gnu: Add python-pypdf.

* gnu/packages/pdf.scm (python-pypdf): New variable.
* gnu/packages/patches/python-pypdf-annotate-tests-appropriately.patch: New
file.
* gnu/local.mk (dist_patch_DATA): Register patch.

Signed-off-by: Maxim Cournoyer <maxim.cournoyer@gmail.com>

3 files changed, 126 insertions, 0 deletions

diff --git a/gnu/local.mk b/gnu/local.mk
index 3da47858ff..310c9f1f90 100644
--- a/gnu/local.mk
+++ b/gnu/local.mk
@@ -1639,6 +1639,7 @@ dist_patch_DATA =						\
   %D%/packages/patches/python-louvain-fix-test.patch		\
   %D%/packages/patches/python-random2-getrandbits-test.patch		\
   %D%/packages/patches/python-poppler-qt5-fix-build.patch	\
+  %D%/packages/patches/python-pypdf-annotate-tests-appropriately.patch	\
   %D%/packages/patches/python-telingo-fix-comparison.patch	\
   %D%/packages/patches/python-w3lib-fix-test-failure.patch	\
   %D%/packages/patches/sdcc-disable-non-free-code.patch		\
diff --git a/gnu/packages/patches/python-pypdf-annotate-tests-appropriately.patch b/gnu/packages/patches/python-pypdf-annotate-tests-appropriately.patch
new file mode 100644
index 0000000000..14f1f73924
--- /dev/null
+++ b/gnu/packages/patches/python-pypdf-annotate-tests-appropriately.patch
@@ -0,0 +1,96 @@
+Origin: https://github.com/py-pdf/pypdf/commit/767047b98ee3ea7aca331cfbd63502a284bfed93
+From 767047b98ee3ea7aca331cfbd63502a284bfed93 Mon Sep 17 00:00:00 2001
+From: dkg <dkg@fifthhorseman.net>
+Date: Sat, 14 Jan 2023 03:32:45 -0500
+Subject: [PATCH 03/14] Annotate tests appropriately (#1551)
+
+By annotating these tests, we can use pytest markers to skip external
+tests and tests that depend on sample-files.
+---
+ tests/test_reader.py | 5 +++++
+ tests/test_writer.py | 4 ++++
+ 2 files changed, 9 insertions(+)
+
+diff --git a/tests/test_reader.py b/tests/test_reader.py
+index 710e6c5..62eb7b7 100644
+--- a/tests/test_reader.py
++++ b/tests/test_reader.py
+@@ -176,6 +176,7 @@ def test_get_outline(src, outline_elements):
+     assert len(outline) == outline_elements
+ 
+ 
++@pytest.mark.samples
+ @pytest.mark.parametrize(
+     ("src", "expected_images"),
+     [
+@@ -866,6 +867,7 @@ def test_get_fields():
+     assert dict(fields["c1-1"]) == ({"/FT": "/Btn", "/T": "c1-1"})
+ 
+ 
++@pytest.mark.external
+ def test_get_full_qualified_fields():
+     url = "https://github.com/py-pdf/PyPDF2/files/10142389/fields_with_dots.pdf"
+     name = "fields_with_dots.pdf"
+@@ -1214,6 +1216,7 @@ def test_zeroing_xref():
+     len(reader.pages)
+ 
+ 
++@pytest.mark.external
+ def test_thread():
+     url = "https://github.com/py-pdf/pypdf/files/9066120/UTA_OSHA_3115_Fall_Protection_Training_09162021_.pdf"
+     name = "UTA_OSHA.pdf"
+@@ -1226,6 +1229,7 @@ def test_thread():
+     assert len(reader.threads) >= 1
+ 
+ 
++@pytest.mark.external
+ def test_build_outline_item(caplog):
+     url = "https://github.com/py-pdf/pypdf/files/9464742/shiv_resume.pdf"
+     name = "shiv_resume.pdf"
+@@ -1253,6 +1257,7 @@ def test_build_outline_item(caplog):
+     assert "Unexpected destination 2" in exc.value.args[0]
+ 
+ 
++@pytest.mark.samples
+ @pytest.mark.parametrize(
+     ("src", "page_labels"),
+     [
+diff --git a/tests/test_writer.py b/tests/test_writer.py
+index 60b4a17..20c4de0 100644
+--- a/tests/test_writer.py
++++ b/tests/test_writer.py
+@@ -930,6 +930,7 @@ def test_startup_dest():
+     pdf_file_writer.open_destination = None
+ 
+ 
++@pytest.mark.external
+ def test_iss471():
+     url = "https://github.com/py-pdf/pypdf/files/9139245/book.pdf"
+     name = "book_471.pdf"
+@@ -942,6 +943,7 @@ def test_iss471():
+     )
+ 
+ 
++@pytest.mark.external
+ def test_reset_translation():
+     url = "https://corpora.tika.apache.org/base/docs/govdocs1/924/924666.pdf"
+     name = "tika-924666.pdf"
+@@ -977,6 +979,7 @@ def test_threads_empty():
+     assert thr == thr2
+ 
+ 
++@pytest.mark.external
+ def test_append_without_annots_and_articles():
+     url = "https://corpora.tika.apache.org/base/docs/govdocs1/924/924666.pdf"
+     name = "tika-924666.pdf"
+@@ -993,6 +996,7 @@ def test_append_without_annots_and_articles():
+     assert len(writer.threads) >= 1
+ 
+ 
++@pytest.mark.external
+ def test_append_multiple():
+     url = "https://corpora.tika.apache.org/base/docs/govdocs1/924/924666.pdf"
+     name = "tika-924666.pdf"
+-- 
+2.39.1
+
diff --git a/gnu/packages/pdf.scm b/gnu/packages/pdf.scm
index 4b9d0623fe..8f3bec35d6 100644
--- a/gnu/packages/pdf.scm
+++ b/gnu/packages/pdf.scm
@@ -1260,6 +1260,35 @@ documents that use the standard security handler.")
 converter using the Poppler and Cairo libraries.")
     (license license:gpl2+)))
 
+(define-public python-pypdf
+  (package
+    (name "python-pypdf")
+    (version "3.2.1")
+    (source (origin
+              (method git-fetch)
+              (uri (git-reference
+                    (url "https://github.com/py-pdf/pypdf")
+                    (commit version)))
+              (file-name (git-file-name name version))
+              (sha256
+               (base32
+                "1qwvjr694sabfblx22zd54b9ny40f2gbv3bv6q43myrlxwvvisk6"))
+              (patches (search-patches
+                        "python-pypdf-annotate-tests-appropriately.patch"))))
+    (build-system pyproject-build-system)
+    (native-inputs (list python-pytest python-flit))
+    (propagated-inputs (list python-typing-extensions))
+    (home-page "https://github.com/py-pdf/pypdf")
+    (arguments
+     (list
+      ;; Disable tests that use the network and non-free assets.
+      #:test-flags #~(list "-m" "not external and not samples")))
+    (synopsis "Python PDF library")
+    (description
+     "This package provides a PDF library capable of splitting, merging,
+cropping, and transforming PDF files.")
+    (license license:bsd-3)))
+
 (define-public python-pypdf2
   (package
     (name "python-pypdf2")