Methods and functions package reusable instructions. So far, our examples have only used assignment and expression-based instructions. Limited to this, our worlds would basically behave the same way over and over again. What we want is to be able to change part behavior as the world's conditions change.

To accomplish this, our expression capabilities must be expanded to be able to evaluate the state of the world. Then we need control flow instructions that change how a method or function performs its instructions based on an evaluation.

Let's dive in!

Evaluating Values

How do we evaluate values against specific criteria? Broadly speaking, Acorn offers several tools:

Note: Evaluation operators have lower precedence than arithmetic operators, but higher than assignment.

Equivalence Operators

Equivalence operators are useful when you want to check whether a particular value has the same content and type as another value. Some examples:

x == 3.0        # Returns true if the local variable x's value is 3.0
x+1 != y        # Returns true if x is not exactly 1 greater than y
2 == 2.0        # Returns false, because the Integer vs. Float types do not match
true != false   # Returns true

Equivalence checking is not always straightforward, depending on the value's type. Since floating point calculations are imprecise and can introduce "rounding errors", two Float values will be declared "equal" if they are sufficiently close. Large, complex value types, such as lists, can take time comparing, as it requires comparing every element in the list for equivalence.

In some cases, you may prefer the '===' operator to quickly know if two values are exactly the same. Two floats would have to match exactly, with no margin for calculation error. Two List or Text values would have to point to the same List or Text value, as a copy made of a list would not be exactly equal using the '===' operator.

Comparison Operators

Comparison operators are useful when you want to see if a value is greater or less than some other value. You can only check greater and less for value types that are Comparable, which Integer and Float numbers are.

x >= 2          # Returns true if x is greater than or equal to 2 (>, <, <= are variants)
2 <= 2.0        # Returns false any time the types do not agree

Float values are compared using the same closeness criteria described above. For one Float value to be less than another, it must be less by at least the comparable closeness.

The '<=>' operator is another way to compare two values, returning an Integer rather than true or false: -1 if less, 0 if equal, and 1 if greater. "Comparable" types use this rocket-ship operator/method to facilitate comparing and sorting.

-2 <=> 4      # Returns -1

A fun example:

frosty +[
	'<=>' other:>
		1  # frosty is always the greatest!

frosty>Float::Infinity  # true

Pattern Matching Operators

Sometimes we want to see if a specific value matches a particular pattern. Later on we will introduce several types of patterns, such as Range, regular expressions, and even custom-built pattern types. In all cases, the way to do the match is the same:

5 =~ 1..7   # Returns true, since 5 is within the range between 1 and 7
1 !~ 5..6   # Returns true, since it fails the pattern match

Patterns will always return false if the match fails. Some will return true, but others, such as regular expressions, may return details about what it found during the match. This will be discussed later, when patterns are introduced.

The pattern match operator is the only one that uses the right hand value (the pattern) to send the '=~' method to, with the left hard value as a parameter. This exception makes the code easier to read, but puts the work on the pattern, where it belongs.

If the pattern value's type does not support the =~ method, the == method is used.

Evaluation Methods

Since operators do not handle all the ways we may want to evaluate values, we use type-specific evaluation methods to all other criteria. For example:

x.null?       # Returns true if x's value is null, false otherwise
list.empty?   # Returns true if list has no members, false otherwise

By convention, methods that return a true/false value have a '?' mark at the end, which improves code readability.

true or false?

How does Acorn evaluate whether a value is true or false? A value is considered false if it is either false or null. All other values of any type, including true, 0 and "" will be considered true.

Boolean Operators

What if you need to make a complex decision involving multiple values and criteria? This is where boolean expressions are useful. These use the logical operators and (&&), or (||) and not (!) to corral several evaluations together to arrive at a summary true/false decision:

0==3 || !2<3 && 3==3   # Returns true (&& is evaluated before ||)

The logical operators work mostly how you would expect:

val1 = null
val2 = 4.5
val = val1 || val2    # Returns 4.5

Note: In case it is not obvious by now, the logical control operators are not the same as the boolean arithmetic operators in other languages that do bit manipulations on Integer values. There should not be much need for them in Acorn, but they are provided as Integer methods (.not, .and, .or, .xor, .shl, .shr) if they are needed.

Control Flow Commands

Having evaluated values against criteria, these Acorn tools can be used to control what happens next:

As with expressions, these all return a value (or null if nothing was done).

? operator: What Value Is Behind Door #2?

The ? operator is a concise way to pick one of two values within an expression. If the expression before the ? operator is true, the first value after is picked. Otherwise the value after the comma is chosen:

x = val<5.0 ? 5.0, val   # sets x to the higher between 5.0 and val

Note: To avoid possible ambiguity, be sure there is a space before the ? operator.

if/switch/match commands: A Fork In The Road

There are several structures for specifying which instructions to perform.

if/unless suffix If you have a single instruction that should execute (or not) based on a condition, append the instruction with 'if' or 'unless' followed by the condition:

val = 5.0 if val<5.0   # if val is less than 5, set it to 5
velocity *= 1.1 unless stop_flag  # unless is the opposite of if

if command If you have several instructions perform or not, use the if command. The instructions to be performed should be tab-indented as a instruction block.

if cmd=='snow'
	sky::weather = cmd
	$.ambient Color::Gray

If you want one set of instructions to run if the condition is true and another otherwise, add an else command and corresponding instruction block.

if actor::mood == 'happy'
	actor::action = 'dance'
	actor::eyes = 'cry'

Use the elif command in between the if and else, should you need to check additional condition and perform the corresponding instructions. An 'if' structure can have arbitrarily many 'elif' sections as needed.

moving = true
if x<0.0
	speed = 0.0
	moving = false
elif y<3.0
	speed = 1.0
	speed = 5.0

switch and match commands These commands are a more concise alternative to if elif commands where we want to evaluate the same single value against multiple possibilities. The switch and match commands are identical, except that switch uses the '==' equivalence operator to compare and match uses the '=~' match operator.

match term_speed
	ship.state = 'landing'
	ship_state = 'crashing'
	ship_state = 'flattened'

while command: Play it Again, Sam!

The while command is used when we want to execute the same tab-indented block of instructions over and over again. A condition is provided, so that we can stop the repetition whenever the condition fails. The condition is checked right away and every time after the block of instructions has been performed.

x = 0
while x<=5
	$env.log "WARNING, WARNING! Will Robinson"

The break statement can be placed anywhere within the while instruction block to offer another way to stop performing the instructions.

while true    # without the break, this loop would run forever
	actor.walkstep  # will happen at least once
	break 'rest' if keypress=='space' # returns the 'rest' value whenever the space bar is pressed

Collection Iterators

Sometimes we have a collection where we want to iterate through each and every element in the collection, performing some operation on that element. The for structure is one way to accomplish this. (Another technique is the .each method, as described in the List chapter).

The for statement lists a local variable to hold the element after 'for' and the value of the collection after 'in'. The tab-indented instructions follow. The break statement may also be used here, if required.

for x in msg_queue
	switch x

As an alternative to the for command, there are several collection methods able to iterate through the members of a collection, and perform a passed function on each one. For example:

list.each txt->   # For every item in the list, this function is called
	$env.log x    # ... logging the item in the log file