twincode.net - News

git & rspec magic (finding the culprit with bisect automagically)

Tue, 27 Sep 2011 09:44:26 +0200

I guess most of you know how to use rspec. And some might have used the power of git bisect. But how can you make them play nicely together?

(In the following example, I am assuming that you use rake, rspec 1.x and bundler, could be a rails project, could be anything)

Well, the easy way is to just run all the spec via the bisect command. Start with

% git bisect start <bad-revision> <good-revision>

and then use

% git bisect run rake spec

until it shows you the bad commit. Use the power of git diff(tool) to compare the last good and the first bad commit, and you'll find the problem in short order.

But what if your test suite is huge, or the startup time for rake spec is not worth it? Well, then you can combine two really interesting flags for rspec: --format e and -e FILE. Example:

Before I prepare bisecting with bisect start ... as above, I am running the specs with the 'failing' formatter via

% spec <directory_or_files_to_run> -f e 2>/dev/null 1>/tmp/failing.txt

which will run the specified files and output the failing specification's descriptions to the file /tmp/failing.txt'. In the next step, I use

% git bisect run spec -e /tmp/failing.txt

thus letting git do the hard work of running the failing specs over and over until the culprit is found. This should be way faster than running the whole test suite altogether, but you also might miss some other failing specs, so beware! Running all specs again will be necessary, and then you might have to start all over again.

Oh, and for those that are missing an example for rspec-2: I'm working on it. It's a bit more involved, because the formatter has to be customized, and the input has to come from the shell (unless I am willing to patch rspec-core), but I am confident that I can get that done in time, too.

Hope that helped some of you, and stay tuned for the rspec-2 method of getting the same stuff done!

Update Hints (rails, ruby, datamapper, ...)

Sat, 02 Apr 2011 17:59:48 +0200

Well, that was an interesting Saturday afternoon. As usual, I upgraded some small stuff, did a little programming on a private project, and then decided to check the versions ...

(The project is using Rails 3.0, Datamapper, Devise and some more things)

And whoo! Datamapper has been upgraded!

So, first things first, I went into my Gemfile and updated the version. Typing bundle install would then give me ...

Errors. Yay.

First, incompatibility in gems. I had to update dm-devise too - and dm-rails (obviously), which lead to upgrading rails to 3.0.5 ...

As soon as I had fixed that, I got more weird errors:

"invalid value for @cert_chain"

Argh! Well, some googling and browsing stackoverflow lead me to the following conclusions:

I need to upgrade rubygems
Following that, I need to downgrade rubygems again, or
I need to upgrade my ruby version.

Going for the full monty, I virtually stripped down my system, dropping gemsets left and right, reinstalling ruby(1.9.2-head) via rvm and then upgrating rubygems (via gem update --system, which seems to work again).

After dropping the vendor/ruby directory, I reinstalled the bundle (bundle install --path vendor/ --rubygems, and everything works again!

... well, almost, now I have to remove the deprecated stuff that Devise introduced ...

... the interesting saturday continues.

Awesome! Rubinius works! (... with RVM)

Fri, 01 Apr 2011 00:02:42 +0200

Long story short:

rvm get head
rvm reload
rvm list known
rvm install 1.8.7
rvm install rbx

WORKS!

Installing Debian squeeze/sid on a HP 620 (Working: 90%)

Wed, 07 Jul 2010 19:59:11 +0200

Installing

Prerequisites

To install any type of Debian on your newly unwrapped HP 620, you have to have the following:

The Debian Installer (d-i) CD (I do use netinstall usually)
Any kind of boot cd that can boot your system again (e.g. Xubuntu installer)
A lot of patience

Preparing the system

First: my system came preloaded with Windows 7. I started the installation process, because I wanted to install Linux in parallel. After futile attempts to resize the partition, create a new one and install Debian squeeze and Xubuntu (one after another), which worked from an install point of view but never managed to boot the system, I gave up and deleted all partitions.

In the BIOS I removed the option to 'protect the disk' and set the controller to IDE (just to be sure).

After that it was pretty straightforward. I installed Debian (minimum, because that's how I roll) and rebooted, just to be greeted by "No System Disk or Disk Error". How inconvenient. But strangely enough, starting the system with the 'boot first disk' option from the Xubuntu worked - and after the first system update, the system was bootable.

The only thing that was left was to activate the wlan device. I installed Xfce and the works, removed network-manager and installed wicd, then downloaded the driver for RTL8192SE from RTL and installed it according to the included readme.

Then I tried to get those pesky extra keys (volume, mute) to work, installed keytouch and the keytouch-editor (via aptitude), but then it segfaulted all the time when trying to create a new configuration. Well, I will leave that for another time.

Out of the important points, the HP 620 works for me. The installation is a bit rough, not all keys work (yet), but after all, everything seems to work now. Now that I know about the pitfalls, I would definitely recommend it so far for running Debian.

Daily Updates (Und täglich grüsst das Fahrrad.)

Tue, 23 Sep 2008 20:55:04 +0200

Hi all,

lately I'm running daily updates of this site, but only in the (german) Bike Section; check it out here.

That's all!

Scrum Workshop (DIY Advanced)

Fri, 04 Jul 2008 10:30:41 +0200

Hello Folks,

I am proud to announce a cooperation between Christian Friedl and me - our first Vienna Scrum Workshop.

Scrum Masters have to weather difficult conditons, gaining valuable insights and deep knowledge - and we want to share our experiences and ideas with others in the same area. Going a step further from Scrum Tuesday, we invite all Scrum Masters to participate and suggest topics, which can be treated in discussions, games etc.

The Scrum Workshop I will take place at the Don Bosco Haus in Wien 13, at July 18 2008.

Please find more information here:

Scrum Workshop I

For questions, please feel free to send a mail to scrum_ws@twincode.net - also if you have suggestions for topics, or just want to notify us that you would like to participate.

See you at the Scrum Workshop I!

tony

Hoax Warning (In german)

Sat, 16 Jun 2007 13:25:59 +0200

Please read my new Hoax Warning.

Thanks.

Life's a bitch (Installation Woes)

Sat, 01 Jul 2006 12:21:45 +0200

Yes, I need to vent a little.

After updating my unstable boxes, my systems went haywire. Very. Can't find the culprit either.

So, after two days of installing and recovering data, I run etch on those boxes.

Works like a charm up to now. Etch installer is great, too.

And Knoppix supports LVM2! Yay!

Security? (Yeah, riiiiight.)

Sun, 04 Jun 2006 12:07:58 +0200

I mean, there's business solutions and Business Solutions (with capital BS!), and then there's things described best as in

UserFriendly Cartoon for Jun 04, 2006

Right? Right.

'Nuff said.

Vanguards goes SOE! (Neverending Frustration)

Mon, 08 May 2006 09:19:10 +0200

You'd think that McQuaid, Butler et al would have learned from past experience: cherishing a viper in one's bosom is never a good idea, especially if the vipers are as big as Microsoft Game Studios or Sony Online Entertainment.

What I read from the press release, titled "Vanguard: Saga Of Heroes finds a new home."¹ is that the relationship between an idealistic game creator and a money-devouring publisher went sour. Oh, I can even hear the wooshing sound of deadlines, and the merciless cries 'program faster, slaves, we have to keep the deadline!' of the marketing droids which again fail to see the prospect of more money in the long run, obviously blinded by the small heap of wealth to be garnered by pressing the last out of their ressources, thus rendering them barren ...

But I digress. Now, you'd expect that they'd try to pull themselves together and start anew with some independent money (which might have been able to be done from starters, but little do I know), they go and throw themselves at exactly the company that already ruined the first McQuaid/Butler title beyond recognition and is already beating the veritable crap out of the successor!

I don't know which is greater: the desperation or the stupidity? Well, Woody, here's one straight for The Zapper: Vanguard: Saga of Heroes. My condolences for the death of the unborn child.

Edit:

I must make amends. As said in this thread about SOE and Sigil on gucomics it shouldn't be that big a problem, but I believe what I can see, and the past has been showing different.

Footnotes:
¹ watch the page title ... someone in a hurry?

The very reason for brewing your own coffee (After coffee in a sac there's ..)

Sun, 07 May 2006 09:45:09 +0200

this.

Never trust coffee in a can!

EQ II - The Obituary (Or: When Mediocre Games Go Bad)

Fri, 05 May 2006 18:52:52 +0200

Once Upon A Time ...

There was this enthusiastic player of MMORPGs. I have played UO eventually, was an avid visitor of MUDs, and almost had a heart attack, when the first really gorgeous and playable (well, mostly) fully 3D game went online - EverQuest (today dubbed EQLive).

Well, I cannot say that my tendency towards hardcore roleplaying (no, not that, you pig! Go, wash your brain with soap!) was always met, but I endured ...

I admit, I have been playing for the better part of six years. Played PvE, PvP, PvE again, was in some small guilds, what can I say, I was a casual player. (My MUD Personalities have been SEK (Socializer - Explorer - Killer), according to the test on http://www.andreasen.org/, which is down at the time of writing this, unfortunately). But I was arduous and loyal. At least to the point where catering to the players who had much more time than me rendered the game unplayable for me. The gaps between me and my friends were too big - and I have always played the unpopular, the underdog classes which were hurt in the process of an ever changing environment.

Dire Times

In short words: disappointed with every aspect of the game, I left. In the meanwhile, the promise of Shadowbane wasn't held, the setting of SWG was none to my liking (oh, how grateful I am that I never played it!), DAOC was too empty and repetitive, and WoW - don't let me get there, agreed?

Promise

But hark, what is that sound in the willows? A shiny, new world opened itself when EverQuest II launched. The realistic world, the movements, the class tree, the character customization options, the sounds! And quests aplenty ... all this made an even bigger promise than every other game out there. It seemed like I had found an old friend, with a serious mien, but the ever present slight hint of a smile around the corners of the mouth, and the warm, shiny eyes which promised a good time together.

So I started out with two of my close real life friends to explore the world of Norrath once again ...

Bliss

I would lie if I would tell otherwise: it was love at first sight. From the User Interface over the quests to the balancing, everything seemed perfect. I happily played my old, grumpy Teir'dal Healer gone Priest towards the ultimate goal: making 'friends' (uh, well, players, not characters), exploring the unknown, doing the undone, you know. Such stuff. There was the one or other quirk, bugs, everything, but with a project like this? And the customer service and developers were responsive ... it seemed too good to be true. Prime example was launch day: I had the game preordered, and couldn't connect, because of my internet provider. SOE employees managed to fix that within hours, which is close to a miracle!

I met people and made some friends, friendships that hold until today. In a fashion, I was as happy as one can be about a game. Honestly, doesn't such a game with it's depth and dedication which is needed, influence our everyday wellbeing?

Disenchantment

Well, it did. The moment my character hit the much anticipated level 20 and was a full fledged Inquisitor in the name of Lucan D'Lere, something started to feel not right. More and more people started to complain about the quirks which were increasingly annoying, the bias between the classes overshadowing the aforementioned balance, and of course: the Nerf Bat being swung mercilessly.

To be fair, SOE tried. They still communicated with the customer audience, they still added stuff, fixed things, and even released so-called Adventure Packs and finally, the first expansion hit the shelves.

Along with that, the proverbial shit hit the proverbial fan: LU (Live Update) #13 was released. The combat system was revised totally, apparently it has been found lacking in multiple severe fashions. Thus, it was changed. To the better, in my honest opinion, but not for others. People left in droves ...

Disappointments

From that day on, the tone changed. The messages on the boards became ruder and nastier, the enforcement of certain policies harder, while others were left completely in the open. The CSRs became unresponsive, and the chat system, where I spent many hours of bantering during downtimes, was effectively locked down and rendered unusable.

Along with that, the boredom creeped in. The quests became repetitive or camp-fests, the competition outrageous, while the monsters were either ridiculously easy or close to impossible to beat. A lot of the promises, for example "Every class can solo equally well.", or "Every profession is as effective as the others in that class." were thrown overboard (well, the latter has some ingrown stupidity, because diversion keeps the games alive, but that's for another rant).

Happily Ever After?

Not. I have decided (and a lot of people with me) to stop paying for something I don't even play anymore. It's become so bad that I haven't been playing for weeks now (except once, when I was coerced to). The quests are lackluster and shallow, communication has been replaced by shouts, shorthands and rudeness, and the game world is challenging noone - except for my graphics card, which was top notch not long ago. There's a bad, bad memory leak somewhere, and honestly, I prefer a good round of old medieval torture carried out on me to what should be a PvP experience.

So, for all that ever wondered, and for those "In The Know", I wish you farewell from

Serath
Natheea
Moriena
Jezarra
Chruf and
Frumuck and Wahgruk

Sayonara Norrath

This time, for real.

Quo Vadis: Security And Safety Awareness (Decision makers on the err.)

Wed, 03 May 2006 22:10:30 +0200

Lately I have come to the conclusion that the average decision maker responsible for IT has little to no idea what all that security and safety talk is about.

Coming from own experience and talking to others in the business, but also to people who professionally do not have to have any clue, the imperative for investing money and time into security considerations is out of most people's grasps.

But why is that so? One of the points is that there is little to no experience of people inside and outside IT about the costs that any violation can accrue. Even more so, the relation between low time and effort attacks and a little spending leads to the fallacy that you have to spend only ten times that to reach one hundred percent security. We all should know how wrong that assumption is. But the problem persists: confronted with the facts that total security means either total isolation or financial measure going through every limit known to mankind, the average manager ofttimes decides to invest nothing.

Another problem I run into always and again is that the terminology along with the technical understanding of decision makers is ... flaky. Playing Bullshit Bingo between 'firewalls' and 'personal firewalls' doesn't leave much for imagination. That the average user does not have the necessary background knowledge to configure, much less react carefully and responsible on the ever changing systems and attack vectors.

Thus the average packet filter dubbed 'firewall' does little to nothing to prevent DOS-es or port scans, much less 'real' attacks. The ever so highly praised 'personal firewalls' are even worse: all they manage to do is to induce a false feeling of security while by continuous selections of 'yes, this Trojan/phising tool may connect whenever to wherever, from now to the end of this computer (which, incidentally, is never far away)' is rendered to a state between little usefulness and total no-op.

Now, the rant is almost off. Who is to blame? That is not the question, after all. What can we do? Well, we can try to enlighten people. Let us inform managers clearly about dangers and possibilities, and weed out the Bad Ones(TM) who deliberately use disinformation to gain money. Let's explain that an early investment can and will prevent much higher reparation cost. I'm still dreaming ...

It's working! (Sometimes Good Things Come To The Patient)

Wed, 03 May 2006 15:11:31 +0200

It's just bad that I am none of them. Patients, of course.

Well, I have my new site up and running, and it seems to do fairly well.

Switching from fcgid to scgi did it, even if I needed a whole build environment and went through some hoops to make that happen ...