doc-reader

To extract text from a PDF or Word document, you can use format-specific parsers that return the content in a structured JSON envelope. This envelope includes the file format, character count, and the extracted text, making it immediately usable for downstream data processing.

Yes, you can use OCR to extract text from images. The system automatically falls back to OCR when text is not readily extractable from image-based formats, ensuring you can digitize and analyze content from scanned documents or image files.

The best way to normalize extracted text from diverse file formats is to use a unified JSON envelope. It standardizes outputs across PDFs, Word, Excel, and images into common fields like status, format, char_count, and text, ensuring consistent data extraction.

Yes, text extraction works with structured data formats like CSV, JSON, and YAML. The system uses format-specific parsers to read these files and returns the content within the same standard JSON envelope, alongside support for HTML, XML, and source code.

A key limitation of document text extraction for large files is truncation. The unified JSON envelope includes a 'truncated' field to indicate if the extracted text exceeds the processing limits, ensuring you know when the output data is incomplete.