In score = base + 6 * rate, what token kind does 6 have?

• A number token
• A name token
• A times token

In this front end, what is the lexer's job?

• Execute machine instructions and record their results
• Group characters into tokens with kinds and text
• Connect tokens into a tree under the grammar rules

Here * binds more tightly than +. The expression a + b * c groups as a + (b * c).

Use a binary-operator tree, where * binds more tightly than +. Which tree represents fee + hours * rate?

• Root * has left child + with children fee and hours, and right child rate
• Root + has left child fee and right child * with children hours and rate
• Root + has three direct leaf children fee, hours and rate, with no * node

Ignore spaces and keep names whole. How does the lexer split score = base + 6 * rate?

• score = base, +, 6 * rate: three tokens containing whole expression pieces
• s, c, o, r, e, =, b, a, s, e, +, 6, *, r, a, t, e: seventeen character tokens
• score, =, base, +, 6, *, rate: seven tokens with names kept whole

Subtraction is left associative. How does unparenthesised a - b - c group?

• It groups as a - (b - c)
• It has no grouping until runtime
• It groups as (a - b) - c

Here term has no unparenthesised subtraction and | separates alternatives. Which rule makes subtraction left associative?

• expr = expr - term | term
• expr = term - expr | term
• expr = term - term | term

Subtraction is left associative. A parser groups 9 - 4 - 1 as 9 - (4 - 1), giving 6. Which diagnosis fits?

• The parser used right grouping; the required left grouping gives 4
• The lexer must delete the final minus; deletion preserves the required grouping
• Subtraction is commutative here; changing operand order always preserves its value