miss char

[beimusky]

There should be 5 characters, including one invisible space. After recognition, only 4 are detected.

555.pdf

[{'matrix': (13, 0, 0, 13, 86.4567, 711.7845), 'fontname': 'IDBHLX+FZY3K--GBK1-0', 'adv': 1.0, 'upright': True, 'x0': 86.4567, 'y0': 708.1575, 'x1': 99.4567, 'y1': 721.1575, 'width': 13.0, 'height': 13.0, 'size': 13.0, 'mcid': None, 'tag': None, 'object_type': 'char', 'page_number': 1, 'ncs': 'ICCBased', 'text': '一', 'stroking_color': None, 'stroking_pattern': None, 'non_stroking_color': (0, 0, 0), 'non_stroking_pattern': None, 'top': 86.7165, 'bottom': 99.7165, 'doctop': 86.7165}, {'matrix': (13, 0, 0, 13, 99.5867, 711.7845), 'fontname': 'IDBHLX+FZY3K--GBK1-0', 'adv': 1.0, 'upright': True, 'x0': 99.5867, 'y0': 708.1575, 'x1': 112.5867, 'y1': 721.1575, 'width': 13.0, 'height': 13.0, 'size': 13.0, 'mcid': None, 'tag': None, 'object_type': 'char', 'page_number': 1, 'ncs': 'ICCBased', 'text': '、', 'stroking_color': None, 'stroking_pattern': None, 'non_stroking_color': (0, 0, 0), 'non_stroking_pattern': None, 'top': 86.7165, 'bottom': 99.7165, 'doctop': 86.7165}, {'matrix': (13, 0, 0, 13, 106.2167, 711.7845), 'fontname': 'IDBHLX+FZY3K--GBK1-0', 'adv': 0.25, 'upright': True, 'x0': 106.2167, 'y0': 708.1575, 'x1': 109.4667, 'y1': 721.1575, 'width': 3.25, 'height': 13.0, 'size': 13.0, 'mcid': None, 'tag': None, 'object_type': 'char', 'page_number': 1, 'ncs': 'ICCBased', 'text': '\u2007', 'stroking_color': None, 'stroking_pattern': None, 'non_stroking_color': (0, 0, 0), 'non_stroking_pattern': None, 'top': 86.7165, 'bottom': 99.7165, 'doctop': 86.7165}, {'matrix': (13, 0, 0, 13, 113.3017, 711.7845), 'fontname': 'ARAGDN+FZY3K--GBK1-0', 'adv': 0.611, 'upright': True, 'x0': 113.3017, 'y0': 708.1575, 'x1': 121.2447, 'y1': 721.1575, 'width': 7.942999999999998, 'height': 13.0, 'size': 13.0, 'mcid': None, 'tag': None, 'object_type': 'char', 'page_number': 1, 'ncs': 'ICCBased', 'text': 'A', 'stroking_color': None, 'stroking_pattern': None, 'non_stroking_color': (0, 0, 0), 'non_stroking_pattern': None, 'top': 86.7165, 'bottom': 99.7165, 'doctop': 86.7165}]

[mkl-public]

The whole content of that page is

BT
    /CS0 cs
    0 0 0 scn
    /Perceptual ri
    /GS0 gs
    /C2_0 1 Tf
    -0.49 Tc
    0.49 Tw
    13 0 0 13 86.4567 711.7845 Tm
    [<032d> -500 <02200001> ] TJ
    /TT0 1 Tf
    0 Tc
    0 Tw
    2.065 0 Td
    (A) Tj
ET

Thus, only the 3 characters from font C2_0 with character codes 0x032d, 0x0220, and 0x0001 and the 1 character from font TT0 with character code 'A' are drawn.
Why do you assume that another character should be detected?

[beimusky]

The whole content of that page is

BT
    /CS0 cs
    0 0 0 scn
    /Perceptual ri
    /GS0 gs
    /C2_0 1 Tf
    -0.49 Tc
    0.49 Tw
    13 0 0 13 86.4567 711.7845 Tm
    [<032d> -500 <02200001> ] TJ
    /TT0 1 Tf
    0 Tc
    0 Tw
    2.065 0 Td
    (A) Tj
ET

Thus, only the 3 characters from font C2_0 with character codes 0x032d, 0x0220, and 0x0001 and the 1 character from font TT0 with character code 'A' are drawn. Why do you assume that another character should be detected?

import pdfplumber
import fitz # PyMuPDF

if name == 'main':
pdf_path = r"D:\code\adobe-pagemaker\app\check_pdf_publish\example\555.pdf"

a = 0
with pdfplumber.open(pdf_path) as pdf:
    for page_number, page in enumerate(pdf.pages, start=1):
        chars = page.chars  # 获取所有字符及其位置信息
        for char in chars:
            a += 1
b = 0
with fitz.open(pdf_path) as pdf:
    for page_number, page in enumerate(pdf.pages(), start=1):

        raw_dict = page.get_text("rawdict")
        for block in raw_dict.get("blocks", []):
            if block["type"] == 0:  # 文本块
                for line in block.get("lines", []):
                    for span in line.get("spans", []):
                        for char in span.get('chars', []):
                            b+=1
print(a==b)

[mkl-public]

Well, as is clear from the content stream quote, there only are 4 characters. So you should not wonder about a missing char but about a char too many.

As there only are very few characters on the page in question, you could improve your code to actually compare the characters you retrieve one way or another.

It may be possible that extra characters are inferred from typesetting details. E.g. a gap (like the one created by the -500 in the TJ argument) might give rise to an extra space character.