Interface DocumentSplitter

All Known Implementing Classes:
DocumentByCharacterSplitter, DocumentByLineSplitter, DocumentByParagraphSplitter, DocumentByRegexSplitter, DocumentBySentenceSplitter, DocumentByWordSplitter, HierarchicalDocumentSplitter

public interface DocumentSplitter
Defines the interface for splitting a document into text segments. This is necessary as LLMs have a limited context window, making it impossible to send the entire document at once. Therefore, the document should first be split into segments, and only the relevant segments should be sent to LLM.
DocumentSplitters.recursive() from a dev.langchain4j:langchain4j module is a good starting point.
  • Method Details

    • split

      List<TextSegment> split(Document document)
      Splits a single Document into a list of TextSegment objects. The metadata is typically copied from the document and enriched with segment-specific information, such as position in the document, page number, etc.
      Parameters:
      document - The Document to be split.
      Returns:
      A list of TextSegment objects derived from the input Document.
    • splitAll

      default List<TextSegment> splitAll(List<Document> documents)
      Splits a list of Documents into a list of TextSegment objects. This is a convenience method that calls the split method for each Document in the list.
      Parameters:
      documents - The list of Documents to be split.
      Returns:
      A list of TextSegment objects derived from the input Documents.
    • splitAll

      default List<TextSegment> splitAll(Document... documents)
      Splits multiple Document instances into a list of TextSegment objects.

      This is a convenience method that allows callers to pass a variable number of Document arguments (using varargs) instead of explicitly creating a list. Internally, it delegates to the splitAll(List) method by converting the varargs array into a List.

      For example:

      
       List<TextSegment> segments = documentSplitter.splitAll(doc1, doc2, doc3);
       
      This is equivalent to:
      
       List<TextSegment> segments = documentSplitter.splitAll(Arrays.asList(doc1, doc2, doc3));
       
      Parameters:
      documents - One or more Document instances to be split. If no documents are provided, an empty list is returned.
      Returns:
      A list of TextSegment objects derived from the input documents. The resulting list is a flat combination of all segments from all input documents.
      See Also: