Alexander Cherednichenko's blog

MySQL failover talk

2011-12-22T15:18:00.005+02:00

Hi! Today I had a talk at local IT user group regarding the MySQL fault-tolerance cluster. No original research neither anything specific - just a bit of experience which may help you setup the same easier ;).

The slides are in Russian: https://docs.google.com/present/view?id=df6sccnc_72gkk5dmck

Alternatively you can look at them locally here:

Wicket 1.4 modal window not hiding contents in hierarchy on close! :o

2011-12-22T07:37:00.003+02:00

Looks like plenty of the posts are for myself not to forget about 'specific' issues found in the frameworks during the development. Well, wicket 1.4.(18?) this time.

The problem

So, we have a modal window. There is a form (call it form2) in this modal window. For the last, modal window itself is within a form (call it form1). Hierarchy looks like form1 <- modal window <- form2.

Both form1 and form2 have validators for their components, and both do have submitting components.

When the MW is shown and visible, everything works just fine according to the Nested form rules of the wicket. However, as you close the modal window by any means, the form2 will still be validated on the submit of form1! However, as it is _not_ visible on client (modal window is closed) we can not show to user what's wrong, they actually don't have access to the components failing.

Wicket modal windows do close by means of Ajax. E.g. when the actual DOM element is hidden by clicking cross, or outside modal window, or close button, wicket issues an ajax-call to server to notify it that window is no more visible.

This call is handled by WindowClose behavior, the subclass of AbstractAjaxDefaultBehavior.For greater detail you may take a look at it in the ModalWindow class. What this behavior does in the original implementation:

Reset shown flag to false, so window will be opened again when next time show(target) is invoked.
In case of the modal window showing pages, not components, do cleanup of the pagemap.
Invoke WindowCloseListener instance (if any has been set to this ModalWindow).

That's it. It means that closing modal window does not hide the components inside it from the wicket hierarchy, and they are treated as visible however they are not really!

The solution

General solution for me was to hide content on the modal window close action.

2 approaches:

Local.
Just implement the windowClosedCallback, pass it to the instance of modal window, and in it, do hide your content on the close action.
Bad things are that if you are developing a reusable component, you may want clients of this component to use windowClosedCallback for their own stuff.
General.
Subclass a ModalWindow. Say, MyModalWindow extends ModalWindow
Override newWindowClosedBehavior() of it
In this method, create your own ajax behavior doing same thing as wicket does, plus hide component if it is there. Pass it on to the return. Voila, we are done.

Now the components will be really hidden, and you won't be getting nested forms edgecase validation issues (or whatever else hiding there - I don't really know).

That's the code snippet:

//... inside our ModalWindow subclass, MyModalWindow

private class MyModalWindowClosedBehavior extends AbstractDefaultAjaxBehavior

implements

IWindowClosedBehavior {

private static final long serialVersionUID = 1L;

@Override

protected void respond(AjaxRequestTarget target) {

respondOnWindowClosed(target); // that'd be a call to the wicket's MW behavior.

if (getContent() != null) {

getContent().setVisible(false);

}

@Override

public CharSequence getCallbackScript() {

return super.getCallbackScript();

}

@Override

protected IWindowClosedBehavior newWindowClosedBehavior() {

return new PoseidonWindowClosedBehavior();

}

Inform user about Ajax failed (no network) in Wicket

2011-12-20T08:33:00.003+02:00

The problem

It is well known to the internet. If you have an Ajax application you have to somehow inform user in case Ajax fails. And in the modern world Ajax may fail very easily - WiFi switched off, bad smartphone connection et al.

Also we have a Wicket which does hide all the Ajax work behind the scenes - you drop components to the page, they do communicate via Ajax.

The (possible) solution

At least we employed the one for us.

1. Create the localized message key in the property file. We are using XML properties, so it'd look like Ajax failed! Reload please

2. Create a function in one of the javascript files you always load. You may do it inline, but the function in the file makes it easier to change. My guess it should survive absence of the Wicket too.

AlertError here is the generic error function showing jQueryUI custom dialog.

function tryRegisterWicketAjaxOnFailure(message) {
   if (!Wicket || !Wicket.Ajax) {
       return;
   }
   Wicket.Ajax.registerFailureHandler(function() {
        alertError(message, 0, 'center', true);
   });
}

3. As our tryRegister takes a message parameter we need to pass it from the Wicket. So, in your BasePage.java or whatever code just add

add(new HeaderContributor(new IHeaderContributor() {
   public void renderHead(IHeaderResponse response) {
       response.renderOnDomReadyJavascript(
           String.format("tryRegisterWicketAjaxOnFailure('%s')",
           getString("ajaxCommunicationFailedMessage")));
   }
}));

4. Just enjoy it. You may put 'refresh' button, or a spinner at this dialog which would constantly check if the network is back. It's up to your creativity and demand.

Passing messages in method names? Why not!

2011-12-17T13:57:00.002+02:00

Have been working with Mockito awesome mocking framework (i LOVE it). It also does use Hamcrest matcher as an option for matching its arguments.

So, say you don't want users of your library to implement interface directly but rather'd like force them to use the Base implementation of this interface and extend it instead? Putting away the reasons which led to this situation, here's how it may be accomplished:

package org.hamcrest;

public interface Matcher extends org.hamcrest.SelfDescribing {

boolean matches(java.lang.Object o);

void _dont_implement_Matcher___instead_extend_BaseMatcher_();

}

Every time you'll try to create a direct implementation of this interface, you'll get this weird method which would tell you what to do.

Good or bad? To me rather bad; something is wrong in design if you offer interface but don't allow users to use it. On the other side it may be good, as BaseMatcher gives some bells and whistles and you don't always know about good base subclass. Also, this may have been caused by some external reasons.

Wicket DownoadLink and AjaxDownload with non-ascii filenames

2011-12-15T23:33:00.005+02:00

The problem - yet unsolved for me :(

You have an Apache Wicket application (in my case, this is wicket 1.4.x). You have a page which asks user to download some content and then lets download it. Content is dynamically generated (file?). It should have a name (default one) on client's machine. Name may have cyrillic (or Chinese) symbols.

Symptoms are well described here http://stackoverflow.com/questions/93551/how-to-encode-the-filename-parameter-of-content-disposition-header-in-http

Behavior and the way to overcome it are well described on http://www.codeproject.com/KB/aspnet/NonUSASCII.aspx

Solutions

First of all, I was unable to solve this completely :(

Coming from StackOverflow post, the good way to do would be to form the URL ending with a filename (URL-encoded), and show no filename in Content-Disposition header. All other tricks and tips work not well under different browsers.

Wicket usual approach is to use either DownloadLink or AjaxDownload (thank you guys for such a good solution! Liked it much). First of them is using usual wicket URL coding strategy and second - pretty much the same, being a AjaxBehavior and thus IBehaviorListener. That far I could see no real way to change these URLs. The issue is that both of those don't allow for custom URLs, they are actually plugged into Wicket's infrastructure of page-component-iBehaviorListener-RequestCycle-RequestCodingStrategy chain, so we don't have direct control over it.

One solution proposed by Martin Grigorov on the wicket-users mailing list was to use

http://wicketinaction.com/2011/07/wicket-1-5-mounting-resources/ for mounting, but that did not work out for me as I needed to use containing page's model to generate the actual data.

Fail :(

Will up if any luck.

Wicket this time

2011-12-15T22:17:00.002+02:00

another incarnation, switching to wicket-related posting as we're doing wicket jobs at the moment.

Hopefully experiences shared here would save someone's time -- that's a good enough reward.

Java SFTP server review - current state

2011-02-18T22:04:00.003+02:00

The life goes on, the new tasks emerge and new research challenges appear.

This time we got the 'heritage' - strangely coupled old system with unclear signal paths (write/poll filesystem instead of doing a local notification call), weird predicates (the file is considered uploaded as soon as 1 hour passes from the start of the upload) and really bad heterohenous design.

The target of the system is get some files uploaded via SFTP (SSH File Transfer Protocol) protocol and process them with some rules, producing output files and making them accessible via the same SFTP. As most of the system is written in Java, we decided to try find the Java SFTP Server to replace part of the existing system. Good things about such an rearchitecturing:

Making signals simple.
Having control over filesystem, clients and authentication.
Introduce uniform build and deploy procedures for the whole system.

A little trouble is that one should distinguish between SFTP and FPTS (FTP over SSL) protocols, as they are completely different.

Having all that in mind, with the good mood we started our little research. Found out some libraries:

Apache Mina SSHD. Sister project of the Apache FtpServer, this one looks really good in terms of the object design and operation. Code is yet a little fresh, but object breakup is really awesome. We collected our server with the Spring beans and that was pretty easy - to replace the key provider, specify key storage, authenticator and more more other things.
If you don't want to bother configuring it yourself, the factory gives you pretty good result as it is.
Being a generic SSH server, Apache SSHD is capable of running SCP and SFTP. Unfortunately, SFTP support is now outdated, and supports only the protocol version 3 (current is 5 or 6). :( We could not use it for the project, as for us it is crucial to have WinSCP support in, and that does not work with protocol v3. After asking about new version guys on the mailing list told me that it'd be awesome if someone implemented that. Who knows if I have some free time, why not? :0
License is strict GPL
SshTools is another implementation of the same. Has declared support of the SFTP, SSH and lots of other stuff. Designed pretty good too, has also notion of virtual file system which is very useful if one needs to move storage to DB or process files on the fly. On the other hand, has few documentation and is also licensed under GPL
JavaSecureFTPd the project is dead as the stone is. GPL, but no one cares :)
J2SSH Maverick commercial; we did not ever try it, but the rumor says it is pretty good. Has all necessary protocol support.

So, for now situation looks following way - if you have a java app and need to embed SFTP server functionality to it, you either need to try SshTools, help Apache SSHD with the new protocol ;) or pay money for the commercial one.

Not yet sure what we stick to...

Reviving for the second time

2011-02-04T11:27:00.002+02:00

Finally, the time has come.
I'm about to change job.
I'm about to change an occupation to something a little different.

Got a new laptop to support my new activities while on the go :) got a new passion and new belief in all the good.

Hopefully will finally establish this blog posting the way i wanted to originally -- one technical posting a week. Think i'll have pretty much of stuff to write on.

Download all java older releases

2010-07-01T11:01:00.005+03:00

Ok, i know that at least once in the life java programmer has to download jdk 1.3.0_u03 to reproduce the old, freaky and disturbing issue on the legacy system.

Because we need to support this system.

Searching for such a download main be a pain in the butt, unless we know the magic url - http://java.sun.com/products/archive/

Yes, I know that's no more Sun, but it anyway allows you to download specific versions of JDK, JRE and other stuff! Which is really useful for real-life experience.

New blog platform - does it worth it?

2010-03-07T13:43:00.002+02:00

Hello again.

I'm thinking of moving to the shared hosting with own domain name and Wordpress installed - seem to like more configuration.

It is now up with the blogger imported.
http://lexaux.net.ua/wordpress

Try it - do you feel it comfortable? Like the styles? Any ideas on whether this change may be good or bad?

Thank you!

Show effect - collecting ideas!

2010-03-07T11:42:00.006+02:00

Hello, my dear fellows.

Show effect - as is
I have recently thought of the 'Show effect'. You know it - when you are presenting your application, or the new iteration to the customer (or anyone external) - you frequently run into situation when the small action results in a large stack trace spanned for the screen?

Imagine the demo to your customer. You:

show nice things - feature A is implemented
show great things - feature B which is very important for the business - is implemented! (applause on the background)
show awesome things - the designs ARE there, even if they were provided 2 days ago!
and then you are about to show little bit of functionality (notifying user about their registration), which is not that important (and thus not that verified) - you get an awful error statring with text Unhandled exception java.lang.NullPointerException

Then all your show becomes a bit compromised.

Minimize your risks with smaller efforts
So, I kept thinking on this for a while - how does one minimize these 'show' risks?
I have been feeling this myself - we had a demo each week for 5-6 weeks with a customer, and we had to demonstrate the progress.
It is pretty hard to show significant feature progress within a week, isn't it? And we were on really tight schedule.
So, these are several things I figured out - sure that I'm not the first who found them, but better would be if you the readers would get them from other's experience than from own:

Time and connection. Be sure that you don't make everyone wait too much and also that everyone is setup for the show. In the distributed teams it is common that dev team is located in different country, so that the technical solutions for presentations are being used (screensharing services and the voice calls usually)
Make sure everyone knows where to get and how to setup the screensharing before you start the show. Provide your help if needed.
Always have a plan of presentation. Even if the things are simple, you need to have written plan before you so you don't waste time thinking on 'what should I click next?'. You should be presenting your work in such a way so your listeners feel that you are completely confident with what you are doing.
Surely, there is always a place for unforseen changes - your customer may ask you to perform some action which does not fit to your plan. Be calm and confident :)
Rehearse the demo, at least 2 times, following the written steps you've done in item 2 of this list.
Try not to show too much - if some features are half-implemented you'd rather not show them than make the whole appearance of the project half-complete.
Make sure no one could compromise your demo without knowing it. Good point here is that you ask no one to touch demo server during the show - by any means. 2 demo users logged in with the same accounts from different machines can lead to catastrophic results.
Don't forget about the user experience. If you are showing something to non-technical people (or even technical people but not of your organization) make sure that you have the unexpected exception page configured correctly - so that even if everything fails, user is still presented with the comfortable page with correct and unambigous steps of what to do next. Making custom pages for HTTP 500, 404 etc is also a nice thing.
Another point here is that if the custom error page is not breaking the overall design feeling. With this little trick subject of your demo would feel much more like a complete application (if it even is a prototype).
Put mocked things in. If you have designed sketches with very nice flash application you don't have yet (and your customer knows you don't) - put them as static pictures. And then as soon as the new feature is completed, replace that static pictures with the functioning parts.
Don't forget to say thanks once the demo is over - the people involved had to put off their ongoing work and tasks (which may be also important) to take participation in your show.
...

The question is still open
It would be great if you could share some of your great experience - cases, practices, ideas - about the show effect.
Thank you! And comment on, I'll add the wise thoughts to the list :)

Testing code hl

2010-03-06T10:07:00.002+02:00

  
public static void main() {
}

Seems to be working now.

Great thanks, dear Heisencoder! I've taken the steps at
http://heisencoder.net/2009/01/adding-syntax-highlighting-to-blogger.html

New York Homicide map

2010-03-04T09:29:00.002+02:00

Here is a strange and terrible example of how maps can be used - http://projects.nytimes.com/crime/homicides/map
The map shows in points the places where murders has happened in New York only, and how they were distributed. Looks like Central Park is the safest place

Be Peaceful!

Time Tracking

2010-02-28T11:22:00.003+02:00

You might have asked yourself a question - what should be your own working day length in order to reach the best performance?
There are different approaches for this.

I always used to work more than intended/I'm paid for. E.g. for the last month or so I've been sitting in the office and performing my dues for 10-12 hrs a day :o)

I started feeling this becoming a mo(:)bius strip - the longer you work the worse results are. I started feeling lack of patience, less creative insight and worse memory.

So, the first experiment I've done is not to work on weekends :) at all.
Next thing to try would be the 'kill' timer - I would start it as soon as I start working, and get off the work not depending on current status when it reaches 9 hours (9 is pretty a resaonable time, giving a hr for all the chit-chat, coffee breaks and lunch)

Will see how it goes.

I've already found a stopwatch for this - XNote Stopwatch - http://www.xnotestopwatch.com/ .
Now need to try how it goes.

Updates would follow soon - and have a good rest of the weekend.

Find a location by IP address

2010-02-22T19:13:00.002+02:00

I was wondering how does one find a user's location (lat lng or city) by an IP address.
Thankfully, found one free and (looks good).

It offers APIs for different platforms (i'm happy with Javascript, yes).

Here is the actual link - http://www.maxmind.com

TDD - what should go first

2010-02-21T16:50:00.002+02:00

Hi.

Plenty of time has passed since the last time I opened The Blog.

The topic of the day is 'how you write code good'. I don't remember times when I had enough time for the unittests. So, i'm trying to figure out several rules for myself to stay when working. Now the following sequence seems to be doing its job:

How this happens now. I want to write a new business-logic method:

I create a method stub in a (bean). With the name only.
I think of what should this method get as input, and what should produce in output, in both normal and exceptional cases.
I modify method to the correct signature.
Then - i generate javadoc ;). In the javadoc I write down all the thoughts I had when was doint 2. of this list.
I go to test. And create 1 test class for this method. In this class, I do write one-by-one all the cases (negative and positive) of how it's done.
Mocking - I'd use it, but yet we have a pretty well-done structure, so I can rely on live objects. Though, I'd try using mockito for the brand new things.
I run tests - they fail. That's good :)
I implement method - tetsts should start working. If one of them fails, I need to correct either the tests themselves or the method - depending on what's error.

/me knows that all the stuff above is well-known, wide-used and so on. The list should be treated as a reminder for myself.

Google Web Toolkit RPC serialization error and Collections.EMPTY_LIST (Collections.emptyList())

2009-09-29T17:13:00.002+03:00

Well, heading made my day.
Actually java.util.Collections.EMPTY_LIST and java.util.Collections.emptyList() are not GWT-serializable and will cause SerializationException being thrown on attempt to transfer them across the wire.

Beware!
Use new ArrayList() instead

Integrating Google Analytics and GWT

2009-09-14T13:22:00.003+03:00

Problem is that GWT apps are single-paged by their nature; they are applications. Though, having way to track user's action would be good. No one cancelled custom-written stuff, but mature analytics tools such as Google Analytics would give you much more functionality.

Fast, and easy. Have not tried it yet - need to get rid of some pending work.
here's link to google groups discussion - http://groups.google.com/group/Google-Web-Toolkit/browse_thread/thread/61912b40a2ca7b2a/

Shortly, you just load analytics script to your single-paged site as external script, and then manually call 'log' method from them to create records in analytics database. Pretty easy, huh?

2009-09-14T12:29:00.002+03:00

Nice thread on GWT and SEO on GWT group.
You might have seen Ian Bambury on the web, with his http://examples.roughian.com/#Home GWT example site.

In this thread Ian describes way his application is indexed. Think this is cool, especially real-world experience in long-running gwt seo :)

Thanks Ian.

Design anti-patterns

2009-09-04T21:46:00.002+03:00

I have not had so much fun for some time.
If you know what are design patterns - take some time to read through the document, I am sure you'll notice many wel-known real-life examples of this.

That's where life meets theory:
http://brajeshwar.com/2005/resign-patterns-ailments-of-unsuitable-project-disoriented-software/

Metrics - good or bad?

2009-08-18T21:04:00.003+03:00

Outside world example
When participating in software projects, people like to speak about metrics.
This is caused by the nature of the software projects. Program develipment oftenly turns to the research task.

You can easily track progress for the bridge construction. It's very easy. You can do 2 things:

Get financial documents and reports and see how it's going.
Go to the construction place and see how it works.
Ask constructors for report. Constructors would tell you, that 'Bridge is complete for 23.56%, first phase is done for 98%, seconds phase - for 33%'
Also - there are some numbers about quailty. For example, 4% of weldings need to be redone, 1.23% of surface paintings fails, etc...

Business people are happy. They have numbers. Using these numbers they can see progress, make plans and feel them pushing the thing.

Metrics in software projects - good or bad?
What about software projects?
You can not go to the construction place. You can ask programmers about 'How it's going?', and you'll get the answer 'Doing ok.'

You need numbers to show to business people. Business people will become upset very soon if you will tell them for a month that 'Progress is good!'

Also, having numbers stimulates programmers.
Here are several examples:

Code coverage. Percentage of code lines 'covered' with unit tests. Covered means here that when tests execute, this line is reached by execution.
Commit percentage. Mr. Alex is committing (putting code to repository) 3 more times than mr. Andrew. We need to punish Andrew!
Usually, work breakdown exists in form of tickets - bugs, feature requests, development tasks. People LOVE having metrics on these. For example - 299 work items done, 1020 left for work. Or - this milestone is completed for 50% - 144 of 288 tickets closed.
Easiest - count of the code lines produced by a developer.

The Main Thought
Metrics can do both things - indicat process, or hide it. As a business person, you can get positive reports during 2 months, showing that codebase grows speeds up and speeds up. But after those 2 month you'll face real f**ck-up.

Metric should be sane and wise.

You would not decide hospital to go to basing on the average temperature among all patients, would you?

GWT Canvas widget - drawing in a browser

2009-08-13T19:00:00.005+03:00

Story

I believe many of you felt that some thing are hard to acquire with pure DOM objects and maniputllation. DOM puts you into hard fram. If you ever tried make something round-cornered, you know that count of the libraries doing this exceeds any reasonable task.

Methods used there are different: some use colored graphic corners, but don't allow you for selecting any color you want. Some of them need JavaScript runing on a page. Some need some weird markup, and tricks with the HyperText controls for IE. Some simply don't work well.

So, why not just to draw?

So, (how) can I draw in browser?
HTML 5 Canvas elements allows some drawings. But, making a web-application today, I'd not use this as a production decision. Simply because some old bad browsers don't know it.

And we don't want to make people feel lost because they are using IE6, do we? There's a solution, though. It's name is VML. It's Micro$oft answer to the SVG. (by the way, SVG is not that a good standard itself, too). Internet E$plorer does support VML, but the API is quite different from Canvas.

If we take 2 coverage areas - one of browsers already covered with Canvas tag (FireFox, all webkits, Opera) and VML (IE's) - we'll have plenty solid browser coverage.

The problems yet to be resolved is a cost of development. In order to get something working, you'll have spent more than twice more time than for single Canvas or for VML only.

So, (how) can I draw in browser? (GWT way)

When someone says about masking differences between the browsers, I always think of GWT. That's the best thing I love in it.

And, I used to search for vector graphics library for Google Web Toolkit.

I found one (just as a proof-of-concept) in the GWT-Widgets library. (There's also an interactive demo for the Canvas widget - you can draw different shapes over image just in a webpage without any flash, java or any other hard stuff).

Actually, I liked it.
Also, after incorporating the GWT-Widgets Canvas to a project, I found another one. This is demo for newer Canvas wrapper, and this is gwt-canvas project home page

The least I wanted to show you is the sample of Canvas working in the real world. The following is a way to see it (it's in Cyrillic, so I'll do verbose notes :) )

The site is about calculation of the metal-plastic windows cost basing on different selections from user.
Open up http://osnova.ks.ua
It loads slowly, but wait until all the stuff is loaded.
In the top navigation menu, select 'Посчитай!' item. This means 'Calculate' in english. It is the fourth item from the left, and it has vivid red background.
You'll see a 'Выберите тип конструкции...' blue link underlined in the center of screen. Click on it.
You'll see basic shape selector. The shapes themselves are drawn with canvas.
Select one - and you'll get it big. Big shape with all sizes is also drawn with the help of canvas widget.

That's all for this site.

What I found good and bad in drawing on the webpage with GWT and Canvas

Pros:

It works
It works in different browsers
It works fast enought
Instrument palette is enough for many basic features

Cons:

As it is PoC, there are some issues. I've seen that sometimes wide lines disappeared in IE, leaving wireframe only.
You can NOT draw text. The text you see there - it is just images with letters. Actually, there's a solution for this - but it's weird. There's a pair of client and serverside script. Client sends request to the server with the phraze it wants to get. Serverside has the text with all the alphabet as an image, and then glues the phraze, thus creating image with the desired text, and then passes that by to client.
That's too heavy for such a simple task
It can be done more easily with Flash, but this is a double-sided penny. You'll need your users to have Flash installed. From the statistics of people coming to this blog, 3 of 80 don't have flash installed. That's too bad for big amount of people. :)

Hope this stuff will be useful for you!

How to deal with big GWT applications

2009-07-24T15:56:00.002+03:00

Well, this is a bit of personal experience.
Instead of making extra-large GWT app having all the functions at one place I would use 'mixed' approach.

'Mixed' means that we install different app features on different pages (yes, user would need to change page from time to time) and exchange data through server session storage.

This would give us several good points:

Page is (oftenly) refreshed, thus freeing memory :)
Not that much code to load/execute
Ability to make more clean code (not so much intermodule communications)
More structured and less complex code (no 'overhead for common'). That's the main, I believe.
Ease of content rendering.

Bad things:

It's not that sleek and smooth.
Lots of requests to server. Additional files loaded (.cache, .nocache, all dependenices all the time. it's cached, yes, but anyway...)
Weird compilation. Having 7 gwt compilations is even worthe than 1, I think. Though, can be well-automated with ant.
Can not use anymore stateless server (so, not that easy clustering/performance scaling)

It's something that can be considered as 'good practice', and 'personal opinion'.
If I were starting large GWT project now, I'd design it as described here.

Btw, take a look at LinkedIn. :)

Turn on English locale in psql PostgreSql client

2009-06-25T20:07:00.004+03:00

Problem with encoding (unicode PSQL and cp1251 terminal)

That's kinda strange.
When you run the psql - PostgreSQL RDBMS console client - it picks up your current language (the one set up in your system). That's really cool, but it works in unicode, and if your terminal is not unicode (for example, cmd.exe on windows), you'll see something weird.
That's the example. (take a look at weird symbols after the 'count' results. I really think this is analog of 'row' in Russian)
The easiest solution of psql/terminal encoding problem you'd think of ever
Just type in

set LC_MESSAGES=en

in windows, or

export LC_MESSAGES=en

in UNIX, and you'll get it working. The cool thing is that it is written in psql documentation, but kinda not-that-easily-found. :)

Enjoy!

News from python struggle field

2009-06-23T14:54:00.002+03:00

Decided not to use templating yet, but generate the html imperatively with the custom module, based on the JAxml.
Module should have simple notions of:

Atomic elements - paragraphs, links, text, &c
Grids (customizable and agile)
Automatic resolution of the attached pictures (through href="cid:picture1")

Don't have much time now really, hope to get on this after we release on the work :)

Searching for a good email/html....

2009-06-19T20:31:00.002+03:00

Well. I'm searching for a good templating engine for python. For filling data (such as reports).
May be not only html, but full reporting library/set of routines.

If anyone knows - let me know.

Once I find it - I'll post results here :)

Learning python

2009-06-19T11:11:00.002+03:00

I think I like Python very much...

I spent several days reading book about 2hrs a day, and wrote small script which reads database, prepares report, generates html content and sends this content as an email message (both html and text version, for old/specific clients).

The script is scheduled with the CRON and it just worked. So, i think this is really good result for the language you don't know.

understand execution model
incorporate database querying support (with named parameters, yeah :)
understand how to generate html (I'm using jaxml module, it's kinda cool. creating html document just looks like doc.html().body(color="red").p(string, align="center") )
understand how to send e-mail message (yes, mime knowledge gives good help, but neretheless. I'm sending this through GMAIL - so, SSL support is also here)
understand how to work with dates (that's a bit ugly now, I'm sure there's better way of obtaining today timestamp than time.mktime(datetime.date.today().timetuple()) )

All this, resulting in a working solution, took for me about 6-7 hrs from the start. Without language knowledge at all :) : )

Learn python!

What's up with me

2009-06-16T14:25:00.001+03:00

One may be interested in what's really going on...
I'm working. Working really hard, and have no minute of free time to drop something to blog :(

So, just as there is free second, I'll continue on this.

Btw, this message should better go to Twitter. Maybe, need to create an account there...

CMD replacement with some nice features

2009-05-27T18:50:00.002+03:00

Just found out how much I dislike the standard Windows XP command prompt (cmd.exe).

So, found project named Console. Here's brief list of its features:

Support for tabs
Transparency (don't think anyone _needs_ this really, but, anyway - that's a feature)
Can make long lines unwrapped. But did not find a way to scroll to the right :)
Good fonts from the start :)
Lines selection model instead of box selection model in cmd.

And that's really strange:

Yes, it supports resizing, but does not support maximization. I found this weird

Generally speaking, this is useful, especially when used together with UnixUtils.

GWT FlexTable bug

2009-05-26T14:40:00.002+03:00

Just faced a complex and really bad bug with GWT's FlexTable.

It tends to forget about the cells removed! So, if I remove some cells, and then add some more, the old ones randomly appear.

Problem is that I was developing really complex layout with dynamic cell manipulation and hard use of rowspan/colspan. And spent 4 hrs fixing algorithm :( :( :(..

Technical details

Methods FlexTable::clearRow() and FlexTable::clearCell() seem to be broken (at least in GWT 1.5). They do their job, but sometimes deleted cells reappear in table. That's frustrating.
This is also stated on the Google code tracker : Issue 1766
I avoided this problem by replacing calls to clearRow and clearCell to FlexTable::clear(). Yes, that's a heavy solution - but it did fit for me :)

Hibernate IN statement

2009-05-19T14:16:00.006+03:00

Problem with hibernate selections of ranged values (sets)

update (2009-05-19):
Thanks to Denis Tsyplakov, found HQL in expression :D. Problem comlpetely closed. So, one should better user HQL, not criteries.

Disclaimer: there's nothing really new in this article. You might know everything which is written here. This posting is mostly for myself and people facing the same problem (with memory :) I just forgot how this should be done. Thanks to colleagues returning me to the right way. Here is the problem described:

We have a mapped Hibernate entity.
This entity has some field. :) For example. Let it be some enumeration (status?).
We need to select the entities with several statuses.

Here's example of such entity:

@Entity
@Table(name="entity")
public class Entity {

@Id
@Column(name="entity_id")
private long id;

@Column(name="entity_status")
private Status entityStatus;

// Getters and setters.

}

and example of the enumeration:
public enum Status {
STARTED, COMPLETED, NOT_STARTED;
}

We need, for example, select all STARTED and COMPLETED entities, not touching NOT_STARTING.

Solution
There are 3 valid ways I see:

We make a query for STARTED entities, and then other query - for NOT_STARTED. After the results are gathered the result objects are mixed.
Cons: you'll need to do N queries for N statuses. Quite bad, isn't it?
Pros: don't really know. Maybe, it's simple? ;)
We construct a query, with OR statement in the middle. That how it should look like:

select e from Entity e where e.status=1 or e.status=2
I'm using the 1, 2 numbers and not the enum names especially to make statement less complicated.
Cons: single query, but static. That means that if you need to form the list of statuses to load at runtime (set of checkboxes?), you'll need to glue HQL. And any query gluing smells very bad.
Pros: simple, and if you are sure the list of statuses to load would never change... if you are SURE.
And the least one I see. We may perform selection using the Criterion Hibernate API, not the query language. The 'query' itself would look like the following code:

List statusList = new ArrayList(); statusList.add(Status.STARTED); stausList.add(Status.COMPLETED);
Criteria c = session.createCriteria(Entity.class). add(Restriction.in("status", statusList)); c.list(); // get the list of entities with corresponding status.
Cons:uses Criteria API, which really is a code. So, if you plan to store HQL queries in a (properties) file, this way is not for you.
Pros: fast, simple and dynamic. That's what I left with.

Splitting blogs

2009-05-12T16:02:00.003+03:00

So, I decided to split the blog to two separate 'threads'.

This leaves for technical things, as well as some interesting reports, everything related to work, IT, outsourcing, java and Unix. And the new one (http://way-up-only.blogspot.com/) is designated for personal things.

Yes, the new one is cyrillic. :)

So - welcome!

The first 'real' posting will be about our trip to Crimea mountains. Soon, really soon... :)

I'm back!

2009-05-12T15:16:00.002+03:00

Just got back from the trip to Crimea mountains. That is really cool!
Photos will be soon.

Also, I'm feeling this blog became less technical. So, I think, I'll split it to two :)

PdfChm - find the freshest books for FREE

2009-05-06T14:14:00.005+03:00

The PdfChm resource is a large collections of books on programming, science, maths, law, photography and lots more. There are even books on crime.

So, enjoy. Registration is free, but you can download no more than 2-3 books per day.

There are pdf, chm :) , ps, djvu, and other formats available.. Readable anyway.

Good practices vs Bad practices

2009-04-27T14:47:00.004+03:00

If you are software developer, or designer, or photographer, or engineer, &c, you may have heard following:

– there are good practices and bad practices, and you should always - i said ALWAYS - use good practices.

That's wrong. You should use good practices in 90% only. Not more. And also you should apply your head when deciding if to use that cool design pattern in this place.

Have a look around, and you'll find tons of the examples.

You may have heard of database normalization. Basically, this is the way of developing DB schema powered by straight mathematic researches form relation theory. Idea is that if you follow that theory you'll never get data inconsistency. Opposed to this, there is denormalization, which is used 10 times more rarely than normalization, but brings beatiful performance shifts.
Rule of thirds in photography - split the frame into 9 common parts and situate the object in the crosses of the splitting lines. You'll concentrate viewers' attention on the object.
- yes, that's true. And now go to any more or less interesting photo museum, and look how frames are organized. Basically, this is the second level of harmony ;)
You know - you should never use magic numbers. NEEEEVER! If i see something in your css code saying offsetWidth = (width - 16) + "px" I would not be happy with that.
But nature owns all of us :) Following picture describes that very well.
Explanation:
Google uses different stylesheets for different browsers.
Internet explorer handles 100% widths awfully. If you embed some element with paddings and margins into other element, it will take more than 100%, because IE does not use padding and margin values when calculating 100%
Width of the scrollbar in IE is 17px
There is a way of doing correct 100% width with IE. It utilizes IE css "expression" facility, and recalculates the width on each page event (even mouse move). This would be canonical way.
Although, Google decided to use -17px right margin just to stop loosing performance. It's easier, and better for both user and developer.

Don't you think it is beautiful?

P.S. There's no attempt to reverse-engineer Google code - just attempt to learn how big guys work. No IP violated :) Google is a registered trademark of Google Inc.

Typetester - a little free service form choosing a font

2009-04-23T12:11:00.004+03:00

So, you need to select a font for your

Blog
Site
Publication
Page on the web
...

Or even worse - you have done that work partially, but now (after 2 month) you want to restore the work. And you just can't figure out what font you've used before :)

If you are not professional in printing, I don't really think you'll be able to notice difference between Bitstream Vera Sans and Sans.

So - typester is a solution for selecting correct font.

Hope this will be useful.

And finally - moved

2009-04-21T11:32:00.005+03:00

All right, we've got where to live now.

The only thing left is to buy something to sleep on. Currently we are sleeping on a half (upper:) of the bed, which lies on the wide piece of cardboard.

By the way, do you know why beggars love cardboard that much? It is really warm :)

And - some photos.

1. View of the facade of our app (looking from the yard)

2. Balcony:

3. And, surely, we still need to make some order in the house: :)

4. Real owner of all the stuff:

Moving...

2009-04-02T15:32:00.001+03:00

I'm moving to new apartment.

This has huge impact on time I can spend in front of monitor, so indexable GWT site demo is delayed...

Google Alerts: watch the web growing

2009-03-25T16:43:00.000+02:00

Google has very cool service, which is now that widely used as mail, or docs...
It is Google alerts. Basically, idea is very simple - you just get to their website and enter keywords you want to get notified about.

Once Google bot indexes page, which may be relevant to the search query you've provided - you get an e-mail message about that. Some kind of 'Google bot news' :)

You can tune the service to send you updates on each encounter of relevant page, or pack these pages to daily/weekly digest.

This things becomes very useful, when you watch some activity on the web - for example, your partner's website.

Google Alerts is an example of minimalistic interfaces (as for me), and that's why it has this tag granted :).

Different TRAC workflow for individual ticket types

2009-03-24T18:12:00.000+02:00

TRAC is a tracking software, used to store and work with different issues during software development process - let it be bugs, development tasks, requirements, or feature requests.
It is free, flexible, has good SVN integration (as for me), and is quite minimalistic in its design.

And - yes it heavily uses wiki markup everywhere, making all linkable to all.
But - if you want to setup custom workflow, and you want it to be different for different ticket types, you'll hit a wall. This is currently not implemented in trac workflow.

The good news are that TRAC is python-based, and thus its sources can be changed easily.

Different workflows for different ticket types in TRAC (TRAC ticket link:) ) describes how one can solve the problem, with minor application of patch command. Have not tried it yet, but will report soon...

Denyhosts for Unix

2009-03-24T09:27:00.000+02:00

I've had a Linux box standing on the Internet (uptime~=180days). It just had Ubuntu server installed, Tomcat and SSHd.
And I never looked at security logs...

Security logs (/var/log/auth.log) were full of invalid login attempt messages!
I apparently got there to see who was accessing the machine during this time, and found that log is full of invalid login attempts through ssh - hjackers seem to try to brootforce the box.

So, I just found Protect your ssh-servers, ban offending hosts article about protection - great thanks to authors. And in two minutes all the offending hosts were in hosts.deny file (denyhosts figured them out, and banned), and auth.log stopped to grow because of invalid attempts.

What I've learned from that.
So basic idea is to setup little tool, which blocks connections from some hosts, and automatically adds hosts to deny list if they have stepped through the threshold of invalid attempts.

And - never. Never open direct root access from the ssh. Ubuntu has deeper facility - AFAIK there's no possibility to perform direct interactive login as root (only sudo is ok), but nevertheless - close this down.

Google Web Toolkit (GWT) Search Engine Optimization (SEO) demo will be soon

2009-03-23T14:39:00.001+02:00

I'm going to create a little demo showing how one can make ajax-loaded content indexable by search bots.

Basic idea is to create "secondary site" - static one - showing EXACTLY the same content as primary one. If bot gets to secondary site, it can index it well because of static nature.
If user gets to the specific page crawled by bot, it just gets redirected to the dynamic version of the site, showing the same content.

Here are some key tools I'm going to use:

OpenSymphony SiteMesh - good decorating filter. It would allow us to wrap the entire page into some kind of template. For example, we could put navigation header/footer for bots and old browsers and redirect statement for modern browsers, deciding on user-agent.
Some Java-based CMS for generating/accessing content. Not decided yet. Even better is to use Wiki engine.
Google Web Toolkit, surely.
Tomcat as a servlet/jsp container.

Right now it seems to me that I don't need anything more.
So, waiting for a demo. Hope I'll have several spare hours to get hands on this interesting stuff.

By the way, there's one interesting problem I can see now: If I put static content inside HTML panel, this would cause problems with navigation and links. For wiki, especially. Any link displayed would lead us out of the dynamic application, which is not desired.

Alternate solution is to put content to the IFRAME element. Don't think I really like it, but...

.NET 1.1 LDAP performance: System.DirectoryServices.DirectorySearcher lags!

2009-03-20T11:38:00.000+02:00

This short posting describes performance problem with .Net 1.1 LDAP (in particular, System.DirectoryServices namespace). Problem was solved.

VB.NET 1.1 performance problem with System.DirectoryServies.DirectorySearcher
While working on one of my projects (VB/C# .Net 1.1), I've run into strange performance problems. Controls were loading slowly. After some research, I narrowed the problem - server made several requests to ActiveDirectory through LDAP protocol. Each of those requests took about 7-8 seconds to complete. Same program, written on Java, worked blazingly fast.
Honestly saying, I spent several days fixing this problem.

Problem is caused by DirectorySearcher implementation
Problem was actually caused by System.DirectoryServices library (really a COM wrapper) having some lags with memory allocation, or something like that. I'll add a link once find it in bookmarks.
So, each time request is performed, it at first spends 7 seconds for authentication, or something like that.

Possible solution (worked for me)
Honestly: I did not manage to make this correct. What I've done, is just a plain caching. So, I just saved link to System.DirectoryServices.DirectorySearcher in the static (even not instance, byt in Class variable!), and pinged it in some period.

Caching for DirectorySearcher gave expected results
Yes. First query was awfully long. But subsequent calls were re-using the initialized object, and thus were working blazingly fast! As there are no side effects yet - application is web application, its uptime is more than 2 months.

Useful links on DirectorySearcher performance problem

Discussion on Bytes - mostly advises to tune up ActiveDirectory. Was not very useful for me
In .NET2 M$ intrdouced a System.DirectoryServices.Protocols namespace, which may have been used for fine-tuning the connection and protocol details (whether use or not encrypted connection). There were no such stuff in .NET 1.1
This is a problem with MembershipProvider which people ran into that far. Something like I had, but not similar.
Also, there are some memory leaks inside .Net DirectorySearcher.
And finally - how to make DirectoryServices use connection pooling, which basically solves the problem.

Afraid of being banned by google for 'cloacking?'

2009-03-18T19:43:00.000+02:00

Cloacking, basically, is idea of presenting different content to normal users and to bots. This malicious technique is used by bad people to increase their ranking at Google, and get traffic, while presenting to user content, which was not really requested by user. And there are some good news in the end of the article :)

Example of Cloacking (do not do like this!)
Here is just a rough example. Imagine, there's a page on the web with url http://exampledomain.url
If Google crawler gets to this page, it is presented with one version of page - clean, structured, full of text, headings, etc. This version of the page has lots of keywords, say 'free software download'. It is ranked good by Google bot.
It is possible to understand, who is 'knocking on the door' - machine or real user.
So, if an internet user comes to this page, he sees a face of some PR company showing up.

This is considered to be cloacking, and site may be removed by Google personnel from Google indices once they discover such bad behavior.

Problems for good guys
Ok, this anti-cloacking technology makes search indexes more clear, readable and reliable. But what if you have fully-dynamical site? Yes, site which is fully powered by javascript, or flash?
If its navigation done with javascript - you have problems.
If your application is build with the Single Page Interface - there's no correct way to make it indexable. (this is how guys from Redmond understand SPI)

Good news (for good guys only)
"The only hard and fast rule is to show Googlebot the exact same thing as your users."
These is very good news (yes, for me it's news. I know that posting has been published @07). This means, that your site will not go banned, if you show the same content for same URLs, but in different way.
So, some links:

Ok. Gone for now - will post something on this topic soon...

SEO and GWT

2009-03-18T18:35:00.000+02:00

I'm going to research a bit what's going on the gwt and seo world now...
If you understand you don't know any of those words, GWT stands for Google Web Toolkit - tooling for creating rich web applications, and SEO stands for Search Engine Optimization.

So, getting hands on it...
In this posting, just 2 links to the documents I'm going to work on at first:

GWT Google Group posting about SEO and GWT. Quite an old thread, something may have changed from those times. People which are well-known to the GWT society are writing, such as Sanjiv Jivan and Ian Bambury.
A whitepaper on GWT and SEO from BackBase. I've only started reading it, so can not yet tell my opinion.

Hope that SEO is applicable for such complex and cool applications as ones built with GWT :)

Spiral Universe school management system

2009-03-17T09:28:00.000+02:00

I'm not interested in school software.

But this startup - Spiral Universe School Management System - seems to be emerging fast and might be interesting.

As usual, such systems tend to solve some specific tasks, like schedule management or class journal or medical log, but this system incorporates social network idea together with the school management tools.
Business model seems to be agile enough - you can use this system for free, but for additional services/support you may pay some money. Also, additional user space for media can be bought.

This startup seems to be fresh-technology-oriented. I found out GWT on Spiral Technologies page, so - here it goes.

Waiting for a live demo or a start to try this out :)

Forests-Forever are CLOSING

2009-03-16T17:28:00.001+02:00

There are only two weeks left to watch out the gratitude of the forest all over the world.
I'm not promoting it - just give it a try.

There are really beatuiful pictures of the forests from all over the world. I mostly liked the ones from Ural, Russia. They give you a little of the fresh air.

While watching those photos I wanted to leave that job, get out of the city and just be there...

So, only two weeks left for Forests Forever. Don't miss it.

Getting UNIX to work

2009-03-16T12:54:00.000+02:00

I like the UNIX operating system family. I would not switch to window$ now, maybe I could live with Mac (it's BSD, :)).
As you could hear before, UNIX can be hard to get working as you want it to work. It may require some ... actions from you :)
I got this picture from Kirill Scorp Malyshev ;).

Example of the job done with GWT

2009-03-16T11:28:00.000+02:00

Just found out site I've been developing some time ago.
It's done with pure gwt, when I was completing it, it seemed to me veery cool.

For english-speaking guys: sorry for cyrillic :)

So, here's link: Металлопластиковые окна в Херсоне: ТМ Основа

It is very easy. It incorporates usage of the GWT history support (try clicking back button) , and it has some heavy-duty stuff done with GWT and modified Canvas widget from GWT widget library.

To see it, you can go to Посчитай!->Выбрать тип конструкции.

The wireframe pictures of the window shapes are drawn here, on the client! For Opera/Safari/Mozilla

Exceptional recursion

2009-02-20T09:09:00.000+02:00

Eclipse.
Error seems to be sometimes caused by not-enough-free-memory. Very valuable diagnostic information, don't you think?

Liquid Rescale gimp plugin

2009-02-18T17:32:00.000+02:00

Liquid rescale is "content-aware resizing". You can make your picture wider or narrower with cutting unnecessary picture element and resampling filling elements, so they keep looking good. This posting is just an example of practical use of the effect. All the mathematical grounding and much more samples can be found on official site of corresponding GIMP plugin

Original picture
Here you can see original picture, with three men walking out of sea. It was taken on the Black Sea coast. We want to remove central man, as his photo is fragmental and has no sense here. It is easily done with Clone Stamp - we just need to fill place taken by the man by water elements taken nearby. Doing so would consume lots of time, if we want picture look realistic. And, I think, there are not that much people who like spending hours stamping the picture. Also, I tried doing this with stamp - and failed :).
There is another solution. We can make this task faster, much faster with Liquid resize. By its nature, LQR cuts off those parts of picture which are considered less valuable. More explanation on how LQR calculates value of a pixel path, can be found here (Russian :) ).

Practical view
Now we are going to enhance this picture with just single effect applied. I already have LQR plug-in installed. If you wish to install it for your Gimp, go here.
Now, we start Liquid rescale plug-in from Layers menu, and create two layers. First layer will hold object masks to be saved during resizing, another one - to be deleted.
I have painted in red in deletion layer, and in green in preservation, in order to see future picture composition easily.
Now, starting the resize...
It takes some time - calculations are not that lightweight.
Also, before running plug-in you can decide to store all the path being cut in separate layer so you could see which pixels are cut.

Resulting image
And this is the image after effect application. Just after single action, without sitting for hours and clone-stamping picture. Et voila.
Better examples can be found on the official plugin example page (even with video).

Here is my useage pattern - if you have some photo with elements that you don't like, and the picture itself is not that valuable to spend a day or two on enhancing it, you can use LQR for quick fixes.

Blog purpose

2009-02-18T12:08:00.000+02:00

For anyone reading this - I am going to use the weblog for storing some interesting and non-straightforward technical solutions - mostly not to forget about them.

Awesome

2009-02-08T14:34:00.000+02:00

Awesome
Awesome is not just an exclamation - it is a name for perfect window manager utilizing tiling windows concept. It can be used for X windows systems on the *nix and BSD systems. I'm using it now, and I would like to share awesome experience.

Background
The idea of multiple windows shown at the same time on the single screen is known for times. Curses, TurboVision creates interfaces capable of showing pseudo-windows with ASCII characters. Bunch of software called terminal multiplexers (see Terminal Multiplexers on Wikipedia) is used for showing several text program outputs on the same text terminal. The best use case I can see is tailing several log files to monitor distributed activity in runtime. Or I can run some commands in the active console, and watch their output (which goes to file :) ) on other virtual terminal.

Nowadays people mostly use graphical user interface. In the Unixes X Window system makes all the low-level work on rendering data and basic manipulating windows. In the Redmond-like GUIs we got used to window decoration, overlapping windows, task bar showing tasks being executed, ability to drag and resize windows with mouse. All this stuff in the Unix world is done by Window Manager (WM). There are different WM's doing their job in different way.

You might have noticed monitor aspect ratio change trend. Monitors are going wider. Resolution goes higher. Have you ever tried maximizing any file manager on the 1920x1200 resolution? One may feel uncomfortable and lost in such a wide space with lots of icons. There is not that much tasks that really require that much information shown on-screen simultaneously. On the other side, if you open lots of windows to utilize informational space, you can run into problems with finding appropriate one.

Tiling window managers
The descendants of the terminal multiplexers are tiling window managers. General idea is to split the screen into non-overlapping areas, and put each window into appropriate area. With this concept, you can see all the information at the same time. And you don't need press Alt+Tab all the time :).
Tiling concept is extended with various features, such as different focus models, static or dynamic window arrangement, virtual desktops/tags, tabs to arrange windows in the single slot, and much, much more.
I am using Awesome window manager.

Awesome features

Dynamic layout model. When new window gets opened you do not need to position it somewhere manually. Awesome decides where to put it based on the windows role, current configuration and layout.
Full keyboard control. You can do everything with keyboard. Move focus forward/backward, switch layouts, rearrange windows, change cell size, etc...
Full mouse control. In addition to keyboard, you can do all the stuff mentioned above with your mouse. It is very handy when working with graphic editors, for example.
Tags. You might have heard of virtual desktops - concept, used heavily in Unix desktop environments. Tags are different. Instead of setting single virtual desktop for window you can assign several tags to it, and window will be shown in each of these tags. This feature allows you to show some important windows on all the tags (so they are visible all the time)
Layout models. Awesome allows to assign different layouts for tags. Windows can be arranged vertically, horizontally, full-screen, maximized, centered (active window is in center of screen, all other are in background tiled vertically), spiral, flying. The least one is the layout all of us got used to - overlapping windows. Yes, for some tasks it is very handy.
Performance. Awesome does no windows decoration. No borders, no headers. Awesome uses XCB - the fastest way of talking to X server by now. Size of the program is 8 MB. Even when I do full compilation of my project, and both kernels are fully loaded, I can manage windows normally.
Programmability. Awesome itself is written in C, but all less important features (menu, key bindings, tag switcher, task bar, tray icon, clock, etc...) are written in Lua - scripting language. You can easily change anything you want. This does not even require recompilation.

Screenshots
Here are some screenshots made on my workstation.
1. Vertical layout. Red border across top window (FireFox) indicates focus. Top line is the task bar. Small icon is the menu icon, 1-9 are tags. Taskbar is long, and spans across all free space. Clock is the widget i wrote myself in Lua. The lease element is the system tray having Skype and Pidgin icon.

2. Dual-monitor setup. The left screen has several IM windows with horizontal tag layout, the right part has Thunderbird and terminal opened. Two monitors are my laptop primary screen (15" 1680x1050) and external Dell monitor (24" 1680x1050). The Awesome setup shown here is outdated for me.
3. Current layout in centered mode. One window is in center, but allows seeing other. Yellow border is a mark border. You can perform single operation to several marked windows. So, window group operations are supported in Awesome, too.

Hope this posting may help someone find own way :) at least in computer interfaces.

Developer hell

2009-02-04T20:13:00.000+02:00

The one who knows would understand me veeery well.

I'm now working on the project which looks in the following way. It started to be developed 8 years ago. There were about 4 teams of long-term hired outsourcers and a couple of consultants sustaining that sh%t-code for these 8 years.

And now - just listen - the code looks the following way.

.Net 1.1 windows application, which has a tree on the left, and different panels showing up on the right, depending on the node selected in the tree.
Sometimes on the right .Net winforms components are shown. Sometimes, ASP.NET 1.1 pages within hosted browser (IE ActiveX) are shown.
There is a couple of situations, when inside that ASP.NET 1.1 pages shown within IE Activex shown within Windows Forms application, Visual Basic 6 ActiveX control gets up.
Haven't you lost the story thread yet?
And - the best one. Where do you think does that custom VB6 ActiveX take data? You'll never believe. From the ASP.NET 2.0 Web service, which is published on the same server where the ASP.NET 1.1 pages containing that VB6 ActiveX are published.

Have you ever thought this is possible.
And - the last.

Users of this system were completely satisfied with this state for almost 2 years, and only after that lifetime system started to slow down...

Would you like such a job? :D

Google Docs

2009-01-26T12:28:00.000+02:00

This is not an advertisement.
It's just a short story of how today's services can be used in order to perform some daily tasks for free.

So, here's an example.
We have:

Document in M$ Word format, created elsewhere
Laptop with Linux installed on it
OpenOffice on linux which does not open correctly the M$ Word document (styles are shifted and paragraphs are misplaced with text
Need to port this document to PDF in order to e-mail as CV.

Instead of installing Word application on the virtualized XP, we can do much simpler thing:

Just forward an e-mail with document attached to the specific email address which is found in the Google Docs setup page, on Upload tab.
Log into Google Docs.
Open the document (it is already imported - Google has imported all the attachments as separate pages.)
And from menu select Save-As PDF File
That's all - you can now dl it!

So, economy is completely understood. How much does Microsoft Word license cost?

MInimalizm

2008-12-10T18:02:00.000+02:00

First of all, sorry for showing plain thoughts, without specific technical background.
I have found, that I like minimal things. I like:

1. Ipod shuffle 2nd generation - it does not have screen. It has only the blinking led. It has only one hole through which it gets charged, filled with data and into which your favourite headphones fit flawlessly.
2. TRAC. Yes, this is an abbreviation. http://trac.edgewall.org/ This is a tracking software (for bugs, change requests, some planning, for vivd documentation and for having all this stuff interlinked well. THis really works hard.
3. Awesome window manager. http://awesome.naquadah.org/ THis drives me crazy. Perfect thing for professional works with computer. It is tiling window manager, it does not have nor window headers neither borders, windows don't get overlapped normally, all is powered with keyboard. Result: you don't spend time on working with windows - dragging, resizing, etc.. You just spend time on working.

I hope to explain more of this stuff in the future posts, but tendency is prominent.
Things take your time. You need some things. So, you need things which take minimal amount of your time - MINIMAL ones.

Audio feeds

2008-12-07T11:46:00.000+02:00

Never used podcasts before. Tried JavaPOSSE.
They seem to be rocking, but again - for me it is too lightweight style.

I'd like to see them more strict and technical.

Time, endless short time

2008-08-19T15:34:00.000+03:00

KISS.
We have all heard of this principle, we try to use whenever we can.
There are tons of books about programming that use this principle - XP KISS system design, KISS code, which is easy-readable.

I have spent about 3 hours fighting with Ubuntu sound systems. I have discovered that there are several different sound servers/systems/modules that do use sound devices on different level of abstraction and HAVE NO COMMON INTERFACES.

Know what? My sound disappeared: I could not call with skype, I could not listen to music; no sound at all. I tried to revert to the original configuration, back and force - no sound at all.

And know what the problem was? My pcm device was muted in the mixer. I spent few minutes checking and fixing that. And three hours finding out what the hell is going on.

So, KISS.

Getting to work e-mail posts

2008-03-30T14:04:00.001+03:00

This is a sample post, trying to check how formatting (code formatting) works within blogger when sending the post from the Gmail. Hope this would be OK.

This is a sample code

public void a() {

}

And this is a sample comment, that how differernt formatting would be seen...

--
Alexander Cherednichenko,
Software Engineer at DataArt

[ the only way out is the way up ]

So what - starting...

2008-03-30T13:44:00.000+03:00

So, I decided to try one more time, and start blogging. Hope, this would bring some good things definitely to me, and possibly to ones who would get here.

I am going to post here nopt only technical notes, but also some interesting links, ideas, solutions...

Hope, I'll get a GWT presentation here soon...