fix: drop redundant final h to recognize rectangles

dhdaines · Nov 25, 2024 · 28463f2 · 28463f2
1 parent 1a8bd2f
commit 28463f2
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 0 deletions.
diff --git a/pdfminer/converter.py b/pdfminer/converter.py
@@ -162,6 +162,11 @@ def paint_path(
                 for o, p in zip(operators, transformed_points)
             ]
 
+            # Drop redundant final "h" on an already closed path
+            if len(shape) > 3 and shape[-1] == "h" and pts[-2] == pts[0]:
+                shape = shape[:-1]
+                pts.pop()
+
             if shape in {"mlh", "ml"}:
                 # single line segment
                 #

diff --git a/tests/test_converter.py b/tests/test_converter.py
@@ -81,6 +81,18 @@ def get_types(path):
             ],
         ) == [LTRect]
 
+        # Same but mllllh variation
+        assert get_types(
+            [
+                ("m", 10, 90),
+                ("l", 90, 90),
+                ("l", 90, 10),
+                ("l", 10, 10),
+                ("l", 10, 90),
+                ('h',)
+            ],
+        ) == [LTRect]
+
         # Bowtie shape
         assert get_types(
             [