Analysis of the Duqu Trojan worm by Kaspersky Labs

This summary is not available. Please click here to view the post.

Computer hardware test/analysis tools

Today I happen to look into the causes for over-heating of my Dell Studio 14 (1458) laptop (i7, 720QM, 4GB RAM, Win 7 with ATI Mobility Radeon  HD 5450 - 1GB) and came across these various tools which people use to test/analyze their hardware.

ThrottleStop
CPUmark
CPU RightMark
Furmark
ORTHOS
HWMonitor
RMClock - Undervolt processor.
GPU-Z

I am yet to use any of these, but I do hope to run them soon and post some results here, JFK. :)

Changing the start up directory of command prompt -- The safe and simple way

By default, the command prompt window (cmd.exe) starts in a particular directory which depends on quite a few factors. Specifically, the env varibales %HOMEDIR% and %HOMEPATH% matter the most or these are the ones which ultimately decide the location. There are some registry values also, but they are not in the game by default. This leads to the command window to start up in something like C:\Documents and Settings\<username>\ which is not particularly useful as this location is rarely used for anything useful. It can get worse. In a corporate environment it might so happen, more often than not, that your home directory is set to a shared network drive using a group policy and the command prompt starts in that remove drive location.. ! The pain can be aggravated if you are on VPN or something similar.

I personally feel that command prompt, which is mainly programmers tool, should not start in %HOMEDIR% as there are rarely (mostly never) any programming related files are kept. Anyways, there are many ways to solve this problem.

The first of them and the most dangerous is fiddling with the registry. It is mentioned here : http://windowsxp.mvps.org/autoruncmd.htm
The problem with this is that it will screw up make based build environments which spawn multiple child shells (aka command prompts), because the command prompt will start in this changed default location instead of the location where the make had to run.

The next is fairly simple and also elegant. You can create a shortcut to the main exe (C:\Windows\system32\cmd.exe) and in the properties of the shortcut you can provide the directory to start in. This good for mouse users. However for those (most programmers) who start command prompt from the run dialog (by typing cmd) this will fail.

The third solution is to deal with the previous solutions shortcoming. You can just create a batch fail in any of the locations where the run dialog looks into. I prefer C:\Windows\systme32\. Put the following line in the batch file: C:\Windows\system32\cmd.exe /K "cd <path-to-dir>" . Just replace the <path-to-dir> with the path where you want the command to start. I generally put it as C:\.
This will start a command window and execute cd <path-to-folder> and will stay for further inputs. I have named this batch file as sh.bat (obviously to get a linux feel :P ). So now when I press the Windows + R key and type sh, I get the command prompt started in C:\.
Done.
And yes, this is totally safe and will not affect any other application using cmd.exe. :)

Vim -- Restoring cursor from previous session

I am sure every Vim user needs this. Its just so frustrating to open a source file to see the #includes (the first few lines) in the file when we actually will be editing many hundred lines later. Today specially I was juggling with several source files, adding something to the .h file and then come back, do something in the .cpp file and then again go to some .c file and so on. Every time I opened a file the cursor was at the first line and every time invariably I had to search for the function I was editing and cycle through the matches to reach it. So I set on a "Search Mission" - A mission to search for the appropriate .vimrc settings to make Vim remember the cursor position from the previous session.

I got a lot of links. In fact there is a separate Vim Tip - Vim Tip #80 for this. But it has so many code lines and I was a little wary to put all that in my .vimrc file. Continued search revealed me a simpler way. Just two lines solution and it is here : Some Mr.Gopinath's .vimrc file. Its very big, but the lines concerning me are:

" VimTip 80: Restore cursor to file position in previous editing session
" for unix/linux/solaris
set viminfo='10,\"100,:20,%,n~/.viminfo

" only for windows [give some path to store the line number info]
"set viminfo='10,\"100,:20,%,nc:\\Winnt\\_viminfo
au BufReadPost * if line("'\"") > 1 && line("'\"") <= line("$") | exe "normal! g`\"" | endif

Looks like he also picked this up from the same Vim Tip #80 but was smart enough to take out only the necessary part. Nevertheless, this works for me. Thank you Mr. Gopinath.

Happy Vimming :-)

Edit:

I read the Vim Tip #80 again and it made more sense this time. I picked up the last line for a user comment. Now the cursor is put back on the same column too.

Recursively rename files in windows

My friend Prasanna B P asked me to solve a bunch of problems that he was facing with his computer. One of them was fairly common and there was a very easy and straight forward brute force solution. But coming up with a "smart" was really interesting.

Essentially some malware had renamed all the movie files that he had to carry a .jpg extension. His initial solution was to associate those jpg files with a movie player like Mplayer123 or VLC and the file would be invariably played without any regard to the extension. But this made the genuined jpg pictures to be opened with the player.

The obvious solution was to rename all the files and change the extension. Manually doing it from the GUI is the brute force idea that I earlier mentioned. And it is a totally crappy one. Next I can rename all files in a directory from the command line. But here the movie files were in directories of their own and hence I would have to move to each directory manually and run the rename command. This makes it as good as the GUI approach. In fact the extra effort of typing the commands might make it worse.

The I searched the internet a little and got to know that the windows command shell supports a "FOR" statement which can be used to recursively traverse directories, amongst many other things it provides. Using that I found this command:

FOR /R %x IN (*.jpg) DO ren "%x" *.avi

from this website : http://stackoverflow.com/questions/210413/command-line-recursive-renamemove-in-windows

People used to Linux Shell scripting might think of this as a wierd syntax, but yeah thats the windows choice.

It was good to do some Windows Shell scripting too. :-)

Making firefox use a little lesser memory

A lot people tell me that firefox uses a lot of memory and it grows like anything when used for a long duration. From the discussions developers and some blog posts/forums it appears the prior to FF3 it was memory leaks which formed a major part of this. Apart from that one of the important features "bfcache - Back/Forward Cache" which, arguably, makes back and forward navigation very fast, uses a lot of memory. In on the pages I read apparently it can be up to 4MB per page. This cache essentially keeps the parsed HTML of the mark-up in the memory along with the Javascript state. Sometimes, or more like most of the times, this can very taxing. You might be the kind of person who does not really uses the Back-Forward navigation and in case you do it you are OK with a slight delay (which may not percievable with a good conenction). If indeed you are that kind then disabling this "bfcache" will probably make your firefox eat lesser memory.

The setting/preference that controls this behavior is : browser.sessionhistory.max_total_viewers

By default it is -1, which is no limit. Setting this to 0 (zero) will disable this cache completely. Essentially you are telling Firefox to not to store the state of any document in the memory. If you want to have this feature with a saner limit you can set it to an integer representing the number of pages you want to be saved in memory.

Like any other pref not exposed through the Options dialog this has to be edited by visiting the page about:config. Open this in a tab and key in (or copy and paste) the preference string. Double click it to edit.

May be you can now lose lesser hair as your other applications will run smoothly. :P

Hari Om.

Undo "Always display images" switch in Gmail

Gmail has this nifty and very useful feature to avoid mostly annoying images being displayed in our incoming emails. Whenever we get an email containing an embedded image, it is not displayed by default. Instead there appear two links at the top of the email saying:

"Display images below"
"Always display images from <this sender's email>"

The first one is a one time thing and the second one is a setting for that particular email id.

I once clicked on the second link accidentally and had unwanted images cluttering my email view. I looked for this thing to be turned off in the settings page, but found nothing. I had a very hard time undoing this accidental setting. Finally I discovered the solution also accidentally. The switch to undo this setting is "hidden" under the "Show Details" link for the email header.

The right end of the very first line of the email, which contains the sender's name/email id", has this "Show Details" link. This shows the email header with several details. At the end is the link saying something like this:

"Do not display images from this sender".

Just use this switch to go back to the good old times of emails without images. :)

An official full-fledged Firefox Add-ons dev guide

Having done a little work on Mozilla Firefox and extensions for it I have always felt the lack of an official, step by step document for extension development. I do accept that there are a lot of resources available on the internet. Innumerable number of blogs and tutorials explaining the process step by step. But most of them become outdated with newer versions of Firefox and the authors are not really keen about updating the info. That is why an official tutorial or guide from Mozilla itself would be the right thing. It will make sure the contents in the guide are up-to-date and worth reading for developing an extension for the currently available version of Firefox.

There are documents for extension development on MDC like the various links present at: https://developer.mozilla.org/en/Extensions but it was either scattered or pretty brief in most places though it did touch most parts of extension development. Nevertheless a wholesome "official" guide was needed and here it is now. I went through the guide and as the blog post says it is still in BETA with a lot of "TO-DO" tags in there. In spite of that it is very much usable. Do visit it and post back your feedback to make it a much better guide.

Add-ons Blog » Blog Archive » Firefox Add-ons Developer Guide (beta release) - Calling all Add-on Developers!

Lets hope we will have more and more quality extensions now. :-)

Changing directory in which command window opens

For some reason my command prompt always opened up on my mapped network drive. That was probably because the network drive was set as my home or something. It was the sys-admins work about which I did not have any clue.

Every time I had to change to my local disk drive and then navigate to the directory I wanted. Also the network drive caused a lot of delay when it was not available, making my PC freeze at times and there by making me very irritated. So I finally decided to get rid of the network drive. But I couldn't and I had to settle for changing the default location in which the command window opens. The link below has all the details for doing the needful. As usual with windows its a registry value :-)

HKEY_CURRENT_USER \ Software \ Microsoft \ Command Processor
              |
              |
              |____ Autorun =

How to change the default startup directory for Command Prompt?

BTW, It mentions that this setting will render the "Open Command Window Here" XP PowerToy. Nothing like that happened to me, at least not yet. So you can be assured that both of these will work in harmony.

Happy Commanding. ;-)

Documentation appears by itself -- from the code.

Beautifully Documented Code at Toolness

I am slightly into web and web related development and as part of that I come across various JavaScript (JS) libraries which  claim to make my life a lot easier and many of them actually do that. But several times there is a big hurdle that I have to cross before actually things become easy. Its understanding how to use that library. This sometimes turns out to be a pain, often bigger than the one if I had done everything by myself without any external library.

This is probably all going to change. The lack of documentation might just vanish and all the necessary documentation might start appearing from the code itself. And the steps for building this, collating things and publishing on website are taken care of. You just need to annotate your code with the right tags and the JavaScript linked at the top (Written by Atul) will do the documentation generation and will also display it alongside code.

This is particularly useful for libraries providing various APIs as the users can see a function in the raw code and understand its meaning and purpose and the arguments it expects and its return value. All in once place at one time.

This is really cool. Go check out the tool and give the world a better library to use :-)

Happy (auto)Documenting

Editing remote files smoothly in vim on windows

I have a laptop and a desktop. Desktop runs Ubuntu, by choice, and laptop has to run Windows, by force. But most of my work happens on the Linux desktop itself. So when I am away from the desktop I would login to my desktop via SSH using PuTTy. It works fine when I am still on the corporate LAN, but the problem starts when I go home and get on to VPN. The speed and the responsiveness simply demotivates me and I tend to waste a lot of time specially when I am editing files with vim, because every keystroke has to travel across the network to my desktop and the response is to be sent back to my laptop. Coding really becomes hell with this.

Recently I got to know that VIM has identified this problem quite some time back and has a solution in store. You can open a remote file over SCP, where in VIM would bring down that file to the local system and store in some temp location. You edit that temp file, with VIM running on your own machine. So do not have to wait for keystrokes to be processed by the remote machine. When you write the file, VIM updates the remote file using SCP.

[Note]: If the file is read only, w! will not work. The user account used for SCP must have write permisson for the file you are editing. Otherwise, obviously, remote write fails and VIM will promptly report it.

Look at this page for the syntax and more details.

This would be straight forward on a linux box as both vim and scp come packed with the OS and they are in the shell execution path and everything is set up by default. Things need some extra work on Windows.

First obvious thing is to install VIM. Then you would need a SCP program. And once again PuTTy comes to rescure. They have a PSCP.exe, which makes you feel at home even on Windows. Get it here.

To improve this a little more you can rename the PSCP.exe to scp.exe and place it in "C:\Windows\System32\" so that it will be picked up from everywhere at the command prompt. Also note that you can use your PuTTy saved sessions directly with PSCP.

Happy remote VIMming. :-)

Hari Om

How libffi actually works?

I came across this libffi when I thought of working on js-ctypes. Though my contribution remained very close to nil, I got to know about this fantastic new thing libffi. But I did not understand how it actually worked when I first read. I just got a high overview and assumed that the library abstracts out several things that I need not worry about when making calls across different programming languages. I just followed the usage instructions and started looking at the js-ctypes code. That was sufficient to understand the js-ctypes code. But today when I was once again going through the README, I actually came across one line that made things a little more clearer. The first paragraph in "What is libffi?" tells it all. Its the calling conventions that this library exploits. I had written a post about calling conventions some time back. As mentioned in the readme file, its the calling convention whcih is the guiding light for the compiler. So when generating the binary code, thecompiler will assume will that the arguments that are being passed to that function will be available in some place and also it knows about a place where the return value will be kept, so that the code from the calling function will know where to look for it.

Now we know that the ultimate binary code that we get after compilation is the machine code. So it is basically a set of machine supported instructions. These instructions are certainly not as sophisticated as those available in the C - like the functions. So what do these functions transalte to? Nothing but a jump of IP(instruction pointer) to an address where the binary code of that function is. Inside this code, the arguments passed are accessed by referencing some memory location. During compilation the compiler will not know the precise memory locations (obviously). But the compiler has to put in some address there when generating the code. How does it decides on the memory location? This is where the calling conventions come into picture. The compiler follows these standard steps. Since its the compiler who is generating the code for both calling a function, where arguments are sent, and the code of the called function, where those args are received, the compiler will be knowing where it has put the arguments and hence generate code to use the data in those locations. From this point of view, the main(or is it the only one) condition for the binary code of a function to work properly is to have its arguments in the memory locations it thinks they are in and and the end place back the return value in the right location. And from what I have understood till now, it is this point that libffi exploits.

When we are forwarding calls from an interpreted language, like JS, to some binary code generated from a compiled language, like C, there are two things to be done:

0. As discussed earlier, whatever arguments are passed from JS are to be placed in the right locations and later take the return value and give it back to JS.
1. Type conversions -- The types used by JS are not understood by C. So someone should rise up to the occasion and map these JS types to their C counterparts and vice-versa.

The first of these is taken care by libffi. But the second one is very context specific and totally depends on the interpreted language. It does not make sense to just have a catalog of type convertors for every interpreted language on this earth. So these two steps were separated and spread out as two interacting layers between the binary code and the interpreted language code. libffi now takes care of the calling convention part and making sure the binary code runs and gives back the results. And the job of type conversion is the job of another layer which is specific to the interpreted language which wants to call into the binary code. And hence we have these different type converting layers for different interpreted languages: ctypes for python, js-ctypes for JS (and probably some more exist). Well with this renewed and clearer understanding I hope to actually contribute to js-ctypes.

Happy calling (with conventions) ;-)

Discussion with biesi, bz and gavin about channel and tabId in #developers

brahmana wonders if biesi was able to examine the UML diagrams

[ 8:24 pm] <biesi> brahmana, sorry not yet
[ 8:24 pm] <brahmana> thought so..
[ 8:24 pm] <brahmana> biesi, anyways another quick question.
[ 8:25 pm] <biesi> brahmana, yes?
[ 8:26 pm] <brahmana> biesi, Can you please have a look at this one: http://wiki.mozilla.org/images/a/a0/Brahmana_URI_Loading_DocShell_Code.jpg -- Which would be the ideal point in the sequence to associate the tabId to the channel created.
[ 8:26 pm] <brahmana> ?

[ 8:26 pm] <brahmana> biesi, I want to know where I will have access to both of them.
[ 8:27 pm] <timeless> brahmana: pretty

[ 8:27 pm] <biesi> brahmana, um

[ 8:28 pm] <biesi> brahmana, nothing in that diagram has access to both :-)
[ 8:28 pm] <brahmana> timeless, thank you.. more here: http://wiki.mozilla.org/User:Brahmana/Netwerk_Docs (just in case)

[ 8:29 pm] <brahmana> biesi, ok.. so how much prior to webBrowser will I have to go to get a tabID ?
[ 8:29 pm] <biesi> brahmana, tabbrowser.xml

[ 8:29 pm] <biesi> its loadURI function or something like that

[ 8:29 pm] <biesi> brahmana, of course some things call loadURI on the web navigation directly...

[ 8:29 pm] <brahmana> biesi, yeah.. thats how I created that sequence diagram..


[ 8:30 pm] <brahmana> biesi, Can't I get hold of the tabID in C++ ?
[ 8:30 pm] <biesi> C++ has no concept of "tab"
[ 8:31 pm] <brahmana> But the <browser> present in each tab corresponds to one nsIWebBrowser, isn't it?
[ 8:31 pm] <brahmana> <browser> == the xul browser element

[ 8:35 pm] <biesi> brahmana, no
[ 8:35 pm] <biesi> there is no nsIWebBrowser in firefox
[ 8:36 pm] <biesi> that's only used for embedding
[ 8:36 pm] <timeless> unfortunately we don't use the same apis everywhere :(
[ 8:37 pm] <brahmana> oh man..
[ 8:37 pm] <biesi> brahmana, there is one docshell per tab
[ 8:37 pm] <brahmana> then the starting of my sequence diagram is wrong..
[ 8:37 pm] <biesi> if that helps you
[ 8:37 pm] <brahmana> yeah.. I am aware of that..
[ 8:38 pm] <brahmana> and I thought it was nsWebBrowser that held a reference to a docShell and made calls on the docShell
[ 8:38 pm] <biesi> ah, no
[ 8:38 pm] <brahmana> But as it appears firefox does not use nsWebBrowser itself...
[ 8:38 pm] <biesi> the browser holds the docshell directly, I believe
[ 8:39 pm] <brahmana> you mean the xul browser ?
[ 8:39 pm] <biesi> yeah
[ 8:39 pm] <brahmana> What C++ object does that map to?
[ 8:39 pm] <brahmana> something under widgets?
[ 8:39 pm] <biesi> the xul browser?
[ 8:39 pm] <brahmana> yeah
[ 8:40 pm] <biesi> um
[ 8:40 pm] <biesi> some xul magic
[ 8:40 pm] <biesi> nsXULElement.cpp perhaps
[ 8:40 pm] <brahmana> oh.. let me see
[ 8:40 pm] <biesi> via the boxObject maybe?
[ 8:41 pm] <biesi> but note that the <browser> is mostly an XBL thingy
[ 8:41 pm] <brahmana> oh man.. this is getting heavily complex..
[ 8:41 pm] <timeless> it really is

[ 8:43 pm] <brahmana> Now the JS call: browser.loadURI() will be a call on the corresponding nsXULElement object, which actually holds a reference to the docShell. Is that right?
[ 8:45 pm] * brahmana requests to put aside the XPCOM stuff that happens in the above sequence..
[ 8:45 pm] <brahmana> sorry, the XPConnect stuff..
[ 8:46 pm] <gavin|> yes
[ 8:46 pm] <gavin|> though the nsXULElement isn't really involved
[ 8:46 pm] <gavin|> apart from being associated with the JS object that implements the XBL methods
[ 8:48 pm] <brahmana> gavin|, Can you please elaborate a little on your last statement..
[ 8:48 pm] <brahmana> ?
[ 8:49 pm] <brahmana> Or is there a doc that I can read up to orient myself a little before asking lots of questions here?
[ 8:49 pm] <gavin|> nsXULElement itself doesn't have anything to do with the XBL implemented methods
[ 8:50 pm] <gavin|> it's just a "container"
[ 8:51 pm] <gavin|> it's not really useful to say that you're interacting with a nsXULElement, because you're really interacting with an XBL bound node
[ 8:51 pm] <gavin|> and the XBL <browser> methods implemented in JS are what matters
[ 8:51 pm] <gavin|> not the nsXULElement class methods
[ 8:52 pm] <brahmana> oh.. ok. so the browser.loadURI() is (most probably) implemented in the JS itself. This JS object holds a reference to the docShell directly and thats how the calls are routed -- makes sense?
[ 8:53 pm] <gavin|> that's about right
[ 8:55 pm] <brahmana> now this XBL/JS implementation is present in tabbrowser.xml?
[ 8:56 pm] <gavin|> and browser.xml, yeah
[ 8:56 pm] <gavin|> the tabbrowser contains <browser>s

[ 8:57 pm] <brahmana> This is the one I should be looking at: http://mxr.mozilla.org/mozilla-central/source/toolkit/content/widgets/browser.xml , right?

[ 8:58 pm] <gavin|> yes


[ 9:04 pm] <brahmana> gavin|, ok.. i figured the exit point to docShell: http://mxr.mozilla.org/mozilla-central/source/toolkit/content/widgets/browser.xml#186 --
[ 9:04 pm] <brahmana> Now I want to get the tabId in which this browser object is, how would I achieve it?

[ 9:04 pm] <gavin|> brahmana: not sure what you mean by "tabId"
[ 9:05 pm] <brahmana> tabIndex, the index of the tab in the tabContainer
[ 9:05 pm] <gavin|> you have a reference to a <browser>, and want to find which tab it's in?
[ 9:06 pm] <brahmana> yes..
[ 9:06 pm] <brahmana> well actually I am inside the browser's definition itself..
[ 9:06 pm] <gavin|> I guess you you need to loop through tabs and compare against their .linkedBrowser
[ 9:06 pm] <gavin|> don't think there's a utility method to do that
[ 9:07 pm] <brahmana> ok.. let me see how I can accomplish that..
[ 9:08 pm] <brahmana> I am thinking this... for(i=0; i < this.parentNode.browsers.length; ++i) if(this.parentNode.browsers[i] == this) return i
[ 9:09 pm] <brahmana> that must work, isn't it?
[ 9:09 pm] <gavin|> probably
[ 9:09 pm] <gavin|> assuming this.parentNode is the tabbrowser
[ 9:09 pm] <brahmana> yeah.. I verified that..
[ 9:09 pm] <gavin|> though the "browsers" getter builds an array by looping through taqbs
[ 9:09 pm] <gavin|> so it would best be avoided
[ 9:09 pm] <gavin|> to avoid having to loop twice
[ 9:10 pm] <brahmana> oh.. instead we directly loop through the tabs..
[ 9:10 pm] <gavin|> yeah
[ 9:10 pm] <gavin|> http://mxr.mozilla.org/seamonkey/source/browser/base/content/tabbrowser.xml#1693

[ 9:11 pm] <brahmana> And if i change the XBL now, is there anything that I need to do during the build?
[ 9:11 pm] <gavin|> you just need to rebuild browser/
[ 9:11 pm] <gavin|> (or toolkit/ if you're touching browser.xml
[ 9:12 pm] <gavin|> why are you changing them, though?
[ 9:12 pm] <brahmana> ok.. thats great.. i can even go for a full build.. :-)

[ 9:12 pm] <brahmana> I want to associate every channel with the tab it is working for..


[ 9:14 pm] <gavin|> brahmana: is therea bug # for this?
[ 9:15 pm] <brahmana> gavin|, oh no.. there isn't.. I saw similar thing in Firebug, Cookie Manager and wanted to explore on that..
[ 9:15 pm] <brahmana> if by any means this is a desirable thing we can have a bug...

[ 9:24 pm] <brahmana> Is the code under xpfe/ still used?
[ 9:25 pm] <gavin|> some of it is
[ 9:25 pm] <brahmana> the window mediator?
[ 9:25 pm] <gavin|> yes

[ 9:26 pm] <brahmana> Along with the tabIndex I would also require some sort of window Id i guess, as tabIndices are not unique across browser windows, isn't it?
[ 9:26 pm] <gavin|> right
[ 9:27 pm] <bz> tabindices need not be unique within a single window either
[ 9:28 pm] <gavin|> er, why wouldn't they be?
[ 9:28 pm] <brahmana> oh.. well if we do not open and close stuff, they should be right?
[ 9:28 pm] <bz> because they're under the control of the page author?
[ 9:28 pm] <bz> And nothing prevents an HTML author from sticking tabindex="2" on every single node in the document
[ 9:28 pm] <gavin|> we're talking browser tabs
[ 9:28 pm] <bz> oh
[ 9:29 pm] <bz> nevermind, then


[ 9:31 pm] <brahmana> gavin|, And about the bug for the stuff I am asking, I assume this isn't really a desired feature, is it?

[ 9:31 pm] <gavin|> brahmana: I still don't really know what the feature is

[ 9:35 pm] <brahmana> gavin|, To observe requests for one tab in observerservice ... there is requirement for coupling tab index with the http channel....

[ 9:35 pm] <brahmana> And that is what I am trying to accomplish, associate the tabIndex with the channel..

[ 9:37 pm] <bz> brahmana: er.... you know the docshell involved in both places, right?


[ 9:39 pm] <brahmana> bz, yeah.. I was talking to gavin and others about the way browser interacts with docShell. But I did not fully understand your question.
[ 9:39 pm] <brahmana> browser as in the xul browser element.

[ 9:40 pm] <bz> brahmana: docshell is he guts of a browser
[ 9:40 pm] <bz> brahmana: the part that actually holds the web page, etc

[ 9:41 pm] <brahmana> bz, yeah.. that was evident from this one: http://www.mozilla.org/projects/embedding/docshell.html and also from the length of nsDocShell.cpp file..

[ 9:43 pm] <bz> ok
[ 9:43 pm] <bz> so you can get from a tab to a docshell
[ 9:43 pm] <bz> you can get from the channel to a docshell (usually)
[ 9:43 pm] <bz> then compare the two
[ 9:43 pm] <bz> for images you're out of luck
[ 9:45 pm] <biesi> brahmana, there can be a <browser> that's not part of a <tabbrowser>
[ 9:46 pm] <brahmana> bz, Whats special about the images?
[ 9:46 pm] <biesi> what's NOT special about images
[ 9:46 pm] <bz> brahmana: they don't so much follow necko rules
[ 9:46 pm] <brahmana> biesi, Are you referring to a situation with single tab?

[ 9:47 pm] <brahmana> bz, biesi yeah I had got the same statement when discussing about the request end notifications..
[ 9:47 pm] <biesi> brahmana, no, I'm referring to extensions or mailnews or whatever that doesn't support tabs
[ 9:47 pm] <brahmana> biesi, oh well.. that is probably not a problem. I don't think we would go beyond firefox.
[ 9:48 pm] <biesi> ok
[ 9:48 pm] <biesi> I have no idea what you're trying to do
[ 9:48 pm] * bz points to "extensions"
[ 9:48 pm] <bz> anyway
[ 9:48 pm] <biesi> I'm just saying, if you want to change browser.xml in mozilla.org
[ 9:48 pm] <biesi> 's repository, you can't assume that there's a tabbrowser
[ 9:49 pm] <brahmana> point fully accepted.. :-)
[ 9:50 pm] <brahmana> bz, I am not sure about getting to a docShell from a channel. Is there is straight forward way?
[ 9:51 pm] <brahmana> s/there is/there a

[ 9:54 pm] <brahmana> moreover the docShell will have reference only to one channel and I assume that is channel corresponding to the base document, i.e the main document request. Is that so?
[ 9:55 pm] <bz> brahmana: the docshell is the channel's loadgroup's notification callbacks
[ 9:55 pm] <bz> brahmana: generally
[ 9:55 pm] <bz> brahmana: we're talking about a channel having a reference to the docshell, not the other way around

[ 9:57 pm] <biesi> the load group has the channels/requests for all the loads
[ 9:57 pm] <biesi> except for iframes of course


[10:00 pm] <brahmana> biesi, bz.. a little out of the current discussion.. When the http-on-modify request is fired, will the connection be already set up? Is there a possibility to change the request URL in that event's listener?
[10:01 pm] <bz> "change" in what sense?
[10:01 pm] <biesi> the connection is not set up yet
[10:01 pm] <biesi> but you can't change the URL
[10:01 pm] * bz really wishes URIs were immutable and that channels' URI were readonly so people wouldn't ask questions like this
[10:01 pm] <biesi> the channel's URI IS readonly
[10:01 pm] <brahmana> ok.. :-)
[10:01 pm] <bz> ah, good

[10:01 pm] <bz> well, that URIs were immutable, then
[10:02 pm] <bz> I thought we were gonna do that for http sometime
[10:02 pm] <bz> flip the bit
[10:02 pm] * brahmana decides not to ask such questions to be compliant with bz's wishes..
[10:02 pm] <bz> (and see what breaks)
[10:02 pm] <biesi> oh we aren't? that sucks :/
[10:02 pm] <bz> well
[10:02 pm] <bz> we're not _yet_
[10:02 pm] <bz> I also wish URIs were immutable by default instead of the nsIMutable mess..
[10:02 pm] <bz> but then again, I could use a pony too


[10:02 pm] <bz> Or better yet, a kayak
[10:02 pm] <bz> it is?
[10:03 pm] * bz looks in his "to check in" folder
[10:03 pm] <biesi> there's so much stuff that I'd like to change if I could...

A web-wizard to create XPCOM components in JS

JavaScript XPCOM Component Wizard

As I might have mentioned in my earlier posts, creating XPCOM components can sometimes be hell. (So wait for js-ctypes to get on-board). Mozilla people know about this pain and hence have published sample code snippets to help you in figuring out what all is necessary and what all is optional. Even the process of creating your component from these snippets can be cumbersome sometimes. So one mozillian, ted, went a step ahead and created a web based wizard to create a component in JS. You just enter a few parameters and a skeletal component is ready for use. It simply saves a lot of time.
Just visit the link at the top and try your hands at it.

Happy Componenting. ;-)

--Brahmana.

Getting bluetooth working on your Thinkpad

I recently started using a Thinkpad T60p and its been good till now. Today I wanted to transfer some files from my cellphone to my laptop via bluetooth. It took my quite sometime before I figured it out(of course with the help of people around me). So I sat down to write this blog post.

Technically the thing is very simple but not apparent. The problem is that the default Thinkpad driver called the enhanced data transfer does not provide any UI to use the device for any sort of communication. Hence Microsoft guys have come up with a driver which will give you a nice System Tray icon from where you can launch settings, send or receive files and all such Microsoft goodies. So to use that one visit this page to download the driver. (Download the one named Microsoft bluetooth support). Run that executable to have the contents extracted to some standard location like: C:\Windows\Drivers\.... . Now do to Device manager -> Bluetooth Devices and click on "Update Driver" and select the above mentioned location where the extracted files are kept. This will install the new driver. Ideally you should be ready to use your bluetooth now. Even now if you are unable to use then go to the bluetooth settings (using the system tray icon) and select the "Hardware" tab at the end. There make sure the "Microsoft Bluetooth Emulator" is selected. After this your Thinkpad bluetooth should work.

Happy toothing.

Frame Relay --- An old but interesting cost-cutting network innovation

[Yet to write this up]

Web protocol handler in FF3

This is one awesome thing to happen. Haven't you crunched your face in disappointment and frustration whenever you clicked a "mailto" link on a webpage and unnecessarily have your desktop mail application (mostly unused and not configured one) popping up, hogging quite a bit of resources? The worst case scenario would be, the above described things happening even when you are logged into one of your web based mail services and thats where you wanted to send an email from.

Now all that go away. You Gmail might just give you a new mail compose page when you click on a mailto link. And this is not just for the "mailto" link. There can be web based protocol handlers to any sort of protocol. So all in all FF3 will rock like anything along with making web applications rock harder. So folks be ready to dance for the beat, for the beat will be pretty high paced. :-)

More details here: mfinkle's blog.

JS-CTYPES Status

Mark Finkles Weblog » JS-CTYPES Status

This was what mfinkle recently talked about js-ctypes. As he said because of the hectic FF3 schedule the devs at mozilla were pretty busy and hence nothing could be done. I being a lazy bot, also did not do anything. In the blog post above, Mark talks about the struct support (which I told that I would be doing). Now I exhibited my initial enthusiasm by writing up a doc, which was not really useful (its here ). After that nothing much happened except for a small bug fix.

Now that mfinkle has rekindled the interest I will also try and contribute. The struct support is still pending and I will start getting XPCOM code exchanging struct information with libffi. After that we can look at the other interface of JS exchanging structs with XPCOM to finally complete the loop. In the mean time I am also looking at embedding mozilla. Will put up a post about that very soon.


Again, Get ready to use deadly binaries in your JS code. JS-ctypes is soon going to be here.. ;-)

Calling Conventions

This is the outcome of today's presentation by Anand - our tech lead.
There are four calling conventions, and the Microsoft terminologies for them are:

1. __stdcall
2. __cdecl
3. __thiscall
4. __fastcall

As its known, the calling conventions will tell who, amongst the caller and callee, will clear the parameters from the top of the stack when the function call completes.

__cdecl is the old 'C' way where the stack is cleared by the calling function. Clearing the stack is nothing but having a statement to move the stack pointer up by some number (which is generally the number of arguments). So when a call is placed, immediately after instruction to call, there will be an instruction to move the stack pointer. This way we have an extra instruction associated with every function call instruction. This leads to the increased size of the executable, because of one extra instruction for every function call.

__stdcall is the new way where every function is responsible for clearing the stack allocations made for its parameters. As in, the callee is the one who will rewind the stack and move the stack pointer. In this approach before the return statement of the function a statement to move the stack pointer is introduced. This way the number of instructions will not increase with the number of the function calls.

__thiscall is the object-oriented way of calling functions. Here the first parameter passed is always the "this" pointer. So any instance function member of a class will follow this calling convention. Unlike other parameters, which go to the stack top, the "this" parameter which is also passed is stored in "ECX" register. Now this is how the "this" pointer is implicitly available to all the instance member functions. No function can be explicitly qualified with this calling convention. Any instance member function will get this implicitly, but it can be over-ridden with any other calling convention.

__fastcall is an approach where in the system will try to push the parameters on to Registers instead of the stack just to get a performance boost. But as with the variable qualifier "register" , if the registers are not available then again stack is used for the parameters.

There are a couple of other posts regarding the specifics of the calling conventions and comparison. If this post made any sense then check out those also.

Mozilla - Bookmarks' keywords & Quick Searches

This world is really full of amazing things. How much ever we discover and find out, there will be something interesting just right next to us which we would not have figured out. This post is about one such thing, which was right here from such a long time and I had not figured it out.

The first hints for this came to me when I saw one of my colleagues, Pramod S, at NI using shortforms for Google(g) and Wikipedia(wp) when searching. I thought he had come across some hack and tweaked his browser. But just today I realized that it is infact a feature in Firefox.

For any URL or webpage that we bookmark we can assign a keyword, a shortform, which can be entered in the URL bar to visit that bookmarked page. This is really a very useful things if we visit certain pages over and over again. In my case http://mxr.mozilla.org/seamonkey/ is one site that I cannot live without. So I have keywords for the 3 different searches available at this page, viz:

• mxrf - for file search(find)
• mxri - for identifier search
• mxrt - for free form text search

This feature is really handy and increases the brwosing speed to a great extent.

How to get this working? or how to assign the keywords?. This again is really simple. For existing bookmarks just go to the Bookmark properties by Right-Clicking on that and you get a small dialog with a text field for Keyword. Key in whatever you want and click on OK. Thats it you are ready to fire the internet with your keywords. For new bookmarks it is still easier. When you are creating a bookmard with the helper dialog you get the keyword field which you can key in or leave it blank (either to be filled later or not to be filled at all).

This is not the end. There is yet another enhancement to this. Certain bookmarked urls have query strings, as in case of search urls. The google url: http://www.google.com/ is followed by a query string of the form "search?hl=en&q=search string" which carries the search string entered. Such strings work with the keywords also. Meaning you can search for something on google using the keyword, 'g' like this:

"g any-search-string" . This will be converted to appropriate URL with a proper query string as specified in the bookmark properties. As it appears this of course requires a little bit of extra work. When saving an URL as bookmark, for instance when saving http://www.google.com/ you can save it along with the query string, but the search string being substituted by "%s" and whatever string you specify with the keyword will replace every occurance of %s in the URL. So this is how the bookmark URL will look like:

http://www.google.com/search?hl=en&q=%s ---- with g as the keyword. Now when using this bookmark, if you enter something like this in the URL bar:

"g bookmarks " -- this becomes-- http://www.google.com/search?hl=en&q=bookmarks -- and finally giving you search results for bookmarks.

Now this is how I have arranged my mxr bookmarks:

• http://mxr.mozilla.org/seamonkey/find?string=%s -- with keyword: mxrf
• http://mxr.mozilla.org/seamonkey/ident?string=%s -- with keyword: mxri
• http://mxr.mozilla.org/seamonkey/text?string=%s -- with keyword: mxrt

Now does this method appear too complex and cumbersome? Well mozillians come to your rescue yet again. You can save such enhanced search bookmarks called "Quick Search" bookmarks very easily like this:

Go to the search page and Right-Click on the search bar, and click on: "Add a keyword for this search" . That will give you the small bookmarking dialog, with a field for keyword. All that you have to do is to just enter the keyword that you want and click on OK. All the URL, query string and %s headache is taken care by Mozilla.

Well mozilla is here just to make your browsing experience easier, safer and worthwhile., isn't it?

And I am proud to be associated with this awesome community. ;-)

Happy and Fast Browsing.