Function get_text

src/utils.py:401–507

Extract text from a page or an annotation. This is a unifying wrapper for various methods of the pymupdf.TextPage class. Args: option: (str) text, words, blocks, html, dict, json, rawdict, xhtml or xml. clip: (rect-like) restrict output to this area. flags: bit swit

(
    page: pymupdf.Page,
    option: str = "text",
    *,
    clip: rect_like = None,
    flags: OptInt = None,
    textpage: pymupdf.TextPage = None,
    sort: bool = False,
    delimiters=None,
    tolerance=3,
)

def get_text(
    page: pymupdf.Page,
    option: str = "text",
    *,
    clip: rect_like = None,
    flags: OptInt = None,
    textpage: pymupdf.TextPage = None,
    sort: bool = False,
    delimiters=None,
    tolerance=3,
):
    """Extract text from a page or an annotation.

    This is a unifying wrapper for various methods of the pymupdf.TextPage class.

    Args:
        option: (str) text, words, blocks, html, dict, json, rawdict, xhtml or xml.
        clip: (rect-like) restrict output to this area.
        flags: bit switches to e.g. exclude images or decompose ligatures.
        textpage: reuse this pymupdf.TextPage and make no new one. If specified,
            'flags' and 'clip' are ignored.

    Returns:
        the output of methods get_text_words / get_text_blocks or pymupdf.TextPage
        methods extractText, extractHTML, extractDICT, extractJSON, extractRAWDICT,
        extractXHTML or etractXML respectively.
        Default and misspelling choice is "text".
    """
    formats = {
        "text": pymupdf.TEXTFLAGS_TEXT,
        "html": pymupdf.TEXTFLAGS_HTML,
        "json": pymupdf.TEXTFLAGS_DICT,
        "rawjson": pymupdf.TEXTFLAGS_RAWDICT,
        "xml": pymupdf.TEXTFLAGS_XML,
        "xhtml": pymupdf.TEXTFLAGS_XHTML,
        "dict": pymupdf.TEXTFLAGS_DICT,
        "rawdict": pymupdf.TEXTFLAGS_RAWDICT,
        "words": pymupdf.TEXTFLAGS_WORDS,
        "blocks": pymupdf.TEXTFLAGS_BLOCKS,
    }
    option = option.lower()
    assert option in formats
    if option not in formats:
        option = "text"
    if flags is None:
        flags = formats[option]

    if option == "words":
        return get_text_words(
            page,
            clip=clip,
            flags=flags,
            textpage=textpage,
            sort=sort,
            delimiters=delimiters,
        )
    if option == "blocks":
        return get_text_blocks(