon!

Also, once we have the final votes for the designs. Selected designer should provide generated versions of mockups in CSS and HTML content, preferable in a public Git repository.

I’m so excited about this that I’m blogging at 4am!

Thanks again to Silviu, Francesco and Pavel to contribute and participate. Also other people that privately sent some feedback about RubyInstaller itself, not just the website

Most of the details are in the Pledgie here , but just to clarify:

Designers that want to participate, need to create a homepage image mockup of the website. These mockups are going to be exposed to everybody and a open poll is going to be held.

The mockup with most votes and that share the idea of the RubyInstaller project and guidelines is going to be awarded with the total money of the pledgie.

Money is going to be send over Paypal when HTML and CSS elements are shared publicly (GitHub or any type of code repository).

Designers will remain the copyright holders of the design and credits will be displayed in the website with permission of usage for RubyInstaller project.

I believe this is fair, and also can let people’s mind fly, so go ahead, mock something that can fit these needs and add those to the following wiki page at GitHub:

Website Contest – RubyInstaller

Once again, if you want to contribute with some money, please use the badge at the bottom of the page or this one:

And for those Linux and Mac users that want to contribute, you can do it anonymously

Good luck everybody!

So, I’m bringing this topic to the table, seeking for ideas on how to improve the Ruby on Windows image beyond just code.

I’m willing to offer money to pay for it, since I’m quite aware that Open Source and contributions don’t put food on the table.

Of course, I’m not rich, so the balance needs to be found

So, what is the idea:

• Build the website with Radiant or a simple CMS over Ruby
• Design needs to be simple and provide access to:
1. News feed (small articles)
2. Download info and links
3. Getting Started Resources (info and links)
4. Contribute (RubyForge and GitHub info)
5. Support (access to mailing list and forums of interest)
• Initial artwork (logo and iso) already exist
• Application icons can be highly improved

For the record: there is no need to be a Windows user or designer, so people on Linux and OSX are welcome

If there is more interest, maybe a Bounty can be opened, but time, feedback and community response will tell.

Please, comment and pass the message!