Function scrub

src_classic/utils.py:4146–4296

(
    doc: Document,
    attached_files: bool = True,
    clean_pages: bool = True,
    embedded_files: bool = True,
    hidden_text: bool = True,
    javascript: bool = True,
    metadata: bool = True,
    redactions: bool = True,
    redact_images: int = 0,
    remove_links: bool = True,
    reset_fields: bool = True,
    reset_responses: bool = True,
    thumbnails: bool = True,
    xml_metadata: bool = True,
)

# Acrobat 'sanitize' function
# ------------------------------------------------------------------------------
def scrub(
    doc: Document,
    attached_files: bool = True,
    clean_pages: bool = True,
    embedded_files: bool = True,
    hidden_text: bool = True,
    javascript: bool = True,
    metadata: bool = True,
    redactions: bool = True,
    redact_images: int = 0,
    remove_links: bool = True,
    reset_fields: bool = True,
    reset_responses: bool = True,
    thumbnails: bool = True,
    xml_metadata: bool = True,
) -> None:
    def remove_hidden(cont_lines):
        """Remove hidden text from a PDF page.

        Args:
            cont_lines: list of lines with /Contents content. Should have status
                from after page.cleanContents().

        Returns:
            List of /Contents lines from which hidden text has been removed.

        Notes:
            The input must have been created after the page's /Contents object(s)
            have been cleaned with page.cleanContents(). This ensures a standard
            formatting: one command per line, single spaces between operators.
            This allows for drastic simplification of this code.
        """
        out_lines = []  # will return this
        in_text = False  # indicate if within BT/ET object
        suppress = False  # indicate text suppression active
        make_return = False
        for line in cont_lines:
            if line == b"BT":  # start of text object
                in_text = True  # switch on
                out_lines.append(line)  # output it
                continue
            if line == b"ET":  # end of text object
                in_text = False  # switch off
                out_lines.append(line)  # output it
                continue
            if line == b"3 Tr":  # text suppression operator
                suppress = True  # switch on
                make_return = True
                continue
            if line[-2:] == b"Tr" and line[0] != b"3":
                suppress = False  # text rendering changed
                out_lines.append(line)
                continue
            if line == b"Q":  # unstack command also switches off
                suppress = False
                out_lines.append(line)
                continue
            if suppress and in_text:  # suppress hidden lines