Function test_2553

tests/test_general.py:627–668

Ensure identical output across text extractions.

()

def test_2553():
    """Ensure identical output across text extractions."""
    verbose = 0
    doc = pymupdf.open(f"{scriptdir}/resources/test_2553.pdf")
    page = doc[0]

    # extract plain text, build set of all characters
    list1 = page.get_text()
    set1 = set(list1)

    # extract text blocks, build set of all characters
    list2 = page.get_text(sort=True)  # internally uses "blocks"
    set2 = set(list2)
    
    # extract textbox content, build set of all characters
    list3 = page.get_textbox(page.rect)
    set3 = set(list3)
    
    def show(l):
        ret = f'len={len(l)}\n'
        for c in l:
            cc = ord(c)
            if (cc >= 32 and cc < 127) or c == '\n':
                ret += c
            else:
                ret += f' [0x{hex(cc)}]'
        return ret
    
    if verbose:
        print(f'list1:\n{show(list1)}')
        print(f'list2:\n{show(list2)}')
        print(f'list3:\n{show(list3)}')
    
    # all sets must be equal
    assert set1 == set2
    assert set1 == set3

    # With mupdf later than 1.23.4, this special page contains no invalid
    # Unicodes.
    #
    print(f'Checking no occurrence of 0xFFFD, {pymupdf.mupdf_version_tuple=}.')
    assert chr(0xFFFD) not in set1

def test_2553_2():
    doc = pymupdf.open(f"{scriptdir}/resources/test_2553-2.pdf")