tolerate duplicate ids in xslt xml input (#396)

Author: de-code

sciencebeam/transformers/xslt.py

@@ -13,7 +13,8 @@
 
 def _to_xslt_input(value: Union[bytes, str, T_XSLT_Input]) -> T_XSLT_Input:
     if isinstance(value, (bytes, str)):
-        return etree.fromstring(value)
+        parser = etree.XMLParser(recover=True)
+        return etree.fromstring(value, parser=parser)
     return value
 
 

tests/transformers/xslt_test.py

@@ -0,0 +1,18 @@
+from lxml import etree
+
+from sciencebeam.transformers.xslt import _to_xslt_input
+
+
+class TestToXsltInput:
+    def test_should_tolerate_duplicate_ids(self):
+        result: etree.ElementBase = _to_xslt_input(
+            '''
+            <xml>
+            <item xml:id="id1">item 1</item>
+            <item xml:id="id1">item 2</item>
+            </xml>
+            '''
+        )
+        items = result.findall('item')
+        assert len(items) == 2
+        assert [item.text for item in items] == ['item 1', 'item 2']