src/__init__.py tests/: fix pymupdf.get_text() with method='single'.

We were ignoring the <pages> arg. Fixes #4524.

Author: julian-smith-artifex-com

src/__init__.py

@@ -20635,7 +20635,10 @@ def apply_pages(
             initfn(*initfn_args, **initfn_kwargs)
         ret = list()
         document = Document(path)
-        for page in document:
+        if pages is None:
+            pages = range(len(document))
+        for pno in pages:
+            page = document[pno]
             r = pagefn(page, *pagefn_args, **initfn_kwargs)
             ret.append(r)
 

tests/test_textextract.py

@@ -309,6 +309,17 @@ def llen(texts):
         pymupdf._log_items_clear()
 
 
+def test_4524():
+    path = os.path.abspath(f'{__file__}/../../tests/resources/mupdf_explored.pdf')
+    print('')
+    document = pymupdf.Document(path)
+    texts_single = pymupdf.get_text(path, method='single', pages=[1, 3, 5])
+    texts_mp = pymupdf.get_text(path, method='mp', pages=[1, 3, 5])
+    print(f'{len(texts_single)=}')
+    print(f'{len(texts_mp)=}')
+    assert texts_mp == texts_single
+
+
 def test_3594():
     verbose = 0
     print()