Ubuntu, Console VIM - weird characters in insert mode

Now that I feel quite comfy with VIM, over the weekend I needed to edit a config file in my Ubuntu 10.10 Virtualbox machine quickly. Instead of GVim, I just opened the file in console VIM. As I hit i to get into insert mode, a bunch of weird character boxes were inserted. That was not good at all :( - just when you think you're comfortable with something if it does something totally weird. In any case, I was in too much of a hurry to bother and went about editing my file with gVim. Also, backspace was wonky (same weird characters) - so I felt better. For some reason that I fail to understand, why must Linux make proper backspace and delete handling such a pain! In any case, it's something that I've dealt with enough times to know that there'll be something on Google.

Later on, tried to see what all the fuss was about. Googling around, I found :help :fixdel and that seemed simple enough. Alas, when I tried it out, it didn't fix the issue at all. Also, I seemed to be getting weird characters just pressing i to get into insert mode - and the VIM wiki page didn't have anything about that. Neither did Google turn up anything that seemed related.

So today early morning, on a whim, read up a little on VIm terminal handling. I have the following in my .vimrc
[sourcecode language="text"]
set t_Co=256
[/sourcecode]
Maybe it was the color escape code that was coming in - so checked out :echoe &term which returned xterm under gnome-console and builtin_gui under gvim. So I've put the following bit in my .vimrc and it seems to have fixed things nicely:
[sourcecode language="text"]
if &term == "xterm"
set term=xterm-256color
endif
[/sourcecode]

Android Annoyances

So yesterday and today while driving back from work, I've had to join conference calls. The conference call provider we use at work has 10 digit passcode numbers. Usually, I have a few bridge numbers with the DTMF codes saved in my contacts so I can just click on the contact to get dialled to the access number and have the participant passcode typed in for me. However, yesterday and today's calls were on a different bridge and I had to try to remember a 10 digit number after dialling the access code - and all that while driving. Needless to say, it took a few attempts and I'm sure at that time my attention wasn't where it should have been - ie on the road and on the traffic. Besides being thoroughly unsafe on Bangalore roads, its just frustrating(thankfully - better sense prevailed today and I pulled over, dialled into the bridge and then started driving again).

So the issue really is that the native parser that parses out email and calendar invites doesn't understand access codes and passcodes. It shouldn't be too hard to do - but then I started digging a bit deeper this evening. Granted that the parser isn't smart enough, at the very minimum if it handles tel: links properly, then its just a matter of educating folks who set up meetings to set them up so that you can click to call with something like <a href="tel:23423432233,,9230233#"> - in fact, in Outlook if you type TEL: and then the number, it will automatically be parsed as a tel: hyperlink. Turns out that its a massive fail - if I click the link, Android will show me the dialler but without the DTMF codes (basically, only the number upto the first comma). TOTAL FAIL.

So, Isn't this something that should have been brain dead simple to do? I mean - this is 2012 after all - and I'm not asking much. All I'm asking is that the tel: url parsing/handling be done in such a way that we can use our phones properly!!! Turns out that there's an open ticket 4575 since Nov 09. And its marked as an enhancement - I find that laughable since its a bug and definitely something that can be done quite easily (esp since a contact that has DTMF codes is dialled properly). However, for the 2 years that the ticket has languished, there have been 73 comments and not a single response from big GOOG :(

At the moment, doesnt look like this is going to be fixed - so I started browsing thru the Android source tree to see if I can find where the implementation for tel: urls - however, given the size of the android source, that's like trying to find a needle in a haystack. Guess I'll have better luck with seeing if CyanogenMod folks can fix this in CM9.

So about that, the other thing that has confounded me is why in the world can't Android bundle a decent T9 dialer/smart dialer out of the box. I know there are tons of apps on the market that do that - but seriously, is smart dialing something so out of the world that I need an app for it? As expected, there's a ticket but no action.

I think its safe to assume that Google isn't interested in fixing these issues as there's no 'benefit' in doing so - though for the life of me, I can't imagine either of them being particularly hard. In any case, I'm eagerly awaiting a CM9 build for the Nexus One (right now, am running an ICS build from XDA).

Facebook publicize is driving me nuts!

I'm thoroughly frustrated with Wordpress.com's facebook publicize feature. In theory, its supposed to post to your facebook wall whenever you publish a new post and that way publicize your post among your friend circle.....if it ever works. I've done all the resets, disconnects and reconnects and it just doesn't. Now, this could very well be a facebook problem rather than a wordpress.com problem - so while my rant might be misdirected, its a rant anyway against the thoroughly frustrating experience. Its like a bucket of cold water on my enthusiasm to be more active on my blog.

You see, with having posted rarely to this blog, I get a measly 70/80 page views per day (yeah - there's no need for the snide looks); So one part of actively persisting on the blog has been to see if I can get to 100+ page views per day. Modest goals, I admit - and getting the linky to a new post on the FB wall is a big part of it. If only it worked as it says on the tin :( :(

Anyway, this post is a test in itself - I've just jumped through the said hoops , mumbled the magic incantations and in other words, followed every bit of direction available to make this work right. And if this post shows up on my FB wall, well and good. If not, then I'm done with trying to get this to work.

A syntax highlighter extension for Deck.js

So for the past few hours, have been playing with Deck.js . I like the idea of a web based presentation format rather than a blob like powerpoint. At the same time, I'm a bit circumspect too - given the state of the tools. At least for my use, there's really no burning need that powerpoint can't solve (though I get the shivers everytime I have to do a presentation). All the web based/HTML5 seem raw at the moment on some much needed features (slide notes, slide prints, scaling issues etc).

Anyway, after a few minutes on Google and StackOverflow, decided to give Deck.js a spin. Deck.js is really nice and you should take a tour if you haven't done so. So finally, after giving the online presentation and the introduction a try, downloaded the latest to give it a more thorough spin. As usual, one of the first things was embedding code snippets and I thought it would be nice to integrate Alex Gorbatchev's SyntaxHighlighter into this. Turned out really simple to do (I'm sure there are other syntax highlighter extensions to deck.js out there) - but since I got something working pretty easily, here it is:

Create a file called deck.syntaxhighlighter.js with code below:

[sourcecode language="javascript"]
(function ($) {
$("head").append(
'<link href="http://alexgorbatchev.com/pub/sh/current/styles/shCore.css" rel="stylesheet" type="text/css" />'
).append(
'<link href="http://alexgorbatchev.com/pub/sh/current/styles/shThemeDefault.css" rel="stylesheet" type="text/css" />'
);
function setupSyntaxHighlighterAutoloads() {
console.log("calling SyntaxHighlighter");
SyntaxHighlighter.autoloader(
'applescript http://alexgorbatchev.com/pub/sh/current/scripts/shBrushAppleScript.js',
'actionscript3 as3 http://alexgorbatchev.com/pub/sh/current/scripts/shBrushAS3.js',
'bash shell http://alexgorbatchev.com/pub/sh/current/scripts/shBrushBash.js',
'coldfusion cf http://alexgorbatchev.com/pub/sh/current/scripts/shBrushColdFusion.js',
'cpp c http://alexgorbatchev.com/pub/sh/current/scripts/shBrushCpp.js',
'c# c-sharp csharp http://alexgorbatchev.com/pub/sh/current/scripts/shBrushCSharp.js',
'css http://alexgorbatchev.com/pub/sh/current/scripts/shBrushCss.js',
'delphi pascal http://alexgorbatchev.com/pub/sh/current/scripts/shBrushDelphi.js',
'diff patch pas http://alexgorbatchev.com/pub/sh/current/scripts/shBrushDiff.js',
'erl erlang http://alexgorbatchev.com/pub/sh/current/scripts/shBrushErlang.js',
'groovy http://alexgorbatchev.com/pub/sh/current/scripts/shBrushGroovy.js',
'java http://alexgorbatchev.com/pub/sh/current/scripts/shBrushJava.js',
'jfx javafx http://alexgorbatchev.com/pub/sh/current/scripts/shBrushJavaFX.js',
'js jscript javascript http://alexgorbatchev.com/pub/sh/current/scripts/shBrushJScript.js',
'perl pl http://alexgorbatchev.com/pub/sh/current/scripts/shBrushPerl.js',
'php http://alexgorbatchev.com/pub/sh/current/scripts/shBrushPhp.js',
'text plain http://alexgorbatchev.com/pub/sh/current/scripts/shBrushPlain.js',
'py python http://alexgorbatchev.com/pub/sh/current/scripts/shBrushPython.js',
'ruby rails ror rb http://alexgorbatchev.com/pub/sh/current/scripts/shBrushRuby.js',
'sass scss http://alexgorbatchev.com/pub/sh/current/scripts/shBrushSass.js',
'scala http://alexgorbatchev.com/pub/sh/current/scripts/shBrushScala.js',
'sql http://alexgorbatchev.com/pub/sh/current/scripts/shBrushSql.js',
'vb vbnet http://alexgorbatchev.com/pub/sh/current/scripts/shBrushVb.js',
'xml xhtml xslt html http://alexgorbatchev.com/pub/sh/current/scripts/shBrushXml.js'
);
SyntaxHighlighter.all();
}
$.getScript("http://alexgorbatchev.com/pub/sh/current/scripts/shCore.js",
function () {
$.getScript("http://alexgorbatchev.com/pub/sh/current/scripts/shAutoloader.js",setupSyntaxHighlighterAutoloads);
});
})(jQuery);
[/sourcecode]

Throw that in your deck.js/extensions folder. IN the slide deck you want to use this extension, include a script line before the end of the body tag:

[sourcecode language="html"]
<script src="../extensions/deck.syntaxhighlighter.js"></script>
</body>
</html>
[/sourcecode]

And you're done. Now to include any code snippets in your deck, just use either of the methods specified in the SyntaxHighlighter page (snippet for the script tag below)

[sourcecode language="html"]
<script type="syntaxhighlighter" class="brush: js">
$(document).ready(function () {
//this is the function body
});
</script>
[/sourcecode]

That's all there is to it. Code above can do with some improvements (conditionally load local copies if the remote loading fails etc) - but this is just a quickie script - so feel free to modify this to your heart's content.

Vim - unmap Esc!!

So I had the bright idea (by no means original, though, as I later figured out) that it'd be great to avoid the Esc key on Vim as its so far away from the home row. The alternative to pressing Esc is Ctrl-[ which, even though I've mapped CapsLock to control, I still find hard. So then, after some more googling around I've settled down on mapping jk to Esc. Its been a few hours with this setup and while its been an absolute pain till now, I think its a great way to avoid the Esc key jump. I can already feel my finger muscle memory relearning and my hand jumps instinctively for the Esc key much less now.

Here's my setup in case you want to try this out. Bung the following into your .vimrc or _vimrc as the case may be:

inoremap  :echoe "use jk"
inoremap jk 

The first mapping makes VIM echo a reminder. Its not friendly since it introduces a pause. However, the idea is to make the Esc so painful that you will shy away from hitting it.

Upgraded to XBMC 11.0 Eden beta

So I upgraded the good ole' media center machine at home to XBMC 11.0 Beta
. XBMC has been one of those software finds that has been just marvellous - to the point where I can't imagine the telly at home without XBMC. I've pretty much stopped watching regular tv/cable and almost exclusively on XBMC.
Also, its been a great way to keep the aging laptop (circa 2006 - core2Duo 1.6 Ghz, 2 Gb RAM and a piddling ATI Radeon X1600) in active duty.
Here's a hat-tip to all the XBMC guys and gals. And if you're not running a media center at home, you should give it a spin - XBMC makes your idiot box smart!

Unit testing Apache CXF RESTful services - code available

So, the original post on the topic written about two and a half years ago had code snippets, but there's been comments and PMs for the complete code. So last week, as I resurrected this blog, decided to get that code out on github. Unfortunately, that was easier said than done; it has been quite some time and frankly, I'd lost the code. I must've switched machines about 3 times in the interim and gone from SVN to github for personal projects. Some hunting around ensued and thankfully, I was able to find the actual code we wrote based on the sample I'd posted. So cleaned that up - and just extracted the unit testing example out of it and pushed it to github - get it here. I haven't updated any of the dependencies - so this is still running against spring 2.5 and cxf 2.2.3 (I think) and things might've changed quite a bit since then (I haven't used the JAXRS bits of CXF much after that)

Running tests:
[sourcecode language="bash"]
mvn test
[/sourcecode]
Running the server:
[sourcecode language="bash"]
mvn jetty:run
[/sourcecode]

Syntax highlighting support in Wordpress.com with markdown

Now that I've cozied up to Vim/VimRepress combo for posting to this blog, there are a few things where I'm finding issues with posting code. With straight wordpress.com, I used to be able to mark up code with the [sourcecode][/sourcecode] tag and syntax highlighting comes in. With markdown - indenting a block of code with 4 spaces renders it as a <pre><code></code></pre> tags, but I don't know if there's a way to let WP.com know what language it is or any way to use the [sourcecode][/sourcecode] plugin from markdown.

Some googling on the topic didn't result in any wp.com specific answers (some folks have posted on using other plugins et cetera with a self hosted Wordpress - but nothing for wordpress.com)

Any ideas/pointers? Guess I should also post the question on the Unix stackexchange

Update on 1/1/2012

Using the sourcecode language="xxx" tag works - but you cant have any empty lines in your source.

Creating an interstitial login page with JqueryMobile

So, at work, we're building a mobile website using JqueryMobile. The app has a bunch of publicly visible pages however, other pages require the user to be authenticated. We didn't want the user to be forced to login on the first page. Instead, whenever a protected page is accessed, and if the user insn't logged into the app, we'd like to take him to the login page. Once he's successfully authenticated, then take him to the page he was navigating to. Doing this in a normal webapp is quite standard - however, with JqueryMobile, query params meddle with the hash navigation model. Also, the page that the user tries to access could be a div in the same physical page or a different url that needs to be fetched.

Trying to solve this was interesting as we were all really just getting started with JqueryMobile - so finding the ideal solution required a few tries. The solution takes a leaf out of JqueryMobile's approach. The outline of the solution is:

1. Any page div that's a protected resource is marked with a data-needs-auth="true" attribute


2. We hook into the document level pagebeforechange event to see if the user is trying to transition to a page requiring authentication. If so, then check if we have the user's authenticated context available.


3. if the said context isnt available,
   
   
   1. Cancel default event handling since we're now going to navigate the user to the login page.
   
   
   2. save the toPage object - so once the user is logged in, we know where to take him.
   
   
   3. navigate to the login page.
   
   
   
   


4. In the login page, the page can call the server apis to autheticate the user. Once the user is authenticated, then
   
   
   1. See if there's a valid returnTo object, if so, take the user to the page.
   
   
   2. If not, take the user to a 'default' page - in our case, this is the app dashboard page.
   
   
   
   

Code below
[sourcecode language="javascript"]
var pageVars = {}
$(document).bind("pagebeforechange", function (event, data) {
if (typeof data.toPage == 'object' &amp;&amp; data.toPage.attr('data-needs-auth') == 'true') {
if (!sessionStorage.getItem("TokenSSKey")) {
if (!localStorage.getItem("TokenLSKey")) {
pageVars.returnAfterLogin = data;
event.preventDefault();
$.mobile.changePage("#Login_Page", { changeHash: false });
}
else {
sessionStorage.setItem('TokenSSKey', localStorage.getItem("TokenLSKey"));
}
}
}
});
[/sourcecode]
The login event handler that handles the server response that's received once we pass the username and password

[sourcecode language="javascript"]
function SuccessLogin(data) {
if (data != null &amp;&amp; data.LoginResult != null) {
if (data.LoginResult.Code === 0) {
localStorage.setItem('UNameLSKey', data.LoginResult.User.AccountName);
if ($("#RememberMeChkBx").is(":checked")) {
ErrorPanel.html("");
localStorage.setItem('TokenLSKey', data.LoginResult.Token);
sessionStorage.setItem('TokenSSKey', data.LoginResult.Token);
}
else {
ErrorPanel.html("");
sessionStorage.setItem('TokenSSKey', data.LoginResult.Token);
}
if (pageVars &amp;&amp; pageVars.returnAfterLogin) {
$.mobile.changePage(pageVars.returnAfterLogin.toPage);
}
else {
$.mobile.changePage("#DashBoard_Page", { changeHash: false });
}
}
}
}
[/sourcecode]

Learning Vim

Is it worth it?

Definitely seems to be. I've looked at VIM in the past, tried it out too a couple of times or more, failed miserably(mostly within a day or two) and then wondered Why nutheads use VI. This would usually be followed with going back to the comfort of Emacs. I think over the years, I've spent more time customizing Emacs than actually getting any work done with it. And somewhere that felt wrong. In light of that, the minimalistic VIM looked attractive and worth another try.

So what was different about this time?

So this time things worked out a bit better. Rather than firing up VIM, spent some time reading through other's experiences on picking up VIM. And the first thing I did right was to disable the arrow keys in normal mode (I still have them in insert mode)

    " disable arrow keys
    noremap   <Up>     <NOP>
    noremap   <Down>   <NOP>
    noremap   <Left>   <NOP>
    noremap   <Right>  <NOP>

Once you have that bit, you're forced to use h/j/k/l. And while h/j/k/l muscle memory is built up within a week, the nice thingthat really happens is that you dont use h/j/k/l much - instead you move to using more efficient movement commands. There're aton of resources/cheatsheets on the web - but the approach I followed was to figure out some small keystroke when I needed it.What that mean't was that I could get work done - but at the same time get more efficient gradually.

Customizations

VIM out of the box is pretty badly configured - and that's part of the reason that people seem to shy away from it. In fact, all the times that I tried out VIM before, I didnt even come close to cusotimizing my .vim. There are folks who have curated vim dotfiles on github etc - but my advice is to stay away from them. You should know what goes in your .vim and be in control of thatrather than getting a bunch of things in your .vim that you dont understand. Just so you know, looking at the github history for my vimfiles repo - the initial commit was 3 months ago - but after that, all the commits have come in only in the last 4 weeks.What that means is that while I put in a vim file initially, I didnt do much with it initially since I was just getting a hang of the basics. Once one becomes comfortable with the basics, one moves to customizing the vim environment more and more.

Parting words

To summarize, VIM definitely seems nice once you invest into it. It's easy to drop off in the initial stage and not go any further - and I believe this is what happens to the vast majority of folks who try it out. However, once you build that initial comfort level,it feels light, fast and easy.Start easy, persist, and customize bit by bit - you'll feel yourself going from struggling with Vim to feeling comfortableand then to customizing your environment for an even better experience with VIM.I've definitely been more productive with VIM than I ever felt I was with Emacs - and these posts to my blog from Vim part of that.Besides that, I've used VIM effectively with a decent sized js code, html markup etc and felt the speed of editing inspite of still beinga noob in Vim terms.

A new look

Changed the theme of this blog and moved around the widgets a bit.
Finally, I can bear looking at this blog :) - hope that holds good for you too.

Compiling VIM

Running ubuntu 10.10 here and ubuntu repos have only vim 7.2. I'm sure there's a ppa out there that has 7.3, but thought
that compiling vim from source would be a good exercise - plus I get to compile it with the options that I'd like
rather than relying on someone's build.

Here's the options that I enabled:
[sourcecode language="text"]
CONF_OPT_PERL = --enable-perlinterp=dynamic
CONF_OPT_PYTHON = --enable-pythoninterp
CONF_OPT_RUBY = --enable-rubyinterp
CONF_OPT_GUI = --enable-gui=gtk2
CONF_OPT_FEAT = --with-features=huge
BINDIR = /usr/bin
DATADIR = /usr/share
[/sourcecode]
Here's hte other dependencies I had to install
[sourcecode language="bash"]
sudo apt-get install libperl-dev ruby-dev python-dev libgtk2.0-dev
[/sourcecode]
Once you have the deps installed, just run
[sourcecode language="bash"]
make
sudo checkinstall
[/sourcecode]

Blogging with Vim

So now I'm in Vim land and this is the first time I've gotten far enough to feel a bit comfy. Decided to dust off my blog and start at it again - what better to do it in than in VIM.

So - TA-DA - here's the first post - courtsey VIM on ubuntu. However, as usual, it was rougher than it's supposed to be. IN any case, I'll forget how I got this far the next time so the next few posts will be around recording how to get VIM to post to WP.com blogs.

But before that - the first thing to to is to get the VimRepress plugin. Better if you have pathogen installed, in which case you can do
[sourcecode language="bash"]
cd .vim
git submodule add https://github.com/raghur/VimRepress bundle/VimRepress.git
[/sourcecode]
That's my fork on Github of https://github.com/connermcd/VimRepress.git which fixes a few things:

• Makes VimRepress work properly through a proxy


• Changes the attachment filename to a '.odt' since Wordpress.com doesn't allow a text file attachment.

I still dont have a clue if doing this will break the plugin - but nevertheless, basic case of posting to my blog works and at this stage that seems good enough for me.

PS as you can see from this post - I've not yet got a hang of markdown syntax :)

Dec 29th - PPS a couple of posts and one more tip for Wordpress.com. WP.com does <br/> for hardbreaks in the markdown text. Obviously, this doesnt leave the post looking very good. I have the following in my .vimrc to get around this

[sourcecode language="text"]
augroup Markdown
autocmd FileType markdown set wrap
\ linebreak
augroup END
[/sourcecode]

PPS
You will also need to have python markdown installed once you have VimRepress running.
[sourcecode language="bash"]
easy_install markdown
[/sourcecode]

Troubleshooting nandroid backup/restore

Have been having all sorts of weird problems with Nandroid backup/restores. Essentially, here's the symptoms of the problem - I'd get a nandroid and restore it successfully (Amon RA/CWM would report success) - however either will get stuck at boot or if it boots successfully, will have tons of FCs and/or data loss. In most cases, I would dread seeing the green Android on boot up asking me to log in to my google account :(

Essentially, my nandroids were useless... to the extent that I had only one nandroid backup that was known to work - and I was keeping 4 backups of that lest I lose it somehow.

So today, thought I'd get deeper into it and see where the problem was

1. It was unlikely that its a problem with CWM/AmonRA - I myself had an old working backup. Since my backups were created and restored successfully with MD5 verification, it seemed that there's something wrong in the backup image itself.


2. Still, that seemed inexplicable, since creating images just doesn't seem that flaky. A couple of times after reboot, I had got a "UID has changed - it is recommended to wipe data" or something similar message - so I thought something was wrong with permissions after the restore. In any case, I tried the CWM menu item of fix permissions - but didnt get anywhere with that. At this point, I was desperate enought to get adb out !!


3. Now in full blown investigation mode, I didnt care if I couldnt restore my data - just wanted to figure this thing out. So restored a  "non working" backup and did a adb logcat while the phone booted... turns out that I was seeing tons of messages like so:


4. 
   

   I/PackageManager(  205): /system/app/ContactsProvider.apk changed; collecting certs
   I/PackageManager(  205): New shared user android.uid.shared: id=10006
   W/PackageManager(  205): System package com.android.providers.contacts has changed from uid: 10003 to 10006; old data erased

   
   


5. So that explained what was going wrong... thought it would be an easy fix to do the 'fix permissions' thing in CWM advanced menu. Restored again and went over to the advanced menu, did fix permissions and rebooted. What I got was a big naught for all my work - same problems and no resolutions. At this point I was stumped enough - but sheer bull headedness forced me to look at the log again... and lo it says 'data erased'. So that explains why fix permissions after boot wont work since the data is erased during boot itself!

At this point, the key to the problem was really understanding how and where android's UIDs are generated,  stored  and regenerated. Headed over to Cyanogen wiki and read up the details on fix permissions which explained the packages.xml file. Somehow the packages.xml was borked in the nandroid (every time) and that was causing it to be regenerated.

Armed with that, got a germ of a solution in place, which is roughly

1. Restore nandroid with borked packages.xml


2. Let the system boot. Will lose data but a new packages.xml will be regenerated


3. Reboot into recovery and adb pull /data/system/packages.xml out.


4. Do an advanced restore and again restore the data only.


5. mount /data and adb pull /data/system/packages.xml to compare differences. Found that packages.xml was indeed corrupt.


6. adb push packages.xml (this is the generated one pulled in step 3) to /data/system. Now you have all the old data but the packages.xml is newly generated one and known to be valid. Obviously UIDs will mismatch - but fix permissions has a valid file to work on.


7. Still in recovery, run fix permissions. It should fix permissions properly.


8. Reboot

It worked like a charm!!!! I'm still a little worried as I dont know what else is borked in my nandroid data.img. And I dont know why that image has the exact same problem every time - I have tried 3 different versions of CWM recovery, Amon RA 2.2.1 and ensured that my sd card was clean (ran chkdsk on it). In any case, since I'm able to restore, will just use the phone for the next few days and hopefully I'll run into any wonkiness soon.

And going forward, I think I'll take my own backup copy of the /data/system/packages.xml file along with each nandroid.

Note To Self: Fixing broken market links on Android after wipe/ROM upgrade

Force stop market and clear data
Launch market again - it will ask you to accept terms. Do so.
Should force it to rebuild the database and you should see all your apps linked to market again.

Note to self: ROM install/upgrade

1. nandroid backup - amon ra recovery


2. Reboot recovery, install zip


3. Install Link2SD-preinstall.zip (only on cyanogen based ROMs)


4. Boot


5. Play around...make sure things work.


6. Install other niceties/Troubleshoot




1. Link2SD - database error.Just uninstall and reinstall.


2. /etc/gps.conf - change to sg.pool.ntp.org




7. Charge to 100%


8. Reboot into recovery


9. wipe battery stats


10. reboot


11. Run down the battery


12. Recharge to 100%

 

A Python mystery

Back after a long time…saw something strange today and think it deserves a post. I was cranking through Problem 47 on Project Euler. As I was optimizing the solution, the optimization actually increased run time – and I’m at a loss to explain it. So here goes:

[sourcecode language="python"]
def problem_47(maxlen = 4):
found = False
i = 2*3*5*7 + 1
while not found:
# facs = [len([j for j in uniq(prime_fac(i+k))]) for k in xrange(0,maxlen)]
d = 4
for t in range(i+3, i-1 , -1):
k = len(list(uniq(prime_fac(t))))
if k < 4:
i = t + 1
break
else:
d -= 1
if d ==0:
found = True
print list(xrange(i, i + maxlen))
# if i % 1000 == 0:
# print i
[/sourcecode]

The run time is about 1m2s.

Now, if I try to optimize it such that I break (line 10) when I find the first number from the end that has less than 4 prime factors, the run time should be lesser (or at least the same). Right?

Turns out Wrong… now the thing takes more than 3m to run. What is going wrong?

If any of you have a clue, drop me a comment.

Fun with python’s decorators

Was in need of a utility function that can retry an arbitrary function a few times before giving up. Essentially something like Gmail or Google Readers behavior when there’s no network connection.

Thought it would be a few minutes job to cook up a decorator utility in Python. Boy! was I wrong! I mean, the basic use case is definitely trivially easy with Python – however, once you want something that’s more useful than that and resembles something that you’d actually use in production, the complexity goes over the top!

Anyway, I’m figuring out all sorts of fun things about decorators – and all of it the hard way! OTOH, its a  lot of fun to write small test code to test & validate assumptions!

Make no mistake – I’m still a python fanboy :) – just that going through some pains with decorators right now. Will follow this up with a longer/detailed post that may have some useful insights I’ve gained till then. Thanks for stopping by!

Back after a long time…

Obviously, I’m not writing enough out here… part of the reason being even though wordpress’s web editor is great, I really like not having to type gobs of text in a text area.

So eventually, looked around and found Windows Live Writer. Its going out on its customary spin :).

So what’s been cooking? Actually a bunch of things over the last several months:

Stuff – on which I mean to put up individual posts

1. Had a fun exercise benchmarking lighttpd with python wsgi
2. Been doing some stuff on mysql cluster – mostly around seeing how it compares with the mysql master-master replication setup I had in place.
3. Dipping my toes into Amazon EC2 finally – though Linode or Rackspace is way easier if you want to just spin up a VM. Amazon’s  EC2 does have some interesting stuff (reliability of back ups, CDN etc). However, it comes at the cost of having a model that initially is hard to understand.
4. Resin server – heard good things about it, had to see if it would fit at  some stuff at work. Disappointed that the free version is really hamstrung.
5. Apache Wicket: I’ve always hated web UI and somehow the action oriented frameworks (Struts and their ilk) never appealed from a coupling/cohesion standpoint. In that respect, seemed like ASP.net got a lot of things right going the component oriented way. However, it seems fatally flawed with stuff like viewstate and postback and so on. On the Java end, tried Tapestry out, but, it comes with too much baggage for my taste. Had been reading of Wicket for sometime now and decided to take the plunge and was pleasantly surprised doing my contrived example:
1. Took much less to get off the ground compared to Tapestry
2. Mentally, a lot easier to understand
3. Managed to realize my goal of exploiting OO techniques to DWIM – even on a simple contrived example.

Books:

1. Steve Souders excellent "High performance websites” book: if you’re doing anything near a high performance website, then grab this book today!
2. Wicket In Action
3. Agile Principles, Patterns and Practices by Robert C Martin: read about the SOLID principles first and then buy this. This is a book to own if you aspire to become a good Agile/OO practitioner. Don’t worry about the C# in the title – it applies universally.

A new tool for the toolbox!

Firstly - my VM setup:

I'm running Virtualbox with Xubuntu 9.10 on Win7 host - and its pretty. Its on a office standard issue Dell D531 - meaning they're AMD Turion X2 TL-60 and 2GB of RAM.

Now the Turion's supposed to have hw virtualization (AMD-V) however, the moment hw virtualization was enabled in virtualbox and I tried starting the vm, the machine would hard reboot!!!

After searching high and low, turns out that its an issue with Dell bioses and they dont have any updates. Here's a page that tracks the issue. Imagine my happiness when a couple of days ago, found that dell had released an unofficial bios update (T12).Well, its gone in, and things are running swimmingly well - my VM now has 2 procs, is stable and I hardly feel I'm in a VM :). In fact, this post is coming from the VM  - firefox with 12 tabs, a few terminals and emacs running on 600 MB of RAM.

Now let me come to the new tool I was talking about

I like to run the VM full screen - feels best that way. After trying out enough and more of multiple desktop softwares, have finally settled on VirtuaWin - beats the crap out of other tools, systray integration is great, has window rules and so on. Over the past couple of weeks, its come close to the ideal tool - does the job well and you hardly know its there :-)