Witting parsers is fun! Mainly when one can concentrate on the semantics of the elements being parsed and can forget the syntax of the language used to write the parser!
The easyness or dificultyness very much depends on the language you're using to write it. That's the reason that if possible, I write them in FORTH. Right now, the application I'm developing is being written in PHP but for the parsers I'm using FORTH; at least I don't have the language syntax in my way as mainly FORTH is a semantic language....
I'm writing a 3 tiered web application and each part comunicates through ??ML languages define for the purpouse;
I have the UIML (User Interface Markup Language) to comunicate between the Root_Server (the application) and the User Interface, and the QML (Query Marckup Language), RSML (Result Set Markup Language) and DBML (Data Base Markup Language) to comunicate between the Root_server and the Back_End processes. And all are carried by the MML (Message Markup Language).... :-)))
Andm of course, it's always easy to design a parser for our own languages than just for other's as we can always define them the way we want and need them!
Prob<abily more about this will apear on my other LJ journal....
amused
Argh
You have to be SMART to write those things!!
*my head hurts just remembering it*
That's the reason that if possible, I write them in FORTH.
Right now, the application I'm developing is being written in PHP but for the parsers I'm using FORTH; at least I don't have the language syntax in my way as mainly FORTH is a semantic language....
I'm writing a 3 tiered web application and each part comunicates through ??ML languages define for the purpouse;
I have the UIML (User Interface Markup Language) to comunicate between the Root_Server (the application) and the User Interface, and the QML (Query Marckup Language), RSML (Result Set Markup Language) and DBML (Data Base Markup Language) to comunicate between the Root_server and the Back_End processes.
And all are carried by the MML (Message Markup Language).... :-)))
Andm of course, it's always easy to design a parser for our own languages than just for other's as we can always define them the way we want and need them!
Prob<abily more about this will apear on my other LJ journal....