Prashant Deva's blog

Talks on technology and the latest developments in Placid Systems.

No more 'Flux' in the Flux Debugger

Friday, February 24, 2006

It turns out that the word 'Flux' is trademarked by fluxbpm.com. They have requested me to change the name of Flux Debugger to something else.

I have agreed to the name change and it will take affect on the day of release of ANTLR Studio 1.1.0, which apparently will be released next week.

The term Flux was coined by me when looking for synonyms for something that would denote the union of Java & ANTLR while debugging.

Another name I had thought for the debugger was-
Conjoined Debugger

But I am open to other suggestions too. So if any of you can think of another name for ANTLR Studio's debugging technology, I am all ears for it :-)

Also could someone tell me how to verify if the new name I choose is already trademarked or not?
posted by Prashant Deva, 11:50 PM | link | 0 comments |  

ANTLR 3 to have inbuilt Incremental Parsing functionality

Friday, February 17, 2006

Havent blogged in a while, have I? Well here goes then.

Lets talk a little about ANTLR 3. I guess a lot of people use ANTLR generated parsers in IDEs. And as we all know, you can make an IDE go real fast if you have an incremental parser instead of the regular one.

For those who dont know, an incremental parser parses only those parts of the file which have changed since the last time it was edited. This is a huge advantage in IDEs, especially when you have large files open, cause you dont want to parse the entire file from top to bottom every time a user types a single key.

But making an incremental parser out of the one generated by ANTLR is pretty tough. As a matter of fact there were no known algorithms for making an incremental parser out of a recursive descent parser (which is what ANTLR generates). You notice the word 'were', instead of 'are'. Yes, there is an algorithm for doing so now. Actually it was stumbled upon by yours truly, when I was designing ANTLR Studio's editor.

The algorithm is already implemented in ANTLR Studio. It is what makes all the cool stuff like TypeOnce in the editor possible.

So I had a talk with Terence and guess what, we are implementing this functionality in ANTLR 3. So in ANTLR 3 all you will have to do is put in an option like 'incremental=true' in the grammar file and voila, it will generate a full fledged incremental parser for you!

For those who cant wait till the release of ANTLR 3 (which is still a long way away), I will be publishing a paper on the algorithm along with Terence in a few weeks. So you will be able to implement an incremental parser with ANTLR 2 too.
posted by Prashant Deva, 3:21 PM | link | 3 comments |