KeywordChoice

The KeywordChoice token in RCParsing matches one of several specified strings, ensuring none are followed by certain characters, such as letters or digits. It’s ideal for parsing multiple keywords efficiently while enforcing word boundaries.

Overview

The KeywordChoice token is similar to LiteralChoice but adds a boundary check to ensure the matched string is not followed by characters that would make it part of a larger word (e.g., letters or digits). It uses predicates to define forbidden trailing characters and produces the matched construction-specified string as its intermediate value. This token is optimized with a trie for efficient matching.

Example

Here’s an example of parsing multiple keywords:

var builder = new ParserBuilder();

builder.CreateToken("keyword")
    .KeywordChoice("var", "int", "double");

var parser = builder.Build();
var result1 = parser.MatchToken("keyword", "int"); // Succeeds
var result2 = parser.MatchToken("keyword", "integer"); // Fails

With a custom predicate and case-insensitive matching:

builder.CreateToken("keyword")
    .KeywordChoice(
        new[] { "var", "int", "double" },
        predicate: c => char.IsDigit(c),
        comparer: StringComparer.OrdinalIgnoreCase
    );

builder.CreateToken("another_keyword")
    .KeywordChoiceIgnoreCase("while", "for");

var parser = builder.Build();
var result1 = parser.MatchToken("keyword", "INT"); // Succeeds, produces "int" as intermediate value
var result2 = parser.MatchToken("keyword", "int123"); // Fails
var result3 = parser.MatchToken("another_keyword", "FOr"); // Succeeds, produces "for" as intermediate value

Use Cases

• Multiple Keywords: Match a set of language keywords like var, int, or double.
• Boundary Enforcement: Ensure keywords are not part of larger identifiers.
• Case-Insensitive Parsing: Handle keywords regardless of case with StringComparer.

Intermediate Values

The KeywordChoice token produces the matched construction-specified string as its intermediate value:

builder.CreateToken("keyword")
    .KeywordChoiceIgnoreCase("if", "else", "while");

var parser = builder.Build();
var result = parser.MatchToken("keyword", "elSE!");
var value = result.GetIntermediateValue<string>(); // "else"

Notes

• Default Predicates: Uses ASCII or Unicode identifier predicates by default (e.g., letters, digits, or underscore).
• Trie-Based: Uses a trie for efficient matching of multiple strings.
• Custom Predicates: Define custom trailing character checks with a predicate function.

Related Tutorials

• Keyword Token
• LiteralChoice Token
• Identifier Token