Skip to content

Parsers API Reference

Base Parser

Parser

Bases: ABC

Source code in src/agentpack/parsers/base.py
5
6
7
8
9
class Parser(ABC):
    @abstractmethod
    def parse(self, file_path: Path, source_id: str) -> SourceDocument:
        """Parses a file and returns a Canonical SourceDocument."""
        pass

parse(file_path, source_id) abstractmethod

Parses a file and returns a Canonical SourceDocument.

Source code in src/agentpack/parsers/base.py
6
7
8
9
@abstractmethod
def parse(self, file_path: Path, source_id: str) -> SourceDocument:
    """Parses a file and returns a Canonical SourceDocument."""
    pass

PDF Parser

Docling Parser

DoclingParser: handle docx, pptx, xlsx, and html via Docling's structured tree. Reuses the module-level DocumentConverter singleton from pdf_parser.

Markdown Parser

CSV Parser

Text Parser