BUG: catch the case where w[0] is an IndirectObject instead of an int (#2154)

Closes #2137

Author: rchen19

pypdf/_cmap.py

@@ -410,7 +410,7 @@ def compute_space_width(
         else:
             w = []
         while len(w) > 0:
-            st = w[0]
+            st = w[0] if isinstance(w[0], int) else w[0].get_object()
             second = w[1].get_object()
             if isinstance(second, int):
                 for x in range(st, second):

tests/test_cmap.py

@@ -191,6 +191,18 @@ def test_unixxx_glyphs():
         assert pat in txt
 
 
+@pytest.mark.enable_socket()
+def test_cmap_compute_space_width():
+    # issue 2137
+    # original file URL:
+    url = "https://arxiv.org/pdf/2005.05909.pdf"
+    # URL from github issue is too long to pass code stype check, use original arxiv URL instead
+    # url = "https://github.com/py-pdf/pypdf/files/12489914/Morris.et.al.-.2020.-.TextAttack.A.Framework.for.Adversarial.Attacks.Data.Augmentation.and.Adversarial.Training.in.NLP.pdf"
+    name = "TextAttack_paper.pdf"
+    reader = PdfReader(BytesIO(get_data_from_url(url, name=name)))
+    reader.pages[0].extract_text()  # no error
+
+
 @pytest.mark.enable_socket()
 def test_tabs_in_cmap():
     """Issue #2173"""