annotate tests/test_mupdf_regressions.py @ 46:7ee69f120f19 default tip

>>>>> tag v1.26.5+1 for changeset b74429b0f5c4
author Franz Glasner <fzglas.hg@dom66.de>
date Sat, 11 Oct 2025 17:17:30 +0200
parents 1d09e1dec1d9
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
1
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
1 import pymupdf
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
2 import os
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
3 import gentle_compare
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
4
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
5 scriptdir = os.path.abspath(os.path.dirname(__file__))
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
6
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
7
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
8 def test_707448():
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
9 """Confirm page content cleaning does not destroy page appearance."""
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
10 filename = os.path.join(scriptdir, "resources", "test-707448.pdf")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
11 doc = pymupdf.open(filename)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
12 page = doc[0]
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
13 words0 = page.get_text("words")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
14 page.clean_contents(sanitize=True)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
15 words1 = page.get_text("words")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
16 assert gentle_compare.gentle_compare(words0, words1)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
17
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
18
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
19 def test_707673():
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
20 """Confirm page content cleaning does not destroy page appearance.
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
21
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
22 Fails starting with MuPDF v1.23.9.
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
23
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
24 Fixed in:
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
25 commit 779b8234529cb82aa1e92826854c7bb98b19e44b (golden/master)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
26 """
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
27 filename = os.path.join(scriptdir, "resources", "test-707673.pdf")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
28 doc = pymupdf.open(filename)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
29 page = doc[0]
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
30 words0 = page.get_text("words")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
31 page.clean_contents(sanitize=True)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
32 words1 = page.get_text("words")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
33 ok = gentle_compare.gentle_compare(words0, words1)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
34 assert ok
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
35
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
36
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
37 def test_707727():
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
38 """Confirm page content cleaning does not destroy page appearance.
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
39
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
40 MuPDF issue: https://bugs.ghostscript.com/show_bug.cgi?id=707727
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
41 """
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
42 filename = os.path.join(scriptdir, "resources", "test_3362.pdf")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
43 doc = pymupdf.open(filename)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
44 page = doc[0]
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
45 pix0 = page.get_pixmap()
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
46 page.clean_contents(sanitize=True)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
47 page = doc.reload_page(page) # required to prevent re-use
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
48 pix1 = page.get_pixmap()
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
49 rms = gentle_compare.pixmaps_rms(pix0, pix1)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
50 print(f'{rms=}', flush=1)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
51 pix0.save(os.path.normpath(f'{__file__}/../../tests/test_707727_pix0.png'))
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
52 pix1.save(os.path.normpath(f'{__file__}/../../tests/test_707727_pix1.png'))
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
53 if pymupdf.mupdf_version_tuple >= (1, 25, 2):
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
54 # New sanitising gives small fp rounding errors.
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
55 assert rms < 0.05
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
56 else:
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
57 assert rms == 0
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
58
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
59
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
60 def test_707721():
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
61 """Confirm text extraction works for nested MCID with Type 3 fonts.
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
62 PyMuPDF issue https://github.com/pymupdf/PyMuPDF/issues/3357
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
63 MuPDF issue: https://bugs.ghostscript.com/show_bug.cgi?id=707721
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
64 """
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
65 filename = os.path.join(scriptdir, "resources", "test_3357.pdf")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
66 doc = pymupdf.open(filename)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
67 page = doc[0]
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
68 ok = page.get_text()
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
69 assert ok
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
70
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
71
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
72 def test_3376():
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
73 """Check fix of MuPDF bug 707733.
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
74
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
75 https://bugs.ghostscript.com/show_bug.cgi?id=707733
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
76 PyMuPDF issue https://github.com/pymupdf/PyMuPDF/issues/3376
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
77
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
78 Test file contains a redaction for the first 3 words: "Table of Contents".
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
79 Test strategy:
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
80 - extract all words (sorted)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
81 - apply redactions
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
82 - extract words again
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
83 - confirm: we now have 3 words less and remaining words are equal.
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
84 """
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
85 filename = os.path.join(scriptdir, "resources", "test_3376.pdf")
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
86 doc = pymupdf.open(filename)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
87 page = doc[0]
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
88 words0 = page.get_text("words", sort=True)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
89 words0_s = words0[:3] # first 3 words
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
90 words0_e = words0[3:] # remaining words
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
91 assert " ".join([w[4] for w in words0_s]) == "Table of Contents"
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
92
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
93 page.apply_redactions()
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
94
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
95 words1 = page.get_text("words", sort=True)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
96
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
97 ok = gentle_compare.gentle_compare(words0_e, words1)
1d09e1dec1d9 ADD: PyMuPDF v1.26.4: the original sdist.
Franz Glasner <fzglas.hg@dom66.de>
parents:
diff changeset
98 assert ok