Jules - improve typing, be able to exclude slow tests, wip on #4928.

src/__init__.py

@@ -13,12 +13,13 @@
 import atexit
 import binascii
 import collections
+import glob
+import importlib.util
 import inspect
 import io
 import math
 import os
 import pathlib
-import glob
 import re
 import string
 import sys
@@ -30,7 +31,6 @@
 import zipfile
 
 from . import extra
-import importlib.util
 
 # Set up g_out_log and g_out_message from environment variables.
 #
@@ -586,6 +586,9 @@ class Annot:
     def __init__(self, annot):
         assert isinstance( annot, mupdf.PdfAnnot)
         self.this = annot
+
+    def __bool__(self):
+        return bool(self.this)
 
     def __repr__(self):
         parent = getattr(self, 'parent', '<>')
@@ -10184,7 +10187,7 @@ def _load_annot(self, name, xref):
             annot = JM_get_annot_by_name(page, name)
         else:
             annot = JM_get_annot_by_xref(page, xref)
-        if annot.m_internal:
+        if annot:
             return Annot(annot)
 
     def _makePixmap(self, doc, ctm, cs, alpha=0, annots=1, clip=None):
@@ -11383,6 +11386,9 @@ def extend_textpage(self, tpage, flags=0, matrix=None):
         mupdf.fz_run_page( page, dev, ctm, mupdf.FzCookie())
         mupdf.fz_close_device( dev)
 
+    def find_tables(self, **kwargs):
+        return table.find_tables(self, **kwargs)
+
     @property
     def first_annot(self):
         """First annotation."""
@@ -25571,8 +25577,7 @@ def _mupdf_devel(make_links=True):
 recover_quad                = utils.recover_quad
 recover_span_quad           = utils.recover_span_quad
 
-from .table import find_tables
-Page.find_tables = find_tables
+from . import table
 
 
 class FitzDeprecation(DeprecationWarning):

tests/test_general.py

@@ -2212,3 +2212,12 @@ def test_4907():
             print(f'{i=}')
             display_list = page.get_displaylist(annots=False)
             text_page = display_list.get_textpage()
+
+def test_4928():
+    path = os.path.normpath(f'{__file__}/../../tests/resources/test_4928.pdf')
+    with pymupdf.open(path) as document:
+        try:
+            document.scrub()
+        except Exception as e:
+            print(f'Ignoring expected exception: {e}')
+

tests/test_memory.py

@@ -1,4 +1,5 @@
 import pymupdf
+import util
 
 import gc
 import os
@@ -17,6 +18,8 @@ def test_2791():
     '''
     Check for memory leaks.
     '''
+    if util.skip_slow_tests('test_2791'):
+        return
     if os.environ.get('PYODIDE_ROOT'):
         print('test_2791(): not running on Pyodide - No module named \'psutil\'.')
         return

tests/test_pixmap.py

@@ -18,6 +18,7 @@
 import pytest
 import textwrap
 import time
+import util
 
 
 scriptdir = os.path.abspath(os.path.dirname(__file__))
@@ -371,6 +372,8 @@ def do(gi):
 
 
 def test_3848():
+    if util.skip_slow_tests('test_3848'):
+        return
     if os.environ.get('PYMUPDF_RUNNING_ON_VALGRIND') == '1':
         # Takes 40m on Github.
         print(f'test_3848(): not running on valgrind because very slow.', flush=1)
@@ -509,6 +512,9 @@ def test_4336():
 
 
 def test_4435():
+    # This is slow, e.g. 224s.
+    if util.skip_slow_tests('test_4435'):
+        return
     print(f'{pymupdf.version=}')
     path = os.path.normpath(f'{__file__}/../../tests/resources/test_4435.pdf')
     with pymupdf.open(path) as document:

tests/test_typing.py

@@ -43,3 +43,6 @@ def test_py_typed():
 
 def _test_4903(page: pymupdf.Page) -> float:  # In 1.27.1, error: Variable "pymupdf.Page" is not valid as a type
     return page.rect.width  # In 1.27.1, error: pymupdf.Page? has no attribute "rect"
+
+def _test_4932(page: pymupdf.Page):
+    page.find_tables()

tests/util.py

@@ -26,3 +26,9 @@ def download(url, name, size=None):
         with open(path, 'wb') as f:
             f.write(r.content)
     return path
+
+def skip_slow_tests(test_name):
+    PYMUPDF_TEST_QUICK = os.environ.get('PYMUPDF_TEST_QUICK')
+    if PYMUPDF_TEST_QUICK == '1':
+        print(f'{test_name}(): skipping test because {PYMUPDF_TEST_QUICK=}.')
+        return True