From 7a65b8c51495cccee4bfe876e5d5ce4ff4e7dc9e Mon Sep 17 00:00:00 2001
From: xxyzz <gitpull@protonmail.com>
Date: Wed, 11 Dec 2024 16:40:54 +0800
Subject: [PATCH] [it] extract example lists

---
 src/wiktextract/extractor/it/example.py | 24 +++++++++++++
 src/wiktextract/extractor/it/models.py  |  7 ++++
 src/wiktextract/extractor/it/pos.py     |  7 +++-
 tests/test_it_example.py                | 45 +++++++++++++++++++++++++
 4 files changed, 82 insertions(+), 1 deletion(-)
 create mode 100644 src/wiktextract/extractor/it/example.py
 create mode 100644 tests/test_it_example.py

diff --git a/src/wiktextract/extractor/it/example.py b/src/wiktextract/extractor/it/example.py
new file mode 100644
index 00000000..6117b854
--- /dev/null
+++ b/src/wiktextract/extractor/it/example.py
@@ -0,0 +1,24 @@
+from wikitextprocessor import NodeKind, WikiNode
+
+from ...page import clean_node
+from ...wxr_context import WiktextractContext
+from .models import Example, Sense
+
+
+def extract_example_list_item(
+    wxr: WiktextractContext, sense: Sense, list_item: WikiNode
+) -> None:
+    example = Example()
+    for node in list_item.children:
+        if isinstance(node, WikiNode):
+            match node.kind:
+                case NodeKind.ITALIC:
+                    example.text = clean_node(wxr, sense, node)
+                case NodeKind.LIST:
+                    for tr_list_item in node.find_child(NodeKind.LIST_ITEM):
+                        example.translation = clean_node(
+                            wxr, sense, tr_list_item.children
+                        )
+
+    if example.text != "":
+        sense.examples.append(example)
diff --git a/src/wiktextract/extractor/it/models.py b/src/wiktextract/extractor/it/models.py
index e0bfacaf..113da01c 100644
--- a/src/wiktextract/extractor/it/models.py
+++ b/src/wiktextract/extractor/it/models.py
@@ -10,11 +10,18 @@ class ItalianBaseModel(BaseModel):
     )
 
 
+class Example(ItalianBaseModel):
+    text: str = ""
+    translation: str = ""
+    ref: str = ""
+
+
 class Sense(ItalianBaseModel):
     glosses: list[str] = []
     tags: list[str] = []
     raw_tags: list[str] = []
     categories: list[str] = []
+    examples: list[Example] = []
 
 
 class WordEntry(ItalianBaseModel):
diff --git a/src/wiktextract/extractor/it/pos.py b/src/wiktextract/extractor/it/pos.py
index c77e427f..590cbd56 100644
--- a/src/wiktextract/extractor/it/pos.py
+++ b/src/wiktextract/extractor/it/pos.py
@@ -2,6 +2,7 @@
 
 from ...page import clean_node
 from ...wxr_context import WiktextractContext
+from .example import extract_example_list_item
 from .models import Sense, WordEntry
 from .section_titles import POS_DATA
 
@@ -43,7 +44,11 @@ def extract_gloss_list_item(
                         sense.raw_tags.append(raw_tag)
                 case _:
                     gloss_nodes.append(node)
-        elif not (isinstance(node, WikiNode) and node.kind == NodeKind.LIST):
+        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)
+        else:
             gloss_nodes.append(node)
     gloss_str = clean_node(wxr, sense, gloss_nodes)
     if gloss_str != "":
diff --git a/tests/test_it_example.py b/tests/test_it_example.py
new file mode 100644
index 00000000..ae66a81e
--- /dev/null
+++ b/tests/test_it_example.py
@@ -0,0 +1,45 @@
+from unittest import TestCase
+
+from wikitextprocessor import Wtp
+
+from wiktextract.config import WiktionaryConfig
+from wiktextract.extractor.it.page import parse_page
+from wiktextract.wxr_context import WiktextractContext
+
+
+class TestItExample(TestCase):
+    maxDiff = None
+
+    def setUp(self) -> None:
+        self.wxr = WiktextractContext(
+            Wtp(lang_code="it"),
+            WiktionaryConfig(
+                dump_file_lang_code="it", capture_language_codes=None
+            ),
+        )
+
+    def test_list_example(self):
+        self.wxr.wtp.add_page("Template:-br-", 10, "Bretone")
+        data = parse_page(
+            self.wxr,
+            "dog",
+            """== {{-br-}} ==
+===Sostantivo===
+# mutazione
+#* ''Da '''dog''', e '''dog'''.''
+#*: Il tuo cappello, il suo cappello.""",
+        )
+        self.assertEqual(
+            data[0]["senses"],
+            [
+                {
+                    "glosses": ["mutazione"],
+                    "examples": [
+                        {
+                            "text": "Da dog, e dog.",
+                            "translation": "Il tuo cappello, il suo cappello.",
+                        }
+                    ],
+                }
+            ],
+        )