Features Overview Leave feedback

GroupDocs.Parser for Python via .NET extracts data from a wide range of supported document formats. Every operation follows the same workflow: open a document with Parser, call an extraction method such as get_text(), get_tables() or get_metadata(), and read the result. A method returns None when the document’s format does not support that operation, so one code path serves every format.

Extract the plain text of a whole document or of one page. The accurate mode keeps the reading order and layout; the raw mode is faster and fits pipelines where speed matters more than layout. Scanned pages and images are read with OCR. See Extract text from documents, Extract text in Accurate mode, and Extract text in Raw mode.

get_formatted_text() returns HTML, Markdown or plain text that keeps the document’s headings, numbered and bulleted lists, tables, hyperlinks, and bold and italic text, ready for an HTML preview or a Markdown pipeline. See Working with Text.

Search a document by keyword or regular expression and get the text around each hit, read its structure (paragraphs, headings, lists, tables and hyperlinks) as XML, or get its text areas with their coordinates and text style. See Search text, Extract highlights, Extract text structure, and Extract text areas.

Extract the tables of PDF, Word, spreadsheet and presentation pages with their rows and cells, from a whole document or one page. See Extract tables from document and Extract tables from document page.

Extract the images of a document, a page or a rectangular page area, and save them to files in their own format or another. Hyperlinks are extracted the same way. See Extract Images from Documents, Extract images to files, and Extract hyperlinks from document.

Read document properties such as title, author and dates with get_metadata(), the table of contents with get_toc(), and the file type, page count and size with get_document_info(). List every format the library reads with FileType.get_supported_file_types(). See Extract Metadata from Documents, Get document info, and Get supported file formats.

Read the comments and notes reviewers leave in PDF, Word and PowerPoint files. Call get_annotations() for the whole document, or get_annotations(page_index) for one page; each annotation carries its text, type, page, author and timestamp. To get the annotations’ text inside the extracted text, set TextOptions.include_annotations. See Working with annotations.

Describe the fields and tables of documents that share a layout, such as invoices, once in a template, then pass it to parse_by_template() to read their values from every such document. See Parse Data from Documents and Working with Templates.

get_container() lists the documents inside ZIP, RAR, 7Z and TAR archives, email attachments, Outlook data files and PDF portfolios, and each item opens with a Parser of its own. See Extract Data from Attachments and ZIP Archives and Working with Containers.

Open a document from a file path or a binary stream, open password-protected documents, and tell the library a document’s format when its name does not. See Load file from stream and Load password-protected documents.

Pass a Python function, or an object with trace, warning and error methods, to ParserSettings to receive the engine’s log messages. A resource handler decides which images an HTML or MHTML document may load. See Logging and Handle loading of external resources.

The package installs the groupdocs-parser command, which extracts text, document info, metadata, images and annotations without writing Python code. See Command Line Interface.

GroupDocs.Parser for Python via .NET runs entirely on your own infrastructure — your documents never leave your environment. The package is a self-contained wheel that bundles everything it needs, so no Microsoft Office, OpenOffice, or separate runtime has to be installed. See System Requirements for the supported platforms and native dependencies.