Module org.seasar.doma.core
Package org.seasar.doma.internal.util

Class SqlTokenUtil

java.lang.Object
org.seasar.doma.internal.util.SqlTokenUtil
public final class SqlTokenUtil extends Object
High-performance SQL token character classification utility.

This class provides optimized character classification methods for SQL parsing using pre-computed lookup tables for maximum performance.

Performance Benefits

  • Constant Time: O(1) lookup for ASCII characters (0-127)
  • Branch Reduction: Eliminates multiple if-else chains
  • Cache Friendly: Small lookup tables fit in CPU cache
  • JIT Optimization: Simple array access is highly optimizable

Design Principles

  • ASCII Optimization: Most SQL content is ASCII, so we optimize for this case
  • Bit Flags: Use byte bit flags for efficient character property storage
  • Static Initialization: Pre-compute all classifications at class load time
  • Compatibility: Maintains exact behavior compatibility with ClassicSqlTokenUtil

  • Method Details

    • isWordPart

      public static boolean isWordPart(char c)
      Determines if a character can be part of an SQL word.

      This method uses the same logic as ClassicSqlTokenUtil: a character is considered a word part unless it's whitespace or one of the specific SQL delimiters. This method is optimized for ASCII characters with O(1) lookup time.

      Parameters:
      c - the character to test
      Returns:
      true if the character can be part of an SQL word

    • isWhitespace

      public static boolean isWhitespace(char c)
      Determines if a character is SQL whitespace.

      SQL whitespace includes space, tab, form feed, and other control characters that should be treated as separators in SQL parsing.

      Parameters:
      c - the character to test
      Returns:
      true if the character is SQL whitespace