camelCase_splitter.md

camelCase_splitter

camelCase_splitter is a character filter that splits camelCase words. It is an alternative to OpenSearch's word_delimiter and word_delimiter_graph, which can be too aggressive about splitting text.

The Elasticsearch docs provide an example pattern_replace filter for splitting camelCase words, but it doesn't handle certain corner cases.

camelCase_splitter also accounts for modifier characters (Unicode regex \p{M}) and formatting characters (Unicode regex \p{Cf}) between the lowercase and uppercase letters it splits on.

• Hidden bidirectional markers, soft hyphens, and various zero-width joiners and non-joiners will not affect the camelCase splitting, so camel-Case with an normally invisible soft hyphen would be correctly split.

• Letters like e̪, which is technically two characters (e + ̪ ), are treated as a single grapheme, so that Me̪Me would be split into Me̪ and Me.

camelCase_splitter also handles 32-bit Unicode characters, so 𝒜𝓍𝔸𝕩 is split into 𝒜𝓍 and 𝔸𝕩.

Example

index :
    analysis :
        analyzer :
            camelCaseless_text :
                type : custom
                char_filter : [camelCase_splitter]
                tokenizer : standard
                filter : [lowercase]

This will produce the token camel and case for the input text camelCase.