[it] handle example list with more than one italic nodes layout

src/wiktextract/extractor/it/example.py

@@ -14,12 +14,15 @@ def extract_example_list_item(
     text_nodes = []
     roman = ""
     translation = ""
+    ref = ""
+    has_zh_tradsem = False
     for index, node in enumerate(list_item.children):
         if (
             isinstance(node, TemplateNode)
             and node.template_name == "zh-tradsem"
         ):
             examples.extend(extract_zh_tradsem(wxr, node))
+            has_zh_tradsem = True
         elif isinstance(node, WikiNode):
             match node.kind:
                 case NodeKind.ITALIC:
@@ -39,17 +42,38 @@ def extract_example_list_item(
                 case _ if lang_code in ["zh", "ja"]:
                     if before_italic:
                         text_nodes.append(node)
-        elif (
-            isinstance(node, str) and lang_code in ["zh", "ja"] and "-" in node
-        ):
+        elif isinstance(node, str) and "-" in node:
+            for t_node in list_item.find_child(NodeKind.TEMPLATE):
+                if t_node.template_name == "Term":
+                    ref = clean_node(wxr, None, t_node).strip("()")
+                    break
             translation = clean_node(
                 wxr,
                 sense,
                 wxr.wtp.node_to_wikitext(
                     [node[node.index("-") + 1 :]]
-                    + list_item.children[index + 1 :]
+                    + [
+                        n
+                        for n in list_item.children[index + 1 :]
+                        if not (
+                            isinstance(n, TemplateNode)
+                            and n.template_name == "Term"
+                        )
+                    ]
                 ),
             )
+            if not has_zh_tradsem and len(examples) > 1:
+                examples.clear()
+                examples.append(
+                    Example(
+                        text=clean_node(
+                            wxr,
+                            None,
+                            list_item.children[:index]
+                            + [node[: node.index("-")]],
+                        )
+                    )
+                )
             break
         elif lang_code in ["zh", "ja"] and len(examples) == 0 and before_italic:
             text_nodes.append(node)
@@ -69,11 +93,23 @@ def extract_example_list_item(
         )
         examples.append(example)
 
+    if not has_zh_tradsem and len(examples) > 1:
+        examples.clear()
+        examples.append(
+            Example(
+                text=clean_node(
+                    wxr, None, list(list_item.invert_find_child(NodeKind.LIST))
+                )
+            )
+        )
+
     for example in examples:
         if roman != "":
             example.roman = roman
         if translation != "":
             example.translation = translation
+        if ref != "":
+            example.ref = ref
         if example.text != "":
             sense.examples.append(example)
 

src/wiktextract/extractor/it/pos.py

@@ -91,12 +91,7 @@ def extract_gloss_list_item(
             else:
                 gloss_nodes.append(t_str)
         elif isinstance(node, WikiNode) and node.kind == NodeKind.LIST:
-            if node.sarg.endswith("*"):
-                for example_list_item in node.find_child(NodeKind.LIST_ITEM):
-                    extract_example_list_item(
-                        wxr, sense, example_list_item, word_entry.lang_code
-                    )
-            elif (
+            if (
                 node.sarg.endswith(":")
                 and len(sense.examples) > 0
                 and sense.examples[-1].translation == ""
@@ -105,6 +100,11 @@ def extract_gloss_list_item(
                     sense.examples[-1].translation = clean_node(
                         wxr, sense, tr_list_item.children
                     )
+            elif node.sarg.endswith(("*", ":")):
+                for example_list_item in node.find_child(NodeKind.LIST_ITEM):
+                    extract_example_list_item(
+                        wxr, sense, example_list_item, word_entry.lang_code
+                    )
         else:
             gloss_nodes.append(node)
     gloss_str = clean_node(wxr, sense, gloss_nodes)

tests/test_it_example.py

@@ -139,3 +139,79 @@ def test_zh_tradsem(self):
                 }
             ],
         )
+
+    def test_double_italic_nodes_with_translation(self):
+        self.wxr.wtp.add_page("Template:-en-", 10, "Inglese")
+        data = parse_page(
+            self.wxr,
+            "water",
+            """== {{-en-}} ==
+===Sostantivo===
+# acqua
+#: ''May I have a glass of '''water'''?'' - ''Posso avere un bicchiere d''''acqua'''''?""",
+        )
+        self.assertEqual(
+            data[0]["senses"],
+            [
+                {
+                    "glosses": ["acqua"],
+                    "examples": [
+                        {
+                            "text": "May I have a glass of water?",
+                            "translation": "Posso avere un bicchiere d'acqua?",
+                        }
+                    ],
+                }
+            ],
+        )
+
+    def test_double_italic_nodes_no_translation(self):
+        self.wxr.wtp.add_page("Template:-it-", 10, "Italiano")
+        data = parse_page(
+            self.wxr,
+            "essere",
+            """== {{-it-}} ==
+===Sostantivo===
+#chi [[esiste]]
+#* ''gli '''esseri''' viventi''; ''gli '''esseri''' animati''""",
+        )
+        self.assertEqual(
+            data[0]["senses"],
+            [
+                {
+                    "glosses": ["chi esiste"],
+                    "examples": [
+                        {"text": "gli esseri viventi; gli esseri animati"}
+                    ],
+                }
+            ],
+        )
+
+    def test_term_ref_template(self):
+        self.wxr.wtp.add_page("Template:-la-", 10, "Latino")
+        self.wxr.wtp.add_page("Template:Term", 10, "({{{1}}})")
+        data = parse_page(
+            self.wxr,
+            "libero",
+            """== {{-la-}} ==
+===Verbo===
+# [[assolvere]], [[liberare]] dalle [[accuse]], [[giudicare]] [[innocente]]
+#* ''et eum omni [[ignominia]] '''liberat''''' - e lo [[assolve]] da ogni [[ignominia]] {{Term|[[:w:Marco Tullio Cicerone|Cicerone]], [[:w:Pro Cluentio|Pro Cluentio]], [[:s:la:Pro_Aulo_Cluentio_Habito|XLVII, 132]]}}""",
+        )
+        self.assertEqual(
+            data[0]["senses"],
+            [
+                {
+                    "glosses": [
+                        "assolvere, liberare dalle accuse, giudicare innocente"
+                    ],
+                    "examples": [
+                        {
+                            "text": "et eum omni ignominia liberat",
+                            "translation": "e lo assolve da ogni ignominia",
+                            "ref": "Cicerone, Pro Cluentio, XLVII, 132",
+                        }
+                    ],
+                }
+            ],
+        )