Unexpected result for apply_redactions()

[slevin1994]

Description of the bug

My pymupdf version is 1.24.14.
I tried invoke add_redact_annot(rect) and apply_redactions(images=0) to remove some words from original pdf file. The locations are shown as below:

However words remanins in these rect areas. Meanwhile some other words were removed unexpectly. The result is shown
as below:

I also attached the files mentioned above, along with the rect positions.

original pdf file: redactions_error_sample.pdf

result file after add_redact_annot(rect):
before_apply_redactions.pdf

result after apply_redactions(images=0):
after_apply_redactions.pdf

rect areas: [{"rect":[164,213,282,227]},{"rect":[282,213,397,233]},{"rect":[434,209,525,243]},{"rect":[169,228,231,243]},{"rect":[377,592,440,607]},{"rect":[373,611,444,626]}]

How to reproduce the bug

doc = fitz.open("redactions_error_sample.pdf")
page = doc.load_page(0)
for rect in rects:
  page.add_redact_annot(rect)
doc.save("before_apply_redactions.pdf")
page.apply_redactions(images=0)
doc.save("after_apply_redactions.pdf")

PyMuPDF version

1.24.14

Operating system

MacOS

Python version

3.9

[jamie-lemon]

I thought your rects array was dubious. This worked for me:

import pymupdf

doc = pymupdf.open("redactions_error_sample.pdf")
page = doc.load_page(0)

rects = [[164,213,282,227],[282,213,397,233],[434,209,525,243],[169,228,231,243],[377,592,440,607],[373,611,444,626]]

for rect in rects:
  page.add_redact_annot(rect, fill=(1,0,0))

# Apply the redactions to the current page, but ignore images
page.apply_redactions(images=0)

doc.save("after_apply_redactions.pdf")

doc.close()

[JorjMcKie]

This is an upstream error. The corresponding MuPDF bug rreport: https://bugs.ghostscript.com/show_bug.cgi?id=708157.