Scanner

Roles: > Source code -> Scanner -> Tokens Only understand the syntax rules not the meaning.

Typical tokens: - Identifier - Keyword - Operators - Separators - literals

Glossary - Lexeme of a token == The character sequence forming the token. - Pattern: a rules describing the set of lexemes that can represent a particular token

Parser

Linear-time algorithm suffices to parse essentially all languages in practice.

There are two types of parsers: - Top-down parsers => easy to implement. - Bottom-up parsers => supports more powerful grammars