The PyMuPDF4LLM API

*property* version Prints the version of the library.

to_markdown

**(doc: pymupdf.Document | str, *, pages: list | range | None = None, hdr_info: Any = None, write_images: bool = False, embed_images: bool = False, ignore_images: bool = False, ignore_graphics: bool = False, dpi: int = 150, filename=None, image_path='', image_format='png', image_size_limit=0.05, force_text=True, margins=0, page_chunks: bool = False, page_width: float = 612, page_height: float = None, table_strategy='lines_strict', graphics_limit: int = None, ignore_code: bool = False, extract_words: bool = False, show_progress: bool = False, use_glyphs=False)

**→ str | list[dict]**Read the pages of the file and outputs the text of its pages in Markdown format. How this should happen in detail can be influenced by a number of parameters. Please note that there exists support for building page chunks from the Markdown text.

Parameters:

• write_images (bool) – when encountering images or vector graphics, images will be created from the respective page area and stored in the specified folder. Markdown references will be generated pointing to these images. Any text contained in these areas will not be included in the text output (but appear as part of the images). Therefore, if for instance your document has text written on full page images, make sure to set this parameter to False.

• embed_images (bool) – like write_images, but images will be included in the markdown text as base64-encoded strings. Ignores write_images and image_path if used. This may drastically increase the size of your markdown text. • ignore_images (bool) – (New in v.0.0.20) Disregard images on the page. This may help detecting text correctly when pages are very crowded (often the case for documents representing presentation slides). Also speeds up processing time. • ignore_graphics (bool) – (New in v.0.0.20) Disregard vector graphics on the page. This may help detecting text correctly when pages are very crowded (often the case for documents representing presentation slides). Also speeds up processing time. Vector graphics are still used for table detection. • image_size_limit (float) – this must be a positive value less than 1. Images are ignored if width / page.rect.width <= image_size_limit or height / page.rect.height <= image_size_limit. For instance, the default value 0.05 means that to be considered for inclusion, an image’s width and height must be larger than 5% of the page’s width and height, respectively. • dpi (int) – specify the desired image resolution in dots per inch. Relevant only if write_images=True. Default value is 150. • image_path (str) – store images in this folder. Relevant if write_images=True. Default is the path of the script directory. • image_format (str) – specify the desired image format via its extension. Default is “png” (portable network graphics). Another popular format may be “jpg”. Possible values are all supported output formats. • force_text (bool) – generate text output even when overlapping images / graphics. This text then appears after the respective image. If write_images=True this parameter may be False to suppress repetition of text on images.

• doc (*Document,str*) – the file, to be specified either as a file path string, or as a PyMuPDF Document (created via pymupdf.open). In order to use pathlib.Path specifications, Python file-like objects, documents in memory etc. you must use a PyMuPDF Document. • pages (list) – optional, the pages to consider for output (caution: specify 0-based page numbers). If omitted all pages are processed. • hdr_info – optional. Use this if you want to provide your own header detection logic. This may be a callable or an object having a method named get_header_id. It must accept a text span (a span dictionary as contained in extractDICT()) and a keyword parameter “page” (which is the owning Page object). It must return a string “” or up to 6 “#” characters followed by 1 space. If omitted, a full document scan will be performed to find the most popular font sizes and derive header levels based on them. To completely avoid this behavior specify hdr_info=lambda s, page=None: "" or hdr_info=False.

• margins (float,list) – a float or a sequence of 2 or 4 floats specifying page borders. Only objects inside the margins will be considered for output. ◦ margin=f yields (f, f, f, f) for (left, top, right, bottom). ◦ (top, bottom) yields (0, top, 0, bottom). ◦ To always read full pages (default), use margins=0.

• page_chunks (bool) – if True the output will be a list of Document.page_count dictionaries (one per page). Each dictionary has the following structure: ◦ ”metadata” - a dictionary consisting of the document’s metadata Document.metadata, enriched with additional keys “file_path” (the file name), “page_count” (number of pages in document), and “page_number” (1-based page number). ◦ ”toc_items” - a list of Table of Contents items pointing to this page. Each item of this list has the format [lvl, title, pagenumber], where lvl is the hierarchy level, title a string and pagenumber as a 1-based page number. ◦ ”tables” - a list of tables on this page. Each item is a dictionary with keys “bbox”, “row_count” and “col_count”. Key “bbox” is a pymupdf.Rect in tuple format of the table’s position on the page. ◦ ”images” - a list of images on the page. This a copy of page method Page.get_image_info(). ◦ ”graphics” - a list of vector graphics rectangles on the page. This is a list of boundary boxes of clustered vector graphics as delivered by method Page.cluster_drawings(). ◦ ”text” - page content as Markdown text. ◦ ”words” - if extract_words=True was used. This is a list of tuples (x0, y0, x1, y1, "wordstring", bno, lno, wno) as delivered by page.get_text("words"). The sequence of these tuples however is the same as produced in the markdown text string and thus honors multi-column text. This is also true for text in tables: words are extracted in the sequence of table row cells.

• filename (str) – (New in v.0.0.19) Overwrites or sets the desired image file name of written images. Useful when the document is provided as a memory object (which has no inherent file name). • page_width (float) – specify a desired page width. This is ignored for documents with a fixed page width like PDF, XPS etc. Reflowable documents however, like e-books, office [2] or text files have no fixed page dimensions and by default are assumed to have Letter format width (612) and an “infinite” page height. This means that the full document is treated as one large page. • page_height (float) – specify a desired page height. For relevance see the page_width parameter. If using the default None, the document will appear as one large page with a width of page_width. Consequently in this case, no markdown page separators will occur (except the final one), respectively only one page chunk will be returned. • table_strategy (str) – table detection strategy. Default is "lines_strict" which ignores background colors. In some occasions, other strategies may be more successful, for example "lines" which uses all vector graphics objects for detection. Changed in v0.0.19: A value of None will not perform any table detection at all. This may be useful when you know that your document contains no tables. Execution time savings can be significant. • graphics_limit (int) – use this to limit dealing with excess amounts of vector graphics elements. Scientific documents, or pages simulating text via graphics commands may contain tens of thousands of these objects. As vector graphics are analyzed for multiple purposes, runtime may quickly become intolerable. With this parameter, all vector graphics will be ignored if their count exceeds the threshold. Changed in v0.0.19: The page will still be processed, and text, tables and images should be extracted. • ignore_code (bool) – if True then mono-spaced text does not receive special formatting. Code blocks will no longer be generated. This value is set to True if extract_words=True is used. • extract_words (bool) – a value of True enforces page_chunks=True and adds key “words” to each page dictionary. Its value is a list of words as delivered by PyMuPDF’s Page method get_text("words"). The sequence of the words in this list is the same as the extracted text. • show_progress (bool) – Default is False. A value of True displays a text-based progress bar as pages are being converted to Markdown. It will look similar to the following: Processing input.pdf… [==================== ] (148/291)

• use_glyphs (bool) – (New in v.0.0.19) Default is False. A value of True will use the glyph number of the characters instead of the character itself.**Returns:**Either a string of the combined text of all selected document pages, or a list of dictionaries.