Fix StopIteration bug for Python 3.7+ (PEP 479 compliance)

mwxml/element_iterator.py

@@ -55,7 +55,12 @@ def __init__(self, element, pointer):
     def __iter__(self):
 
         while not self.done and self.pointer.depth() > self.depth:
-            event, element = next(self.pointer)
+            try:
+                event, element = next(self.pointer)
+            except StopIteration:
+                # Stream exhausted - this is normal completion
+                # PEP 479: Catch StopIteration to prevent it from escaping the generator
+                break
 
             if event == "start":
                 sub_iterator = ElementIterator(element, self.pointer)
@@ -69,7 +74,12 @@ def __iter__(self):
     def complete(self):
 
         while not self.done and self.pointer.depth() > self.depth:
-            event, element = next(self.pointer)
+            try:
+                event, element = next(self.pointer)
+            except StopIteration:
+                # Stream exhausted - this is normal completion
+                # PEP 479: Catch StopIteration to prevent it from escaping the generator
+                break
             if self.pointer.depth() > self.depth:
                 element.clear()
 

mwxml/iteration/tests/test_stopiteration_bug.py

@@ -0,0 +1,202 @@
+"""
+Test suite for StopIteration bug fix (PEP 479 compatibility).
+
+This module tests that the mwxml library properly handles StopIteration
+exceptions in Python 3.7+ where PEP 479 converts StopIteration raised
+inside generators to RuntimeError.
+"""
+
+import io
+import pytest
+import sys
+
+from ..dump import Dump
+
+
+# Sample XML with valid MediaWiki structure
+MINIMAL_XML = """
+<mediawiki xmlns="http://www.mediawiki.org/xml/export-0.10/">
+  <siteinfo>
+    <sitename>Wikipedia</sitename>
+    <dbname>enwiki</dbname>
+  </siteinfo>
+  <page>
+    <title>Test Page</title>
+    <ns>0</ns>
+    <id>1</id>
+    <revision>
+      <id>100</id>
+      <timestamp>2021-01-01T00:00:00Z</timestamp>
+      <text>Test content</text>
+    </revision>
+  </page>
+</mediawiki>
+"""
+
+MULTI_PAGE_XML = """
+<mediawiki xmlns="http://www.mediawiki.org/xml/export-0.10/">
+  <siteinfo>
+    <sitename>Wikipedia</sitename>
+    <dbname>enwiki</dbname>
+  </siteinfo>
+  <page>
+    <title>Page 1</title>
+    <ns>0</ns>
+    <id>1</id>
+    <revision>
+      <id>100</id>
+      <timestamp>2021-01-01T00:00:00Z</timestamp>
+      <text>Content 1</text>
+    </revision>
+  </page>
+  <page>
+    <title>Page 2</title>
+    <ns>0</ns>
+    <id>2</id>
+    <revision>
+      <id>200</id>
+      <timestamp>2021-01-02T00:00:00Z</timestamp>
+      <text>Content 2</text>
+    </revision>
+  </page>
+</mediawiki>
+"""
+
+
+@pytest.mark.skipif(sys.version_info < (3, 7), 
+                    reason="Bug only affects Python 3.7+")
+def test_stopiteration_bug_reproduction():
+    """
+    Reproduce the StopIteration RuntimeError bug in Python 3.7+.
+
+    This test demonstrates the bug that occurs when the XML stream is
+    exhausted and StopIteration propagates through a generator, which
+    PEP 479 converts to RuntimeError.
+
+    NOTE: This test is expected to FAIL before the fix is applied and
+    PASS after the fix is applied.
+    """
+    dump = Dump.from_file(io.StringIO(MINIMAL_XML))
+
+    # Before the fix, this should raise RuntimeError in Python 3.7+
+    # After the fix, this should complete normally
+    try:
+        pages = list(dump)
+        # If we get here, the fix is working
+        assert len(pages) == 1
+        assert pages[0].title == "Test Page"
+    except RuntimeError as e:
+        if "generator raised StopIteration" in str(e):
+            pytest.fail(
+                "Bug reproduced: RuntimeError raised due to StopIteration in generator. "
+                "The fix has not been applied yet."
+            )
+        else:
+            # Some other RuntimeError, re-raise it
+            raise
+
+
+def test_iteration_completes_normally():
+    """
+    Verify iteration completes without RuntimeError after fix.
+
+    This test verifies that the fix properly handles stream exhaustion
+    and allows iteration to complete normally.
+    """
+    dump = Dump.from_file(io.StringIO(MINIMAL_XML))
+
+    # Should complete without raising RuntimeError
+    pages = list(dump)
+
+    # Verify the page was extracted
+    assert len(pages) == 1
+    assert pages[0].title == "Test Page"
+    assert pages[0].id == 1
+    assert pages[0].namespace == 0
+
+
+def test_multiple_pages_iteration():
+    """
+    Verify iteration works correctly with multiple pages.
+
+    Tests that the fix doesn't break normal iteration over multiple
+    pages in the XML dump.
+    """
+    dump = Dump.from_file(io.StringIO(MULTI_PAGE_XML))
+
+    # Should complete without raising RuntimeError
+    pages = list(dump)
+
+    # Verify all pages were extracted
+    assert len(pages) == 2
+    assert pages[0].title == "Page 1"
+    assert pages[0].id == 1
+    assert pages[1].title == "Page 2"
+    assert pages[1].id == 2
+
+
+def test_iteration_with_generator_pattern():
+    """
+    Verify the fix works with generator iteration pattern.
+
+    This tests that the fix works when using the dump as a generator
+    rather than converting to a list.
+    """
+    dump = Dump.from_file(io.StringIO(MULTI_PAGE_XML))
+
+    # Iterate using generator pattern
+    page_titles = []
+    for page in dump:
+        page_titles.append(page.title)
+
+    # Should have collected all pages
+    assert page_titles == ["Page 1", "Page 2"]
+
+
+def test_empty_dump_iteration():
+    """
+    Verify iteration handles empty dumps correctly.
+
+    Tests edge case where there are no pages in the dump.
+    """
+    empty_xml = """
+    <mediawiki xmlns="http://www.mediawiki.org/xml/export-0.10/">
+      <siteinfo>
+        <sitename>Wikipedia</sitename>
+        <dbname>enwiki</dbname>
+      </siteinfo>
+    </mediawiki>
+    """
+
+    dump = Dump.from_file(io.StringIO(empty_xml))
+
+    # Should complete without error
+    pages = list(dump)
+
+    # Should have no pages
+    assert len(pages) == 0
+
+
+def test_partial_iteration():
+    """
+    Verify partial iteration doesn't cause issues.
+
+    Tests that breaking out of iteration early doesn't cause problems
+    with the StopIteration handling.
+    """
+    dump = Dump.from_file(io.StringIO(MULTI_PAGE_XML))
+
+    # Only iterate over first page
+    first_page = None
+    for page in dump:
+        first_page = page
+        break
+
+    # Should have gotten the first page
+    assert first_page is not None
+    assert first_page.title == "Page 1"
+
+
+if __name__ == "__main__":
+    # Allow running tests directly
+    pytest.main([__file__, "-v"])