Ok, here is the exact assignment. * *
*Tiny Scheme Details**:* The Scheme or LISP that you are going to implement should include the following primitives: *T, NIL, CAR, CDR, CONS, ATOM, EQ, NULL, INT,* *PLUS, MINUS, TIMES, LESS, GREATER COND, QUOTE, DEFINE.* T and NIL represent true and false. In general atoms may be identifiers or integers. An identifier will start with a letter of the alphabet, and maybe followed by one or more letters or digits; only uppercase letters will be used. No underscores or other characters are allowed in identifiers. Integers may be signed or unsigned, i.e., 25, +25, -25 are all legal. CAR, CDR, CONS are the standard Scheme primitive functions. ATOM returns T if its argument is atomic, and returns NIL otherwise. INT returns T if its argument is an atom that is a number. EQ works only on atomic arguments; it returns T if its two atomic arguments are the same [or equal, for numbers] and NIL otherwise. NULL returns T if its argument is NIL and NIL otherwise (non-atomic arguments should be acceptable to NULL). PLUS, MINUS etc., take two numbers [integers] as their arguments and return the result [an integer]. LESS and GREATER allow you to compare two integers and return T if the first is less or greater than the second respectively. COND and QUOTE are the standard Scheme forms. DEFINE allows the user to define a new function and use it later. A function definition looks like (define (f x) fb ) where f is the function being defined; its formal parameter(s) is named x; and its body is the Scheme expression fb. When this is "evaluated", it should add a pair (f . ((X Y) . fb)) to the d-list [the list of definitions that your interpreter should maintain internally; actually, how you store function definitions on the d-list is up to you; the above is only one possibility]. The features of Tiny Scheme that have not been included are: LAMBDA, LABEL. These are not needed because Define is sufficient for defining new functions. I thought it would be easier to post it all instead of trying to explain it. On 9 November 2010 15:37, John Clements <cleme...@brinckerhoff.org> wrote: > > On Nov 9, 2010, at 12:32 PM, Peter Breitsprecher wrote: > > > Ok, here is the code I have written so far for the function I am having > trouble with... > > > > (define (rep-loop) > > (newline) > > (display "repl>") > > (read-token)) > > I'm going to ignore rep-loop. > > > > > (define (read-token) > > (let ((expr (read-char))) > > ((cond [(char-whitespace? expr) (read-token1)] > > [(eq? expr #\() (read-token2)] > > [(eq? expr #\)) (read-token3)] > > [(char-alphabetic? expr) (read-token4)] > > [(char-numeric? expr) (read-token5)] > > [else (display "Exiting Loop")])))) > > Okay! we've got a name: read-token. > > The way it's written, it's going to be very hard to test. This is because > it operates on typed-in input. > > This program would be much easier to develop if you could write a function > that accepts strings as input. Can we break the problem up this way? > > John Clements > > > > The cond calls other functions I have written just so I could see it if > was reading the output letter by letter... Now I need to make read-token > read the whole word if it sees a character and the entire number if it sees > a number. > > > > Is that getting closer? > > > > On 9 November 2010 15:28, John Clements <cleme...@brinckerhoff.org> > wrote: > > > > On Nov 9, 2010, at 12:27 PM, Peter Breitsprecher wrote: > > > > > I'd lie but no.. I was just trying to clarify what I was looking for. > I gave a the code for one of the functions I had written, and then an > example of the input the user is giving below, and then tried to describe in > more detail what I needed help with... > > > > Well, I'm glad you're not lying :). > > > > Can you start by picking a name for the function that you're trying to > write? > > > > John Clements > > > > > > > On 9 November 2010 15:23, Shriram Krishnamurthi <s...@cs.brown.edu> > wrote: > > > Hi, > > > > > > That would not be considered an "example" for at least two reasons. Do > you see why? > > > > > > > > >> On Nov 9, 2010 12:17 PM, "Peter Breitsprecher" <pkbre...@lakeheadu.ca> > wrote: > > >> > > >> As an example, I want have most of the functions for Plus, Minus > already written. > > >> > > >> Here is the code for it. > > >> > > >> (define (plus num1 num2) > > >> (+ num1 num2)) > > >> > > >> That is childs stuff... Here is where I am having the problem, and > please keep in mind I am new to DrRacket... > > >> > > >> If my program accepts input such as > > >> (plus 13 5) > > >> > > >> I need to parse it so that I know what it is asking for. If It is a > left parent, then I need to move on, if it is a right parent, move on to the > next character, once i get to the next character I can see it is an > alphanumeric character, but then I need to somehow read the whole word, so > that I can store that value in a variable so that I can read the next > number. If the numbers are single digits, it is easy, because they are only > one charater each and read-char works perfect, but I need something like a > read-word, or read-number and I can't seem to figure out how to do it. > > >> > > >> So really I need to know how do I read the words when there are > characters present, and read the whole numbers when there is a number > present. > > >> > > >> > > >> > > >> On 9 November 2010 15:00, John Clements <cleme...@brinckerhoff.org> > wrote: > > >> > > > >> > > > >> > On Nov 9, 2010, a... > > >> > > >> -- > > >> Kurt Breitsprecher > > >> (807) 474-9601 > > >> pkbre...@lakeheadu.ca > > >> > > >> > > >> _________________________________________________ > > >> For list-related administrative tasks: > > >> http://lists.racket-lang.org/listinfo/users > > > > > > > > > > > > -- > > > Kurt Breitsprecher > > > (807) 474-9601 > > > pkbre...@lakeheadu.ca > > > > > > > > > > > > > -- > > Kurt Breitsprecher > > (807) 474-9601 > > pkbre...@lakeheadu.ca > > > > -- Kurt Breitsprecher (807) 474-9601 pkbre...@lakeheadu.ca
_________________________________________________ For list-related administrative tasks: http://lists.racket-lang.org/listinfo/users
