The DrScheme Programming Environment

One of the main activities of many computer scientists is that of programming, expressing algorithms in a form understandable to the computer. (Even computer scientists who emphasize other issues often end up needing to build programs to help support arguments or to provide concrete evidence for theories.) Much of this semester, you will be writing programs as a way of learning about important concepts. Programming is also a skill that you can apply in a variety of contexts.

At this point, most programmers develop their code in what is called an program-development environment. Program-development environments let you write code, test bits of the code, format your code for easy readability, obtain documentation on built-in procedures, and so on and so forth.

In this class, we will use a program-development environment called DrFu. DrFu is an environment developed by a group of students and faculty at Grinnell. DrFu takes the user interface of DrScheme, a leading program-development environment for Scheme, and uses it to interact with GIMP, a leading open-source raster graphics drawing environment. DrScheme was developed for teaching, and GIMP lacks a good program-development environment, so the pairing seemed to be useful, particularly for this course. Since DrScheme provides the interface for DrFu, we tend to use the two terms interchangeably. However, we will try to reserve the term “DrScheme” for the aspects primarily relevant to DrScheme (whether or not it is used in DrFu), “GIMP” for those primarily relevant to GIMP, and “DrFu” for cases in which you really need both.

You start DrFu by clicking on the icon for DrFu that you selected in the introductory Linux lab. If you have no such icon, find a teaching assistant or teacher for help.

You will find that DrScheme has many similarities to most applications, including an Edit that supports operations like cut and paste. You may, however, notice that DrScheme has a primary window that looks somewhat different than other programs.

As you'll note, that window has two panes. The top pane is called the Definitions Pane and the bottom pane is called the Interactions Pane. As the names suggest, the top pane is used for writing definitions (more on those later), and the bottom pane is used for your primary interactions with DrScheme. We will start with the interactions pane.

For you initial use of DrScheme, you can treat the interactions pane as a fancy calculator. That is, you'll enter expressions and it will respond with the values of those expressions. For example, you can ask it to add 3 and 4, or to find the square root of 144.

The expressions you write in the interactions pane have to use the syntax of the Scheme programming language. Scheme uses a consistent syntax that makes a lot of sense to the computer, but that takes a bit for humans to master. (Fortunately, Scheme's syntax is much simpler than that of almost any other computer language.)

Here are the two key points to remember when writing Scheme expressions:

Why does Scheme use these two key ideas? Parentheses help Scheme resolve potentially ambiguities. For example, you'll note that different calculators compute different results for 3+4*5, depending on whether or not they are designed to accommodate precedence. In Scheme, since you must parenthesize every subexpression, you must write either (+ 3 (* 4 5)) or (* (+ 3 4) 5). Each expression indicates precisely what you want computed.

Prefix order, on the other hand, is intended to make life easier for the programmer. In particular, in languages in which some operations appear before their operands (such as sqrt) and others appear between their operands (such as + or modulo), programmers must reflect on which order to use for each operation. In Scheme, there's never a question, because you always put the operation first.

When DrScheme is ready for you to type an expression in the pane, it prints a greater-than sign (also known as a right angle bracket). You type an expression you want evaluated and then hit the Return key. If you have not closed all the open parentheses, it will let you type more on the next line. Once you have completed an expression, it will evaluate it and present the result.

> (+ 3 4)
7
> (sqrt 144)
12
> (string-append "Hello" " " "Sam")
"Hello Sam"

While the first two expressions are numeric, the last shows that DrScheme can work with words, too. (As long as those words are enclosed in non-curly quotation marks.)

At times, you will find that you made a small typo in a long expression. For example, in computing the average of the grades 90, 67, 80, 85, 100, and 91, you might write

> (/ (+ 90 67 80 85 100 91) 5)
102.5

The result suggests, of course, that there was an error in the expression. (After all, the average of grades that are all 100 or less should not be more than 100.) A bit of analysis suggests that you really wanted to divide by 6, rather than 5. What can you do? You can retype the whole expression, but that takes a bit of time and gives you an opportunity to mistype something. You can cut and paste, and then edit the 5 (using the arrow keys and the backspace or delete keys).

Alternately, you can take advantage of DrScheme's Command History. DrScheme lets you scan through the previous expressions entered. To back up, you enter Esc (which appears in the upper-left-hand corner of the keyboard) and then P for “Previous”. You can type Esc and P multiple times to back up more. You can also type Esc and N for “Next”.

We've seen what the interactions pane is for, so what should we use definitions pane for? The simplest use is to assign names to values to make it easier to write our expressions. For example, if we wanted to work with grades on six homework assignments, we might name the grade on the first assignment grade1, the grade on the second assignment grade2, and so on and so forth. That way, the expression to compute the average grade would be the clearer

> (/ (+ grade1 grade2 grade3 grade4 grade5 grade6) 6)

In Scheme, we name values using the define operation, which we put in the definitions pane.

(define grade1 90)
(define grade2 67)
(define grade3 80)
(define grade4 85)
(define grade5 100)
(define grade6 91)

The definitions in the definitions pane are evaluated when you click the Run button. Once you've done so, you can use the names in the interactions window.

If you spend a bit of time naming values, you may find it useful to save those definitions into a file that you can then restore later. As you might guess, you save definitions using a menu item from the Save. Since there are two panes in the window, DrScheme provides both Save Definitions and Save Interactions (as well as a host of variations thereof). Since the definitions are more likely to be of permanent use than the interactions, the default is to save definitions.

The custom is to save Scheme files with a suffix of .ss or .scm. Note that DrScheme does not add the suffix automatically - you need to add it yourself.

Once you have saved definitions, you can quit DrScheme, go off and do other work, and then restart DrScheme and reload the definitions. As you might expect, you can load old definitions by using Open... or Open Recent from the File menu.

If you end up creating lots of definitions files and want to use them simultaneously (but don't need to edit them), there's a command you can type in the interactions window to get definitions from a file rather than from the definitions pane, load. You need to put the name of the file in quotation marks. For example,

> (load "mygrades.ss")

You have now learned enough to begin to interact with DrScheme. In a subsequent laboratory, we will ground these abstract instructions in concrete exercises.