Fix index out of range error in fr/pronunciation.py

tatuylonen · Nov 29, 2023 · b8ff771 · b8ff771
1 parent 201c4d5
commit b8ff771
Show file tree

Hide file tree

Showing 3 changed files with 24 additions and 11 deletions.
diff --git a/src/wiktextract/extractor/fr/page.py b/src/wiktextract/extractor/fr/page.py
@@ -72,7 +72,7 @@ def parse_section(
                 wxr.config.capture_pronunciation
                 and section_type in wxr.config.OTHER_SUBTITLES["pronunciation"]
             ):
-                extract_pronunciation(wxr, page_data, level_node)
+                extract_pronunciation(wxr, page_data, level_node, base_data)
             elif (
                 wxr.config.capture_linkages
                 and section_type in wxr.config.LINKAGE_SUBTITLES

diff --git a/src/wiktextract/extractor/fr/pronunciation.py b/src/wiktextract/extractor/fr/pronunciation.py
@@ -10,19 +10,25 @@
 
 
 def extract_pronunciation(
-    wxr: WiktextractContext, page_data: List[Dict], level_node: WikiNode
+    wxr: WiktextractContext,
+    page_data: List[Dict],
+    level_node: WikiNode,
+    base_data: Dict[str, str],
 ) -> None:
     sound_data = []
+    lang_code = base_data.get("lang_code")
     for list_node in level_node.find_child(NodeKind.LIST):
         for list_item_node in list_node.find_child(NodeKind.LIST_ITEM):
             sound_data.extend(
                 process_pron_list_item(
-                    wxr, list_item_node, page_data, defaultdict(list)
+                    wxr, list_item_node, defaultdict(list), lang_code
                 )
             )
 
     if len(sound_data) == 0:
         return
+    if len(page_data) == 0:
+        page_data.append(deepcopy(base_data))
 
     if level_node.kind == NodeKind.LEVEL3:
         # Add extracted sound data to all sense dictionaries that have the same
@@ -53,10 +59,10 @@ def extract_pronunciation(
 def process_pron_list_item(
     wxr: WiktextractContext,
     list_item_node: WikiNode,
-    page_data: List[Dict],
     sound_data: Dict[str, Union[str, List[str]]],
+    lang_code: str,
 ) -> List[Dict[str, Union[str, List[str]]]]:
-    pron_key = "zh-pron" if page_data[-1].get("lang_code") == "zh" else "ipa"
+    pron_key = "zh-pron" if lang_code == "zh" else "ipa"
 
     for template_node in list_item_node.find_child(NodeKind.TEMPLATE):
         if template_node.template_name in PRON_TEMPLATES:
@@ -81,7 +87,7 @@ def process_pron_list_item(
         ):
             new_sound_data = deepcopy(sound_data)
             process_pron_list_item(
-                wxr, nest_list_item, page_data, new_sound_data
+                wxr, nest_list_item, new_sound_data, lang_code
             )
             if pron_key in new_sound_data:
                 returned_data.append(new_sound_data)

diff --git a/tests/test_fr_pronunciation.py b/tests/test_fr_pronunciation.py
@@ -27,7 +27,7 @@ def test_pron_list(self):
         root = self.wxr.wtp.parse(
             "=== Prononciation ===\n* {{pron|bɔ̃.ʒuʁ|fr}}\n** {{écouter|France (Paris)|bõ.ʒuːʁ|audio=Fr-bonjour.ogg|lang=fr}}"
         )
-        extract_pronunciation(self.wxr, page_data, root.children[0])
+        extract_pronunciation(self.wxr, page_data, root.children[0], {})
         self.assertEqual(
             page_data,
             [
@@ -60,13 +60,18 @@ def test_pron_list(self):
         )
 
     def test_str_pron(self):
-        page_data = [defaultdict(list, {"lang_code": "zh"})]
+        page_data = []
         self.wxr.wtp.add_page("Modèle:Yale-zh", 10, body="Yale")
         self.wxr.wtp.start_page("")
         root = self.wxr.wtp.parse(
             "=== {{S|prononciation}} ===\n* '''cantonais''' {{pron||yue}}\n** {{Yale-zh}} : nei⁵hou²"
         )
-        extract_pronunciation(self.wxr, page_data, root.children[0])
+        extract_pronunciation(
+            self.wxr,
+            page_data,
+            root.children[0],
+            defaultdict(list, {"lang_code": "zh"}),
+        )
         self.assertEqual(
             page_data[0].get("sounds"),
             [{"tags": ["cantonais", "Yale"], "zh-pron": "nei⁵hou²"}],
@@ -78,14 +83,16 @@ def test_no_ipa(self):
         files.
         Test wikitext from https://fr.wiktionary.org/wiki/mars
         """
-        page_data = [defaultdict(list)]
+        page_data = []
         self.wxr.wtp.start_page("")
         root = self.wxr.wtp.parse(
             """=== {{S|prononciation}} ===
 {{ébauche-pron|sv}}
 * {{écouter|lang=sv|Suède||audio=LL-Q9027 (swe)-Moonhouse-mars.wav}}"""
         )
-        extract_pronunciation(self.wxr, page_data, root.children[0])
+        extract_pronunciation(
+            self.wxr, page_data, root.children[0], defaultdict(list)
+        )
         self.assertEqual(
             page_data[0].get("sounds"),
             [