You can find Kohana RPM for the current stable release 3.0.5 here, and the source RPM is available here in case you want to rebuild it yourself (and you might, details follow). New releases to correspond with new releases from Kohana will be updated there as needed.

This package is built on a CentOS 5.4 machine, with pretty much default settings.

Please note that Kohana is not tested with the PHP version 5.1.6 which is what RHEL/CentOS 5 uses by default, so there could be some problems with using it on such a default installation of the operating system. There was one place where Kohana uses a function in a way that is not supported on PHP 5.1.6, and I had to patch that with a workaround.

This package includes additional patches that I felt where necessary, some to work around Kohana issues and some to add functionality that I felt is missing or or lacking. If there is interest I can expand on the work there. I do intend to submit all patches upstream, of course – once I’m more familiar with the community process of Kohana.

In the mean time, you can try it out and let me know if it works for you or not.

Related articles by Zemanta

• Top 5 Free Linux Distributions for Servers in 2010 (taragana.com)

Yep, that sounds about right.

The Linux Action Show review is useful, and good, but its not really fair – Chris and Bryan ranted on a lot of things that do not work well for Fedora, such as not a lot of applications pre-installed and some new and immature applications being introduced, Flash being hard to install, codecs missing and obscure instructions on Fedora’s wiki pages on how to address these issues, and more.

The thing is, is that all those comments are fair when looking at an operating system that is geared towards the general public – like Ubuntu – but Bryan and Chris themselves mentioned that Fedora is not aimed at that crowd but is meant for power users and developers (the debate about what is the target audience for Fedora is raging – I think the best description I heard so far, is from Máirín Duffy – heading Fedora’s design team – where in an interview she said “Fedora is aimed at people who want to work on Fedora).

When taken into the context that Fedora is an operating system for developers and power users, and specifically those that want to work on Fedora, then most of these problems are not important – its like harping on MS-Windows that it doesn’t come with a lot of software out of the box – its not supposed to.

Fedora people are very cautious with everything that is proprietary software and/or patent encumbered – hence the very ambiguous “You can find many interesting things using a search engine like Google” quote Chris mentioned: Fedora does not want to get involved in anything remotely DMCA applicable, and as much as I don’t like it – them being based in the USA and sponsored and hosted by a commercial company, I guess it is understandable.

I would love it if Fedora will put up detailed “follow these easy steps to enable patent-infringing and possibly copyright-violating codecs on your new Fedora system”, but I doubt they can get away with it and there is no point ranting. Chris and Bryan, if you’re listening – I’m not saying its all your fault, but that’s what you get for living in that crazy country you call home .

My Fedora Experience

One of the things that works mostly well for me in Fedora 13 is KDE: I really like the KDE implementation in Fedora 13 (based on KDE SC 4.4.3) – it is very very pretty. While the 3D desktop effects are not as polished as Compiz on the GNOME side, they are still very nice (though I had to tweak the wobbly windows effect quite a bit and it still not exactly what I want, but hey – at least I can tweak it!) and the Plasma based desktop itself is extremely nice. I’m a very big fan of the “desktop activity” model, and I really dig the default activity which is a widget board with a single “folder containment” widget showing the desktop icons at the top left corner. I easily added a battery monitor, trash can, quick launcher (which I rotated 90° and put vertically on the right side), a system monitor and a weather applet.

The folder containment has a very useful effect where you can hover on a folder and then the folder will open in a sub-widget and will let you browse the things inside the folder – into whatever depth you want:

Here is a full picture of my current desktop, without any application running – because my laptop has a small 1024×768 resolution, I ran applications on the entire screen without any panels: the bottom panels is set to “windows can cover” so that it is visible only if there is no full size window (though you can bump the mouse cursor on the bottom edge to get it up) and the small top panel I use for status notifications is auto-hiding.

(Also, please don’t mind the background – one of the setting for the desktop activity is to choose a slideshow for the background, so I have several hundreds wallpapers and the desktop keeps changing every 15 minutes).  As usual, a static picture doesn’t do it justice – you have to see how it behaves: all the UI elements have eye-catching transition effects such as sliding and fade, notifications are transparent bubbles that pop up form the bottom and stack on top of each other, windows have transparency effects, menus slide up and down.

What is not to like about the KDE desktop – the integration is not up to par with the current GNOME setup – for example, I use bluetooth a lot and it simply doesn’t work well or at all in KDE: you can’t right click a file and send it to bluetooth and bluetooth pairing is problematic at best and impossible at worse. Don’t even talk to me about dial-up. The bluetooth software isn’t even running unless you run it manually. Another thing which I also like a lot in GNOME is that you can get into any part of the system through GConf or DBus, while in KDE 4.4 – even though they pioneered the message bus with their DCOP bus which was the basis for FreeDesktop.org’s DBus – currently not many services are exposed to that kind of manipulation, specifically Plasma stuff (which is mostly what you have on a KDE desktop today) is very hard to control programatically.

The lack of an SSH agent for KDE is also bothering me that makes using pass-phrase protected SSH keys as annoying as using passwords directly. When you try to access a remote file system over SSH you get a dialog that looks like a password entry dialog – including a username field – but that asks you for your SSH key pass-phrase. This is really easy to miss a few times until you figure out what to type in, and when you ask it to “save it in the wallet”, it will still prompt you for that next time. SMB access has even more problems. As I said – integration is not great, which is why I still pop into GNOME from time to time, though I’m using KDE now more then I ever used it since version 3.5 – I still can’t say its my primary desktop, but its getting really close.

Now about GNOME in Fedora 13: its OK. There’s nothing new and exciting there, and the GNOME experience in Fedora 13 is somewhat worse then what you get in even Ubuntu 9.10, but its useful and works well and they finally got rid of that annoying Nodoka theme – the replacement is the old Clearlooks style which is easy on the eyes, but as the Linux Action Show guys noted – it looks old.

My main gripe with Fedora 13 is that it keeps crashing! The operating system itself is rock solid, and they added a service that monitors for crashes and allows you to easily report them to the developers as bugs (somewhat like Ubuntu’s apport, but works better and doesn’t require you to use a browser to actually submit the bug, although it has the horrible name “abrt”), but I would have like it a lot better if applications wouldn’t keep crashing all the time – to the point that with most crashes I don’t bother reporting on them.

I’ve ranted on Evolution crashing in a previous post, but that is nothing new – with Fedora 13 everything crashes, I even had Nautilus (the GNOME file manager) crash on me multiple times! When has Nautilus ever crashed? – it was probably the most stable GUI application ever created. It is very important that they extended the maximum size for crash reports, because I sure as hell need it.

Interestingly enough – and back to the Linux Action Show review commentary – I don’t miss Mono at all. While on principal I don’t approve with Fedora’s move from Mono (regardless on which basis it was done), I wasn’t horrified by Fedora 13 dropping Mono completely from the default installed. I was expecting that I’d go and install Mono at one point or another, but it hasn’t happened yet. I’ll probably install it again at one point when I go back to playing with developing in Mono, but at the mean time I don’t need it.

Update:

Technology-wise its fine and has many useful and interesting features, but from the stability point of view it is computing hell – I’ve used many alpha versions that were more stable then Fedora 13 GA release. Heck, the beta I’ve been using since December was more stable then the GA release! And it seems to be getting worse, not better – the last update to Evolution (2.30.2) completely destroyed my ability to manage my calendar: ever since the update, whenever I receive a calendar invitation or try to create a new calendar entry, the Evolution data-server process crashes and the calendar access fails.

I’m currently running Fedora 13 on two different machines (a desktop and a laptop), but I’m going to move both to Ubuntu 10.10 Alpha 2 that was released recently – I’m sure it will be more stable .

Related articles by Zemanta

• The Joy of Betas: Fedora 13 Beta Released Today (ostatic.com)


• What Is The best Linux For Beginners? KGB Says It Is Ubuntu or Fedora (techie-buzz.com)


• Fedora 13 – Ubuntu’s smart but less attractive cousin (go.theregister.com)


• The Microsoft/Novell Mono Lobby (techrights.org)


• Bryan Quigley: So.. you want to try out the Alpha version of Fedora on a usb stick? (gquigs.blogspot.com)

I mean – if you are a simple e-mail user: have one account with which you send a few emails, receive a few e-mails and sometimes forward something to your list of friends – then you have several good options including a few web-mail systems.

If, on the other hand, you are more serious in your communication requirements and you need to:

• Work with multiple e-mail accounts and manage them separately but with the option of moving stuff between accounts.


• Connect with MS-Exchange (still most companies in Israel use MS-Exchange for their groupware backend)


• Work with multiple mailing lists with different policies and different internal filing requirements


• Keep all your past communications for reference


• Inter-operate with multiple shared calendaring systems, and specifically with other people’s calendars in a heterogeneous environment (some people do not believe there are e-mail clients other then MS-Outlook)


• Do all this on multiple computers so that e-mail, address books, calendars and what-not are transparently available on all computers


• Use Linux as your computing platform of choice.

Currently I use Evolution as my main PIM solution, and while on paper it has everything I need, in practice it gives me more frustration then any other computing system I used ever.

For first thing – if you want to work with MS-Exchange as a server (and in my company I’m forced to do so), then its unstable as nothing you’ve ever seen out of beta in the last 10 years. The old OWA method requires a separate process that tends to crash a lot, leaving you without access to your e-mail, address book and calendar until your do the “force-shutdown” rain dance, and even if it doesn’t crash it still manages to give you grief with the “Formatting Message” bug. The new MAPI method, while understandably immature and a pre-release preview is crashy to the point of absurdity – sometimes you can’t even get into the main window before it crashes. The Fedora 13 bug list for Evolution crashes gets longer by the day (I do feel sorry for Matthew Barnes that has to deal with all this) and I do not believe these are very complex issues: its a segmentation fault, for crying out loud. I can’t believe that at this point – the year 2010 – we still have software that segfaults with invalid memory access on a daily basis! Don’t people check their pointers before dereferencing them?

And don’t start with me about shared calendars! The weird thing is that if you want your Evolution client to use a shared calendar system, you have the option of MS-Exchange support, and.. well, that’s it.   Its not like there aren’t other shared calendar systems in the world (for example – Zimbra and Zarafa to name a couple. Interesting how both start with “Z”), is it just that if you aren’t a corporate Exchange user then you aren’t really someone interesting to Evolution developers ?!?.

But often this is not really what you want – an administrator might want to see what happens in the last X minutes and the common practice to do this is to run tail with a guessed number of lines, see if you get what you want and if its not enough increase the number and try again.

Here’s another approach that works well if the log file you want to trace has time stamps for its lines – use UNIX date command to format the earliest time stamp you want to see and then filter on that.

For example, to list the last 10 minutes of logs from an Apache log file, one can use this command:

Here we take advantage of the fact that date is very user friendly about interpreting date definitions to format an easy to understand relative time stamp – You can use almost any plain English time definition to tell date what time you need. After formatting a time stamp according to your log convention ( ”%d/%b/%Y:%H:%M:%S” is for Apache’s access log time stamp. Syslog logs or other logs will need different format specifications), we use grep to find a line that matches and show as many lines after that as we need (hopefully all of them choose a -A large enough for your purposes).

The one major caveat of this approach is that it relies on your log to have at least one message every second, otherwise the exact second that date computed may not have a corresponding message in the log and then grep will not find it and will not output anything. Working around that problem is not that simple though – we’d have to read every line in the log, actually parse the time stamp in the log and compare it to the time stamp we provide and if the log time stamp is later – only then display the line. grep is useless for this because it can’t do any comparison other then comparing texts to check that they are identical.

My solution for this is to use date again to parse the date format in the log file (maybe with some modifications so date won’t get too confused) and then simply compare epoch time values:

1. cat /var/log/httpd/balancer-access_log | ( date=$(date -d “10 minutes ago” +”%s”) while read line; do [ ”$(date -d”$(echo $line | cut -d’]’ -f1 | sed -e ’s/.*\[//;s/\// /g;s/:/ /;’)” +”s”)” -ge ”$date” ] &x%x echo $line done)


   What I do here is to pipe the log file through a while read loop – this will read each line into the $line variable, consuming all the log lines until the end of the file is reached. Each line is then put through some sed to extract the time stamp in a way that date will like it (again – the code above works for Apache access logs. Other log types may need more or less massaging to get at a parsable time values, but either way a different code will be needed). We then put the time stamp through date and use the format code +”%s” to get an epoch time stamp and compare it to the epoch time stamp we generated at the beginning from our “free language” time specification. If the parsed time value is greater (or equal) then we display the line.

   The one major disadvantage of giving up on grep is the performance that will be orders of magnitude worse when you actually have to do some processing of data .

   ]]> http://geek.co.il/wp/2010/04/08/script-day-output-the-tail-of-a-log-based-on-time/feed 0 http://geek.co.il/wp/2010/03/21/firefox-3-6-automatic-upgrade-seems-to-be-paying-off http://geek.co.il/wp/2010/03/21/firefox-3-6-automatic-upgrade-seems-to-be-paying-off#comments Sun, 21 Mar 2010 13:00:41 +0000 Oded http://geek.co.il/wp/2010/03/21/firefox-3-6-automatic-upgrade-seems-to-be-paying-off Mozilla’s decision to offer users of Firefox 3.5 to automatically upgrade to 3.6 seems to be paying off quite well:

   From looking at some web statistics that I generated from leading web sites, this is how the browser market place looks for the top 95% of web users:

   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   

   Browser	Market share last week
   Internet Explorer 7.0	26.69%
   Internet Explorer 8.0	22.55%
   Firefox 3.6	13.67%
   Firefox 3.5	9.36%
   Safari 4	7.84%
   Chrome 4	6.87%
   Internet Explorer 6.0	6.00%
   Firefox 3.0	3.46%

   
   

   There is some kind of anomaly here with Internet Explorer 7 growing compared to last month, which I’m not sure how it happens but I’ll keep my eye on it.

   Other then that we can see that the migration from Firefox 3.5 to Firefox 3.6 has been much stronger then the migration from Internet Explorer 7 to Internet Explorer 8.

   On the other hand – even disregarding the IE anomaly, we can see that Internet Explorer 8 struggles to gain market share against Internet Explorer 7, taking more then 3 months to take less then half of IE7’s share, while Firefox 3.6 took up about 65% of its slightly older sibling in about two weeks.

   This is a major success for Mozilla and I for once am very happy with that as 3.6 brings some changes that I’m very interested in like support for Microsoft’s document.readyState feature.

   I can only hope that Microsoft can achieve such a success with the migration to Internet Explorer 9, which promises to bring many standards support improvements including finally support for normal DOM events and will allow web developers to finally use a proper event based programming model without all this attachEvent compatibility junk.
   

   Related articles by Zemanta

   
   
   
   
   • News Story First look at Internet Explorer 9 (daniweb.com)
   
   
   • Firefox 3.5 wins top dog browser crown – sort of (theregister.co.uk)
   
   
   • The Second Coming of Internet Explorer (java.sys-con.com)
   
   
   
   
   
   ]]> http://geek.co.il/wp/2010/03/21/firefox-3-6-automatic-upgrade-seems-to-be-paying-off/feed 0 http://geek.co.il/wp/2010/03/14/how-to-build-a-chroot-jail-environment-for-centos http://geek.co.il/wp/2010/03/14/how-to-build-a-chroot-jail-environment-for-centos#comments Sun, 14 Mar 2010 14:57:02 +0000 Oded http://geek.co.il/wp/2010/03/14/how-to-build-a-chroot-jail-environment-for-centos A chroot environment is simply a directory – inside which you can find a file system hierarchy exactly like your original operating system. You can then use the UNIX chroot command to open a shell in that directory so that command running under that shell see only the chroot environment and can’t mess up your system. This is very useful for many different reasons – for example if you want to build some software packages and you don’t want their build dependencies to pollute your real system.

   Building a chroot environment is not difficult at all using the right tools, and YUM – the CentOS installation tool – has what you need.

   To start, log into your CentOS system and create yourself a directory where you want to build your chroot jail. I usually choose something like /var/tmp/chroot, so I would run
   
mkdir -p /var/tmp/chroot


   Next step is to initialize the RPM database so that we can install all of the software we need in the chroot jail: we will need to create the directory for the database because RPM expects it to exist, and then use the RPM “rebuilddb” command:
   
mkdir -p /var/tmp/chroot/var/lib/rpm
rpm -rebuilddb -root=/var/tmp/chroot


   In order for YUM to manage to install software into our chroot system it needs to know which CentOS version to install – for this to work it needs the package centos-release to be installed in the root. Doing this properly will take a lot of time and require us to manually download and install many packages – and that is boring. What we will do instead is just to download the centos-release file from CentOS mirrors:
   
wget http://mirror.centos.org/centos/5/os/x86_64/CentOS/centos-release-5-4.el5.centos.1.x86_64.rpm

   and just install it forcefully:
   
rpm i root=/var/tmp/chroot -nodeps centos-release-5-4.el5.centos.1.x86_64.rpm


   Finally we can call on YUM to install the rest of our system:
   
yum—installroot=/var/tmp/chroot install -y rpm-build yum


   The last step will take a bit of time to complete as YUM has to download all the software that goes into your chroot jail environment, but when its ready – you can run chroot /var/tmp/chroot to see how it looks inside.

   Optional

   
   It probably feels pretty bare inside, with a default bash prompt and no aliases. To make the chroot jail a bit more comfortable you can populate the chrooted root directory with the skeleton account files so it feels more like a proper system – before you chroot, copy the files from /etc/skel to the chroot /root home die:
   
cp /var/tmp/chroot/etc/skel/.??* /var/tmp/chroot/root

   Then run the chroot command and tell it to start Bash as a login shell:
   
chroot /var/tmp/chroot /bin/bash -l

   Now you should have a nice bash prompt just like in a real system.

   The special file systems

   
   If you try to run some stuff inside the chroot, for example – yum, you’ll see that some things are not really functional because they want access to /proc or /dev. In order to work with these you will want to mount the real /proc and /dev file systems into the chroot, like so (but get out of the chroot first):
   
mount—bind /proc /var/tmp/chroot/proc
mount—bind /dev /var/tmp/chroot/dev


   As you can surely figure out – this will let processes from inside the chroot (especially if you’re running in the chroot as the root user) to see and tap into various parts of the real system even though they do not have access to the actual files – this is a security issue and if you plan to run software inside the chroot that you do not trust, then I suggest not doing a simple chroot, and instead using a virtual machine or something like User-Mode Linux.

   Of course binding the /proc and /dev file systems is optional and if you don’t run any software inside the chroot that needs these, you can skip that part.

   Network

   
   Network access should work fine inside the chroot, but we didn’t setup name resolving – fortunately this is rather easy: just copy the nameserver file from your real system into the chroot:
   
cp /etc/resolv.conf /var/tmp/chroot/resolv.conf

   And now you can use YUM to install additional software from inside the chroot jail. Of course you can always exit the jail and install from outside using the yum —installroot switch.
   
   ]]> http://geek.co.il/wp/2010/03/14/how-to-build-a-chroot-jail-environment-for-centos/feed 3 http://geek.co.il/wp/2010/03/07/another-mac-styled-update-for-ubuntu-window-buttons-to-the-left http://geek.co.il/wp/2010/03/07/another-mac-styled-update-for-ubuntu-window-buttons-to-the-left#comments Sun, 07 Mar 2010 17:13:33 +0000 Oded http://geek.co.il/wp/2010/03/07/another-mac-styled-update-for-ubuntu-window-buttons-to-the-left As we’ve know for quite a long time now, Ubuntu is aiming to look and behave as much like Mac as possible – we’ve already seen the Growl-like on screen notifications (which are rather cool) and here is the next major user interface change:

   In Ubuntu 10.04 – due to be released on late April of this year – the window buttons (close, maximize,minimize) will be on the left side of the window! See here for the branding screenshots.

   In the previous release, Ubuntu already changed the order of the minimize and maximize buttons so instead of the MS-Windows-like order (minimize, maximize, close) Ubuntu has been putting the maximize button first. A minor change but it looks now to be a preparation to this very major change – starting with the next version, all the buttons are moving to the left side, like in Macintosh:

   

   And just to compare, here’s the MacOS X implementation:

   

   Care to tell the difference?

   OK, on the Mac, the button order is reversed then in the new Ubuntu: close, minimize, maximize. So I’m not really sure what’s going on.

   Anyways, quite likely people who are used to Ubuntu, other Linux desktops and even MS-Windows are all going to hate this setup, and there is no configuration option to turn this off – this is the Ubuntu way, so so either suck it up or change your operating system.

   Or you can start messing with GConf (GNOME’s configuration system which is very much like the MS-Windows registry). If you do want to upgrade to the latest Ubuntu and keep your window buttons where you are used to, you can install the GNOME configuration editor (“Configuration Editor” in Ubuntu’s Software Center, or “gconf-editor” in Synaptic), look under /apps/metacity/general for button_layout, read its help description and modify what you want.

   Or if you want to just go back to the old behavior without messing with stuff, follow these easy steps:
   

   
   
   1. Invoke the run dialog by pressing ALT-F2
   
   
   2. In the command text, put this: gconftool -s /apps/metacity/general/button_layout -t string “menu:maximize,minimize,close”
   
   
   3. Click on “Run”
   
   
   
   Your windows buttons should now immediately revert to the old setup. If you ever want to try the new setup again, use this command: gconftool -s /apps/metacity/general/button_layout -t string “maximize,minimize,close:”.
   

   Update

   
   Thomas Thurman – a GNOME developer working on Metacity – commented on the button layout issue in his blog and about Ori’s button_layout theming bug (see the comments below). Other discussions from around the web:
   
   
   
   • Ubuntu developer Ivanka Majik with more details on reasons for the change
   
   
   • Launchpad bug report requesting to change back and a lot of discussion on that, including Mark Shuttleworth weighing in.
   
   
   • GUI designer Ian Cylkowski has some interesting comments on the new theme and its button layout and has a better suggestion for the button order, that I think better addresses Ubuntu’s ideas as detailed by Ivanka Majik.
   
   
   • A simple application to easily change the button location or layout without needing to learn GConf.
   
   
   • Ubuntu Tweak adds a nice UI to change the button layout (scroll about halfway down). I think this is now my recommended application for changing the button layout.
   
   
   • Scott Ritchie has a list of problems with the new layout (and a funny animation).
   
   
   • A favorable opinion of the new buttons.
   
   
   
   

   Related articles by Zemanta

   
   
   
   
   • Ubuntu dumps the brown, introduces new theme and branding (arstechnica.com)
   
   
   • Ubuntu Gets New Themes, Logo, Boot Screen, Indentity and So Much More! (omgubuntu.co.uk)
   
   
   • Ubuntu Lucid Lynx changes its spots (go.theregister.com)
   
   
   
   
   
   ]]> http://geek.co.il/wp/2010/03/07/another-mac-styled-update-for-ubuntu-window-buttons-to-the-left/feed 12 http://geek.co.il/wp/2010/02/12/upgrading-centos-5-to-fedora-12 http://geek.co.il/wp/2010/02/12/upgrading-centos-5-to-fedora-12#comments Fri, 12 Feb 2010 04:32:51 +0000 Oded http://geek.co.il/wp/2010/02/12/upgrading-centos-5-to-fedora-12 http://geek.co.il/wp/2010/02/12/upgrading-centos-5-to-fedora-12/feed 12 http://geek.co.il/wp/2010/02/09/15-years-in-the-making http://geek.co.il/wp/2010/02/09/15-years-in-the-making#comments Tue, 09 Feb 2010 00:47:06 +0000 Oded http://geek.co.il/wp/2010/02/09/15-years-in-the-making One of the most annoying issues with Linux’s graphical system (and any UN*X), is that if you have some keys setup for switching the keyboard layout – when using two or more keyboard layouts, such as for writing English and Hebrew – then that key combination cannot be used in any other keyboard shortcut.

   Its most annoyingly present when setting the keyboard layout switching command to ALT+SHIFT (like in MS-Windows), then you can’t do any keyboard shortcut that has ALT+SHIFT in it – such as ALT+SHIFT+TAB to move to the previous window.

   Well, finally there is a solution! As detailed in FreeDesktop.org’s Bugzilla and Ubuntu’s Launchpad , and thanks to Ilya Murav’jov we now have a patch to solve this ancient X11 problem.

   By changing the X server behavior to trigger the layout switch only when releasing the shortcut keys and only if no other key has been pressed, we can now enjoy a perfectly working “switch to previous window” action Also all of the (default) ALT+SHIFT+ keyboard shortcuts in Eclipse will finally work for me out of the box. This will also solve the Firefox keyboard shortcut problem I talked about in this post.

   Now, this is just a preliminary patch. It has not been accepted anywhere yet and X.org developers will most likely want to have this configured and have the old behavior by default – at least for a while. So expect more changes and a lot more time until you see this in a repository near you. But if you are of the adventurous kind, and are running Ubuntu Karmic (9.10) or Lucid (10.04 alpha), then you can try to add my PPA to your system and upgrade your X.org server to the patched version.

   On the other hand, if you’re really into pain ( ) here’s how to build it yourself on your Ubuntu system:
   

   
   
   1. Start by enabling the Ubuntu source repositories: open the “System” menu and go to “Administration” and start “Software Sources”. In “Software Sources”, click the checkbox labeled “Source code” – it may turn in to a dash instead of a check box, that’s ok. Click “Close” and when it asks if you want to update the repositories, click “Reload”.
   
   
   2. Now go to “System” -> “Administration” again and start “Synaptic Package Manager”. When it loads, use the quick search to look for “pbuilder” and check that for installation, then “devscripts” and also choose to install this. Click the “Apply” button to install them and wait until the installation is done.
   
   
   3. Create yourself a directory where you can put the source to be built. Now we need to go to the terminal for the rest of the work, so run it from “Applicatoins” -> “Accessories” and then use it to change into the directory you just created.
   
   
   4. run apt-get source xserver-xorg-core to download the source for the X server.
   
   
   5. Once the download is done, go to the FreeDesktop.org Bugzilla bug I linked above and download the patch from there (it should be the first one under attachments) and save it – in the directory you created you should now find a directory named xorg-xserver-someversion/debian/patches – save the patch there. Its a good idea to make sure the file ends with “.patch”.
   
   
   6. In the same directory where you saved the patch, edit the file “series” and add the file name of the patch file in the last line.
   
   
   7. Now back in the terminal, go into the xorg-xserver-someversion directory and run dch -n. This will open an editor that will let you write a changelog entry for the change you have just done. Write something meaningful and save it.
   
   
   8. To start the build, we want to first create a build chroot as described in the PBuilder Ubuntu Wiki page: run sudo pbuilder create --debootstrapopts --variant=buildd and wait for it to finish – it will take some time. When it is done, you can start the build by running pdebuild from the directory where you’ve put the source package. If you have a PGP key that you want to sign the new package with, you probably want to run pdebuild—auto-debsign so it will automatically sign the new package for you at the end of the build, otherwise you can sign it later or not sign it at all.
   
   
   9. Once the build is done, you should find your new X server DEB packages under /var/cache/pbuilder/result and you can install them by double clicking or by running sudo dpkg -i from the terminal.
   
   
   
   If there is interest, I can also build Fedora 12 packages with the patch – let me know.
   

   [Update]

   
   I’ve finished building Fedora 12 packages for the patched X.org. If anyone is interested in proper shift key handling on Fedora 12, then you can find updated X.org RPMs here. These are compiled for 64bit Fedora 12, but you can also find there the source RPM from which you can build 32bit packages if you are so inclined.
   

   [Another Update]

   
   In preparation for Fedora 13 to be released, I’ve made RPMs with the bug #865 patch for Fedora 13 X.org.
   
   
   ]]> http://geek.co.il/wp/2010/02/09/15-years-in-the-making/feed 1