-
October 21st, 2004, 06:41 PM
#21
I will join in and help, but I can only help if we use Perl or C++. My C++ is not the best.
Let me know if you guys want help though. Anything in perl would work great. I can do C++ **** though, let me know.
Oh and a Instant messenger in C++ would be kind of cool.
-
October 21st, 2004, 10:52 PM
#22
I'm interested in this. I'm open to about any kind of project had have a very broad skillset when it comes to languages and technologies used (check my profile if you want a 50k foot overview).
"When I get a little money I buy books; and if any is left I buy food and clothes." - Erasmus
"There is no programming language, no matter how structured, that will prevent programmers from writing bad programs." - L. Flon
"Mischief my ass, you are an unethical moron." - chsh
Blog of X
-
October 22nd, 2004, 07:25 AM
#23
Hi,
h3r3tic :
I'd like to help or at least follow along and have access to the work.
I think thats the whole purpose of this.............Learning and sharing what we have learned
whizkid2300 :
I will join in and help, but I can only help if we use Perl or C++. My C++ is not the best.
I too don't have any experience in Compilers........All i have is read theorital concepts about it in school never implemented them.........it would be a challange ......so we still now sure about the tools plateform required.........lets first do a requirements analysis and see whats required or suitable.
Juridian :
I'm interested in this. I'm open to about any kind of project had have a very broad skillset when it comes to languages and technologies used (check my profile if you want a 50k foot overview).
With your experience you could be a great help .........can you tell us about requirements for compilers building .....where should we start?...............what do we require.?.........major problems to be ready to face? etc.
And a straight quiestion do you think it is feasible to go for this project?. .............Should we stick to it or we change over to something else?
--Good Luck--
-
October 22nd, 2004, 10:48 AM
#24
Hi,
Thanks for the response guys it will be surely of great help if u people join us.
And a straight quiestion do you think it is feasible to go for this project?. .............Should we stick to it or we change over to something else?
SwordFish : I also have only the theory for compilers and never ever thought of trying one but still what i feel is we should not skip this coz afterall compilers are also written by humans and we are 100% humans heheheh.. so why can't we do this ? plus we will have many people helping us around in case we get stuck up somewhere.
Hence it will be better if we first do the requirement analysis and fix up the platform and language which will be used and i am sure we will surely see light on the other end of the tunnel .. what say ??
In
A laptop, internet connection and beer.
-
October 22nd, 2004, 12:42 PM
#25
Hi,
I am on it ................
-
October 22nd, 2004, 01:00 PM
#26
Hey me too
A laptop, internet connection and beer.
-
October 23rd, 2004, 07:57 PM
#27
Hi,
This is the first time I am attempting to build a compiler……………Previously I had only read about the theoretical concepts involved this will be the first time I am going to implement them
And above all I am being able to deveote little time for this ……my Uni exams are coming and I have to study AI and OS , LISP and Oracle/D2K with this…….I am doomed…
I have been trying to collect information about similar projects didn’t get much ……..
I have got conflicting definations of a compilers..........i.e A compiler is a program that takes it’s input in a form of a high language and gives output in another language………… it can be assembly or some form of binary or executable ……….some defined it as a program or group of program that translates one language into another (doesn’t specify anything so I am taking ti can be another higher level language?)…………it says generally the output language is in Assembly ( but some said you can also try to convert it into another higher language and then use their compiler to get the binary) ………..and that assembly goes through assembler and linker to get the binary and that’s I think out of our current score ?……..our scope is just up till conversion?
The compiler works in stages also called passes…….A pass is a complete traversal of the source program, or some internal representation of the source program.……..generally there are 4 passes in a C compiler called the 4 pass compiler but there are 1 pass and 2 pass and 3 pass compilers as well.
The major process are Lexical Analysis , Parsing……..there are a few tools available for each process like the age old Lex (lexical Analyzer) For Lexical Analysis and YACC(Yet Another Compiler Compiler) for parsing……..there a few freeware tools also available now that work like them………but using them would kill the fun
Here is what I recommend
- Rather than using pre-built lexical and parsing Tools we make our own Lexical Analyzer and Parser in C…..…and implement them as functions one function calls another in a sequence?
- We make it a 1-pass compiler.?
- And we should stop saying we are making a C compiler ………rather than that a compiler for a language that looks like C……….. I mean making a full C compiler might take more than year or so. ….L
- It would work something like this
Source Code --> Lexical Analysis-->Syntactical Analysis-->Symantec Analysis-->Code
This compiler that takes the source code as input does the then does the Lexical Analysis and then the Syntactical analysis and semantic analysis then generates a Intermediate code(what this intermediate code will be haven’t worked out yet) and that’s it .
- We facilitate one or two or at-most 3 data types in our language? Thereby reduce the complexity……..although I recommend just two one is Integer and the other raw which takes anything other than integer………just make it for mathematical type of things first
What do you people think?
Corrections ?
Recommendations ?
Comments?
Before we go on to write a full analysis report …….....I say we try to maintain proper documentation of whatever we do starting for analysis till the last phase..........it will help others who join or see it later.
--Good Luck--
-
October 24th, 2004, 06:56 AM
#28
Member
Its a great idea to start a project on compilers development. I would like to follow this discussion and contribute the best i could.
Thanks intruder and swordfish for all that u r doin'
I suggest some links below to continue with R&D on compilers
http://www.research.att.com/~bs/compilers.html
http://www.gnu.org/software/gcc/gcc.html
First it is must to chalk out the plan what ur compiler should be capable of handling. Also u would need to go into the details of libraries and header files of the compilers which r already available. Also give a thought to Linker, how exactly u are going to convert ur compiled assembly code into binary executable.
Good luck to all of u.
Gaurav
-
October 24th, 2004, 07:32 AM
#29
Member
Wow, I would love to help...but this all sounds very very complicated and I am not sure my limited programming skills would help a bunch. But I would still love to follow along and see how you all do it. Hope we get something started tho, cant wait.
-Ep
01001001001000000100110001101111011101100110010100100000010000100110010101110100011101000111100100100001
-
October 24th, 2004, 08:09 PM
#30
Epison it doesn't take much programming knowledge to help on any kind of project, the basic thing behind a project, with a group is that everyone knows a little something. Not many people will know everything. So you take a small problem like. Writing code to search for libraries. Simple enough. Take small things, and then once your done with that take on something else. Shortly you are getting better, and being given harder tasks, Which in turn helps you out more, but also helps out the project more.
Pretty much, my main requirement for whenever I do a group project, is atleast one person that doesn't know a **** load, they like to work, and they tend to work hard. So give that person bugs to fix, (Which trust me is a pain in the ass.) You learn from doing that, so you become better.
If it is limited programming experience which is your problem, then start programming now. Something tells me with how AO works, this thing won't start for a little while.
Posting Permissions
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts
-
Forum Rules
|
|