acrucetta / c_compiler Goto Github PK

1. Write a Jack Tokenizer
2. Write a basic compilation engine that handles all the features, except for expressions, and array-oriented statements.
3. Extend the compilation engine to handle expressions.
4. Extend the compilation engine to handle array-oriented statements.

Keyword:

• class
• constructor
• function
• method
• field
• static
• var
• int
• char
• boolean
• void
• true
• false ...

Symbol:

• { or }
• ( or ) etc...

Identifier:

• String of letters, digits, and underscores that does not start with a digit.

Integer Constant:

• A sequence of digits from 0 to 9.

String Constant:

• A sequence of characters not including double quote or newline.

Stage 1: Symbol Table

• Add to each identifier: name, type, kind, and running index.
• The symbol table is basically a list of hash tables, one for each scope.
  • Each reflects a single scope nested within the next one in the list
  • Compilers look for symbol in the first table in the list, if it fails, it moves to the next table and looks for it (inner scopes outward)
• We have different types:
  • static: class level
  • field: class level
  • local: subroutine level
  • argument: subroutine level
• A list of hash tables, each representing a single scope, nested within the next one on the list.
• We need to define if the identifier is being defined (var declaration) or used.

https://github.com/timiscoding/JackCompiler/blob/master/SymbolTable.js