epd-parser

To extract structured data from Environmental Product Declaration PDFs, you can use a parser that processes individual files or entire folders to output normalized JSON containing core EPD metadata like manufacturer, product name, and declared units.

Yes, you can batch process an entire folder of EPD PDFs. The parser handles large document sets by iterating through directories to extract environmental impact data and produce normalized JSON output for AEC sustainability workflows.

Extracted environmental impact fields include manufacturer, product name, declared unit, GWP A1-A3 values, validity dates, and LEED eligibility. This structured data is output as normalized JSON for downstream material specification.

The parser handles scanned or low-text EPD documents by automatically flagging them for manual review. It also flags expired or incomplete documents to ensure data quality and reduce downstream errors in the extraction process.

You can parse EPD PDFs directly or use pre-extracted text fixtures. The tool supports processing both raw PDF files and plain text extracts, providing flexibility depending on your document library setup.

You can validate the accuracy of your EPD data extraction using the built-in self-test mode. This feature checks the extraction logic against document fixtures to ensure the output JSON accurately reflects the source PDFs.