System and methods for improved sorting with national language support

In a computer system for sorting text strings comprising international characters according to a locale-specific sort order, an improved method for comparing characters in text strings which expands from one character to two characters for sorting, the method comprising:

• storing a table comprising a plurality of one-to-two character expansion entries, each entry storing first and second expanded characters which a given one-to-two character expands into for sorting, each entry further storing a flag for indicating whether said given one-to-two character is sorted before or after said first and second expanded characters, said plurality of entries being stored in said table in an order which indicates a sort order for each one-to-two character in the table relative to other one-to-two characters in the table;
• receiving a request for comparing first and second one-to-two characters, said first and second one-to-two characters expanding into identical first and second expanded characters for sorting;
• determining, based on respective entries in said table for said first and second one-to-two characters, which of said first and second text strings is ordered first in a sort performed according to said locale-specific sort order by comparing the order in said table of the respective entries;
• and
• in response to said determining step, specifying which one-to-two character is ordered first in a sort performed according to said locale-specific sort order.

In a computer system for sorting text strings comprising international characters according to a locale-specific sort order, an improved method for comparing characters in text strings which sorted on the basis of character case, the method comprising:

• storing a collation table comprising a plurality of character entries, each entry for a character storing primary, secondary, and tertiary collation weightings, said primary collation weighting indicating for the character a collation weighting based on character family, said secondary collation weighting indicating for the character a collation weighting based on diacritic marks, said tertiary collation weighting indicating for the character a collation weighting based on character case;
• receiving a request for comparing first and second characters;
• and
• determining, based on respective entries in said table, which of said first and second characters is ordered first in a sort performed according to said locale-specific sort order by substeps comprising:
• (1) if a difference in respective primary collation weightings exists, indicating that the character to be ordered first is the character with the greater primary weighting, otherwise
• (2) if a difference in respective secondary collation weightings exists, indicating that the character to be ordered first is the character with the greater secondary weighting, and otherwise
• (3) if a difference in respective tertiary collation weightings exists, indicating that the character to be ordered first is the character with the greater tertiary weighting.

A system for comparing text strings comprising culturally-dependent information, the system comprising:

• a computer having a processor and a memory, said memory storing a collation table and an expansion table;
• said collation table comprising:
• a plurality of character entries, each entry for a character storing primary, secondary, and tertiary collation weightings, said primary collation weighting indicating for the character a collation weighting based on character family, said secondary collation weighting indicating for the character a collation weighting based on diacritic marks, said tertiary collation weighting indicating for the character a collation weighting based on character case;
• said expansion table comprising:
• a plurality of one-to-two character expansion entries, each entry storing first and second expanded characters which a given one-to-two character expands into for sorting, each entry further storing a flag for indicating whether said given one-to-two character is sorted before or after said first and second expanded characters;
• and
• collation means for comparing text strings, said collation means including:
• means for testing whether a net difference in primary collation weightings exists between text strings;
• means for determining whether a character expansion for a particular character is treated as a difference in secondary weighting;
• means for testing whether a net difference in secondary collation weightings exists between text strings;
• and
• means for testing whether a net difference in tertiary collation weightings exists between text strings.