I bought abhiomkar.in domain about 2 years back at Godaddy. And mapped it to Webfaction servers.

My new sub-domain for my blog: http://blog.abhiomkar.in

I will keep my new blog as active as possible. Please subscribe to my blog if you haven’t via Feedburner

See you there,
Abhinay Omkar
@abhiomkar

Question:
Given the following javascript code:

function countdown (num) {
        for (var i = 0; i <= num; i += 1) {
            setTimeout(function () {
                alert(num - i);
            }, i * 1000);
        }
    }

    countdown(5);

The desired result is a countdown from 5 to 0 using alert messages. Explain why the code only alerts -1, then fix the code so it works as expected.

Answer:
Let’s debug the code:
Debugging:

function countdown (num) {
    for (var i = 0; i <= num; i += 1) {
            setTimeout(function () {
                console.log("\nnum = " + num);
                console.log("\ni = " + i);
                alert(num - i);

            }, i * 1000);
        }
    }

countdown(5);

Output:
num = 5
i = 6
num = 5
i = 6
num = 5
i = 6
num = 5
i = 6
num = 5
i = 6

Hence, the code in question always alerts ‘-1′.

My fix for the above problem is:

function countdown (num) {
        for (var i = 0; i <= num; i += 1) {
            setTimeout(function () {
                if(i > num){
                    i = 0;
                }else{
                    i += 1;
                }
                alert(num - i);

            }, i * 1000);
        }
    }

countdown(5);

Explanation:
The alert box gives -1 every time (6 times), because the in-line function which has the alert() statement gets executed only after the loop is completed. By the time the loop is completely executed the value of ‘i’ will be 6. The program continues to execute and loop until the i<=num condition is satisfied. The in-line function which has alert() statement gets executed only after the specified time (i*1000) (that is, 0, 1000, 2000, 3000, 4000, 5000 in milliseconds) for six times with 1 sec delay for each alert box. So, The execution of loop gets'll be executed already before the first trigger of setTimeout().

So, the code is modified to reset the 'i' value to 0 when it is '6' and then keep increasing the 'i' value before alert() statement.

It was fun! :D

Note:
Note that, most of the browsers have a minimum delay set for setTimeout() call. Such as, 4ms to 10ms of minimum delay is set (depending on the browser) even though the timeout value (second parameter) is mentioned as ’0′ (zero). Check this out for more details – https://developer.mozilla.org/en/DOM/window.setTimeout

I was inspired by the write up in the book called  ’Learn Python The Hard Way’, I didn’t read the whole book but, this page caught up my attention. It’s inspiring for programmers, so thought I would share with you all. Please drop me a comment if you do like it :)

gist: 398604 – GitHub

Explanation:

I’ll walk through the code to explain how it works.

We need urllib and urlparse python modules so we import them -

from urllib import urlopen, unquote; from urlparse import parse_qs, urlparse;

youtube_watchurl is a self explanatory, we store the Youtube Watch URL here -

youtube_watchurl="http://www.youtube.com/watch?v=NeSuirvA6UE&playnext_from=TL&videos=MS3Hq4oBj08"

We need the v parameter value from the above Youtube Watch URL (Example: v=NeSuirvA6UE )

urlparse(youtube_watchurl).query returns the query part of the above URL i.e., “v=NeSuirvA6UE&playnext_from=TL&videos=MS3Hq4oBj08″. We need the value of ‘v’ parameter from this query string. parse_qs() function returns parameter name and value in dictionary format. parse_qs(…)['v'][0] returns the value of ‘v’ parameter.

video_id = parse_qs(urlparse(youtube_watchurl).query)['v'][0];

The below GET request will retrieve all video response like video title, description, tag, view count, rating etc and most importantly it also includes token string.

Just try

curl "http://www.youtube.com/get_video_info?&video_id=NeSuirvA6UE"

to see the response from youtube. We need to parse the token string from the above response.
 

urlopen('http://www.youtube.com/get_video_info?&video_id=' + video_id).read().decode('utf-8')

Similarly, as we have retrieve ‘v’ parameter value we retrieve the ‘token’ parameter value using below piece of code.

parse_qs(unquote(urlopen('http://www.youtube.com/get_video_info?&video_id=' + video_id).read().decode('utf-8')))['token'][0])

example token value:

vjVQa1PpcFNWkxk6GpwXYrIxVZ-1iDcF_E9emDZ4UKA%3D

Now we have video id and token string, let’s put these value into the direct url to download the video. The format of the Direct URL:

http://www.youtube.com/get_video?video_id=video_id&t=token&fmt=18

Example Direct URL:

http://www.youtube.com/get_video?video_id=NeSuirvA6UE&t=vjVQa1PpcFNWkxk6GpwXYrIxVZ-1iDcF_E9emDZ4UKA%3D&fmt=18

The above URL redirected me to below URL

http://v12.lscache4.c.youtube.com/videoplayback?ip=0.0.0.0&sparams=id,expire,ip,ipbits,itag,algorithm,burst,factor,oc:U0dWSFZUVl9FSkNNNl9OTlpB&fexp=903801,903206&algorithm=throttle-factor&itag=18&ipbits=0&burst=40&sver=3&expire=1273698000&key=yt1&signature=5F8E3C3EABC00FE77AB5699811B13957CA95CFFD.14D37482C6E22D91D08770C6630EE3BE3BF9732D&factor=1.25&id=35e4ae8abbc0e941

And that’s it we use the above direct link to download and write it to a file

open(video_id+'.mp4', 'wb').write(urlopen("http://www.youtube.com/get_video?video_id=%s&t=%s&fmt=18"%(video_id, parse_qs(unquote(urlopen('http://www.youtube.com/get_video_info?&video_id=' + video_id).read().decode('utf-8')))['token'][0])).read())

Here is the complete code:

gist: 398604 – GitHub

Inspired from Downloading YouTube Videos with a Perl One-Liner - http://www.catonmat.net/blog/downloading-youtube-videos-with-a-perl-one-liner/

and

Stackoverflow: http://stackoverflow.com/questions/2678051/cant-download-youtube-video/2679042#2679042   (Coded in Python 3)

As of now It doesn’t require any non-builtin python modules to run this script except flickrapi for python (Get it from here – http://stuvel.eu/projects/flickrapi). You just need a Python. You can suspend the download in-between & resume download later. I code it later for EXIF data.

Give a try and let me know your suggestions :)
Code is at Github – http://github.com/abhiomkar/FlickrBird

Run flickrbird.py

Screenshot running FlickrBird on Mac OS X - Flickr Backup Tool

How to use on Mac OS X / Linux?

Open Terminal and enter below commands

mkdir MyFlickrBackup
cd MyFlickrBackup
#Download FlickrBird Script
wget http://github.com/abhiomkar/FlickrBird/raw/master/flickrbird.py
#Give execute permissions to run it
chmod +x flickrbird.py
#Yah! run it...
./flickrbird.py

I just wanted to post a blog on Homebrew because, I started loving it already. Homebrew is the new modern package management system for Mac OS X wich beats MacPorts. Yah! Homebrew is faster.

Homebrew is a Packaging system for Mac OS X 10.5 and above; heavy optimisations, no redundant packages and a bonus beer theme.

Here how you can install Homebrew on your Mac:

I prefer installing homebrew in /usr/local directory.

sudo curl -L http://github.com/mxcl/homebrew/tarball/master | sudo tar xz --strip 1 -C /usr/local


to avoid using sudo when installing packages, change the owner recursively on /usr/local directory

sudo chown -R `whoami` /usr/local


That’s it. Homebrew installed.

Now, Let’s install wget using homebrew

MacBook:~ abhinay$ brew install wget
==> Downloading http://ftp.gnu.org/gnu/wget/wget-1.12.tar.bz2
File already downloaded and cached to /Users/abhinay/Library/Caches/Homebrew
==> ./configure --disable-debug --prefix=/usr/local/Cellar/wget/1.12
==> make install
/usr/local/Cellar/wget/1.12: 9 files, 784K, built in 37 seconds


Really easy! isn’t it?

Have fun!

Note:
If you are facing problem because of MacPorts, try to uninstall it – Uninstall Instructions

and Do check out homebrew homepage on github: http://github.com/mxcl/homebrew

And one more thing, You should have xcode installed on your Mac – if already don’t have xcode, you can download from here – http://developer.apple.com/technology/xcode.html (Or you can install xcode from your second installation CD of Mac OS X)

WoW! My blog crossed 50,000 Views. Thanks to all of you! Here is my new year greetings to you! :)

Yah, that’s my Firefox looking just like Google Chrome Browser! Running on Ubuntu 9.10 Karmic.

Simple – I used, Chromixfox, Total ReChrome to customize my Firefox to look like Chrome browser.

And also, changed the tabMaxWidth value in about:config to 175

Checkout my Firefox Addon Collection – Firefox Addons that I use -

https://addons.mozilla.org/en-US/firefox/collection/e2a5c8ad-c048-eb04-05de-586ce207f773

^ Must have Firefox Add-ons according to me ;)

What do you need to do this?
Ubuntu CD (I used Ubuntu 8.10 Desktop Edition Install/Live CD)
Mac
Internet Connection (Wired)

Few days back I had installed Debian 5.0 lenny on my MacBook. Because, I don’t have much hard disk capacity (only 80 GB), I had to delete the Linux partition (waiting for Ubuntu 9.04).  I used BootCamp Utility (in Mac) to delete the Linux partition. It wipes off all the linux data & restores Mac partition (ex: I had 60GB Mac + 20GB Linux, After restore 80GB* Mac).
I was happy that it was restored. But, the partition had errors after restores. I was not able to copy/store the any data if it exceeds 60GB on Mac, it says the Mac startup disk is full. Then I realized that the Mac file system (i.e., HFS+) had errors. Yes, the Ubuntu CD is there to rescue me. Enough story… follow this to proceed…

1. Disable journaling on Mac parition before booting into Ubuntu CD
Use this command on your Mac terminal

• sudo diskutil disableJournal /dev/disk0s2

where disk0s2 is your Mac partition (use ‘df’ command to find out the root disk name, it should be mounted on ‘/’)

2. Boot from Ubuntu CD
Insert the Ubuntu CD, press ‘C’ on startup to boot from CD.

3. Connect to Internet & Install HFSprogs
After Boot connect to Internet, Yeah! I know the wireless won’t work now. I suggest you to connect from Wired.
I assume that you are connected to Internet now.
Then enable Universe packages using ‘Software Sources’.
Then, Install hfsprogs package

• sudo apt-get install hfsprogs

This will enable you to check & repair the HFS+ paritions.

4. Check & Repair

You need to do fsck check on your Mac partition. Use fdisk command to find the Mac device.

abhinay@ubuntu:~$ fdisk -l
WARNING: GPT (GUID Partition Table) detected on ‘/dev/sda’! The util fdisk doesn’t support GPT. Use GNU Parted.

Disk /dev/sda: 80.0 GB, 80026361856 bytes
255 heads, 63 sectors/track, 9729 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0×00000000
Device Boot Start End Blocks Id System
/dev/sda1 1 9730 78150743+ ee GPT

using fdisk didn’t show the Main Partition of mac, It shows only one partition. The above partition sda1 is the EFI partition (boot partition).
That means, /dev/sda2 is your Mac partition.

Perform fsck on sda2

abhinay@ubuntu:~$ sudo fsck.hfsplus /dev/sda2
** /dev/sda2
** Checking HFS Plus volume.
** Checking Extents Overflow file.
** Checking Catalog file.
** Checking multi-linked files.
** Checking Catalog hierarchy.
** Checking Extended Attributes file.
** Checking volume bitmap.
Volume Bit Map needs minor repair
** Checking volume information.
** Repairing volume.
** Rechecking volume.
** Checking HFS Plus volume.
** Checking Extents Overflow file.
** Checking Catalog file.
** Checking multi-linked files.
** Checking Catalog hierarchy.
** Checking Extended Attributes file.
** Checking volume bitmap.
** Checking volume information.
** The volume Macintosh HD was repaired successfully.

Hurray! It is repaired now…!

One more last thing you ‘ll have to do.

5. Enable the journaling on Mac Partition
use this command on Mac to enable the journal back.

sudo diskutil enableJournal /dev/disk0s2

You are done! :)

**This worked for me!

Comment me if you have any troubles or suggestions. Thanks for reading my blog!

But, I hope that this helps!