well i finished the xpcom labs; here are the outputs I got for these labs.
xpcom (part 1)
xpcom (part 2)
Sunday, November 25, 2007
Sunday, November 11, 2007
thunderbird lab
thunderbird lab
In this lab we were supposed to fix a bug where by a person who enters an '@' in the middle of two words; weather it be an email address or not; is mistaken as an email address
for example:
we can set up a meeting location at seneca@york; where seneca@york will be mis identified as an email address.
the location of the patch is on my wiki under Lab notes section link below
http://zenit.senecac.on.ca/wiki/index.php/User:Vbala#Lab_Notes
for example:
we can set up a meeting location at seneca@york; where seneca@york will be mis identified as an email address.
the location of the patch is on my wiki under Lab notes section link below
http://zenit.senecac.on.ca/wiki/index.php/User:Vbala#Lab_Notes
Monday, November 5, 2007
Vladimir Vukicevic's presentation on Debugging
According to Vladimir; the most important parts of debugging are setting up break points and watching how data is being manipulated through the program. Vladimir compared two major types of debuggers Visual Studio, and GDB. GDB is a more command line based debugger which is harder to use, but Vlad found this to be his debugger of choice over Visual studio debugger (or atleast thats what I took from it)
The advantages of using Visual Studio debugger was the ability to use the GUI for easier break point setting compared to GDB which is all command line and harder to implement break points. The advantage of setting break points or watch points is they are a good way to track data changes in bytes withing an address location. Vlad also talked about EMAX which has a hook up to GDB which offers a certain level of user interface for GDB. Vlad also stressed on that fact that debugging does not solve a bug; it merely helps you to retrieve a bunch of data that helps in removing bugs; sometimes debugging might just lead you to dead ends but there are those times when it can be very useful
Vlad also mentioned Venkeman and Firebug which are useful Java script debuggers. He then talked about dTrace or Dynamic Trace which is a way to track or trace execution of code /files over a running program. It is often a good way to show when files are being used, and or accessed, and network connections are being made.
He then talked about probes used within GDB. His purpose for a probe or my understanding of his purpose of a probe is that they are used to identify target areas where the code addresses certain data variables and where the changes are being made or updated. Probes are used as a way of defining the "target" for a certain file area to be addressed during debugging. Vlad also noted that GDB has a tendency to be buggy or crash in a Linux based environment.
Vlad also touched on Chronicle which is a framework that provides complete registry and memory level recording for linux processes and efficient queries over the recorded traces. Chronomancer is another program that offers a GUI front end for Chronicle.
Overall it was a very informative session and was definitely a unique and unparalled opportunity to watch one of the programming elites of Mozilla going through and breaking down what he does for a living.
The advantages of using Visual Studio debugger was the ability to use the GUI for easier break point setting compared to GDB which is all command line and harder to implement break points. The advantage of setting break points or watch points is they are a good way to track data changes in bytes withing an address location. Vlad also talked about EMAX which has a hook up to GDB which offers a certain level of user interface for GDB. Vlad also stressed on that fact that debugging does not solve a bug; it merely helps you to retrieve a bunch of data that helps in removing bugs; sometimes debugging might just lead you to dead ends but there are those times when it can be very useful
Vlad also mentioned Venkeman and Firebug which are useful Java script debuggers. He then talked about dTrace or Dynamic Trace which is a way to track or trace execution of code /files over a running program. It is often a good way to show when files are being used, and or accessed, and network connections are being made.
He then talked about probes used within GDB. His purpose for a probe or my understanding of his purpose of a probe is that they are used to identify target areas where the code addresses certain data variables and where the changes are being made or updated. Probes are used as a way of defining the "target" for a certain file area to be addressed during debugging. Vlad also noted that GDB has a tendency to be buggy or crash in a Linux based environment.
Vlad also touched on Chronicle which is a framework that provides complete registry and memory level recording for linux processes and efficient queries over the recorded traces. Chronomancer is another program that offers a GUI front end for Chronicle.
Overall it was a very informative session and was definitely a unique and unparalled opportunity to watch one of the programming elites of Mozilla going through and breaking down what he does for a living.
Saturday, October 27, 2007
FSOSS
Well; what an event! for a first time newbie who has had little or no exposure into the OS world; this was quite an opener. I attended more than two presentations; and volunteered in more than one role for this event, so it was quite an interesting place to be.
The presentations I attended were
The presentations I attended were
- Benjamin Smedberg's reading code presentation
- Mike Beltzner's presentation on user experience design in open communities (funny and intriguing presentation; one of his lines is still stuck in my head "Well since no one has any questions; Ill show you Leopard and my Icons....weeeeeeeeeeee" =))
- Mark Surman's open sourcing education in South Africa and rest of the world
- Lawrence Mandel's & Jeffrey Liu's open commercial development
- Bob young's applying open source concepts to non-software industries -- (by far the most friendly, humble and most experienced one of them all in my opinion; not that all the other people who presented were bad; its just he was so much more captivating in the way he presented his topic)
- Community Management as open Source's core Competency by David Eaves
- Community as a Core competency by Bryan Krischner
catching up on the labs
Friday, October 19, 2007
Mark Finkle Presentation
Watching him in action was like watching a seasoned race car driver pushing his car to limits of handling through the toughest of circuits; that only a few can even fathom of reaching. To me it was a learning experience; and a reassurance that in case I had some doubts or questions that there are people out there who can possibly answer any little or major snag I come across in my extension development. I wish I had approached on him on suggestions and questions I had about my extension; since I was terribly sick at the time; I decided it would not be a good time to approach him, after all it would only make for an awkward first time discussion.
Subscribe to:
Posts (Atom)