fix: Preserve XML tree order in Formex parser

tests/test_parser.py

@@ -0,0 +1,52 @@
+import pytest
+from lxml import etree as ET
+
+from formex_viewer.formex4 import FormexArticleConverter
+from formex_viewer.main import Language
+
+
+@pytest.fixture
+def converter():
+    return FormexArticleConverter(language=Language.ENG)
+
+
+def test_convert_tree_order(converter):
+    """Test that the order of HTML blocks in the converted article matches the order of elements in the XML tree."""
+
+    xml = """
+    <ARTICLE>
+        <SUBDIV>
+            <TITLE>
+                <TI>Subdivision Title</TI>
+                <STI>Subdivision Subtitle</STI>
+            </TITLE>
+            <PARAG IDENTIFIER="001.001">
+                <NO.PARAG>1</NO.PARAG>
+                <ALINEA>Paragraph 1 text.</ALINEA>
+            </PARAG>
+            <COMMENT>Comment text.</COMMENT>
+            <ALINEA>Alinea text.</ALINEA>
+            <QUOT.S>Quotation text.</QUOT.S>
+            <SUBDIV>
+                <TITLE>
+                    <TI>Nested Subdivision</TI>
+                </TITLE>
+                <ALINEA>Nested alinea.</ALINEA>
+            </SUBDIV>
+        </SUBDIV>
+    </ARTICLE>
+    """
+    parser = ET.XMLParser(remove_blank_text=True)
+    el = ET.fromstring(xml, parser)
+    html = converter.convert_article(el)
+
+    # Check that the order of HTML blocks matches the order of elements in the XML tree
+    idx_title = html.index("Subdivision Title")
+    idx_parag = html.index('class="paragraph"')
+    idx_comment = html.index("Comment text.")
+    idx_alinea = html.index("Alinea text.")
+    idx_quot = html.index("Quotation text.")
+    idx_nested = html.index("Nested Subdivision")
+
+    # The order in the XML: title, parag, alinea, comment, quot, nested subdiv
+    assert idx_title < idx_parag < idx_comment < idx_alinea < idx_quot < idx_nested