Extracting Chinese some texts are not right.

[buptyyf]

Please provide all mandatory information!

Describe the bug (mandatory)

In one pdf, pymupdf extracts some chinese character not right.

Example PDF：
example.pdf

code

doc = fitz.open(pdf_file_path)
doc[1].get_text()

Many characters are not right.

How can I recognize and fix it?

[JorjMcKie]

This is a "Discussions" item, so I will first covert it.

[JorjMcKie]

This problem is caused by the fonts themselves. A text extraction software is dependent on the font's back-translation information, which is contained in the /ToUnicode table or directly given as a standard character encoding scheme, e.g. /WinAnsiEncoding.
Any wrong information there cannot be detected as such.
The only thing you can try is using OCR, sorry.

[buptyyf]

So, I think it's no solution. I have no way to detect this problem, so that I can use OCR to solve it.

[buptyyf]

I have another question. Why pdf reader in mac or chrome can read this pdf correctly?

[JorjMcKie]

I have another question. Why pdf reader in mac or chrome can read this pdf correctly?

Showing this PDF is not the problem. You talked about text extraction. If you create a page pixmap with PyMuPDF, you will get the right picture.
But if using e.g. Adobe Acrobat or any other PDF viewer and then selecting the text with the cursor you will get the same wrong result.