Extract span node's title attribute from expanded qualifier template

tatuylonen · Jan 30, 2024 · ef6ac38 · ef6ac38
1 parent 19f65b4
commit ef6ac38
Show file tree

Hide file tree

Showing 2 changed files with 33 additions and 4 deletions.
diff --git a/src/wiktextract/extractor/zh/translation.py b/src/wiktextract/extractor/zh/translation.py
@@ -121,9 +121,19 @@ def process_translation_list_item(
                 continue
             else:
                 # qualifier template
-                tag = clean_node(wxr, None, child)
-                if len(tag) > 0:
-                    tr_data.tags.append(tag.strip("()"))
+                expanded_template = wxr.wtp.parse(
+                    wxr.wtp.node_to_wikitext(child), expand_all=True
+                )
+                find_title = False
+                for span_node in expanded_template.find_html("span"):
+                    tag = span_node.attrs.get("title", "")
+                    if len(tag) > 0:
+                        tr_data.tags.append(tag.strip())
+                        find_title = True
+                if not find_title:
+                    tag = clean_node(wxr, None, child)
+                    if len(tag) > 0:
+                        tr_data.tags.append(tag.strip("()"))
         elif isinstance(child, WikiNode) and child.kind == NodeKind.LINK:
             if len(tr_data.word) > 0:
                 page_data[-1].translations.append(tr_data.model_copy(deep=True))

diff --git a/tests/test_zh_translation.py b/tests/test_zh_translation.py
@@ -180,8 +180,21 @@ def test_language_name_template(self):
 
     def test_l_template(self):
         self.wxr.wtp.start_page("茄子")
+        self.wxr.wtp.add_page("Template:cs", 10, "捷克语")
+        self.wxr.wtp.add_page(
+            "Template:l",
+            10,
+            """<span>{{{2}}}</span>
+{{#if:{{{g|}}}|<span class="gender"><abbr title="陽性名詞">m</abbr></span>}}""",
+        )
+        self.wxr.wtp.add_page(
+            "Template:口", 10, '<span title="口语词汇">口</span>〉'
+        )
         page_data = [WordEntry(word="茄子", lang_code="zh", lang="漢語")]
-        node = self.wxr.wtp.parse("* 南非語: {{l|af|eiervrug}}")
+        node = self.wxr.wtp.parse(
+            """* 南非語: {{l|af|eiervrug}}
+* {{cs}}: {{l|cs|patližán|g=m}} {{口}}"""
+        )
         extract_translation(self.wxr, page_data, node)
         self.assertEqual(
             [
@@ -194,5 +207,11 @@ def test_l_template(self):
                     "lang": "南非語",
                     "word": "eiervrug",
                 },
+                {
+                    "lang_code": "cs",
+                    "lang": "捷克语",
+                    "word": "patližán",
+                    "tags": ["陽性名詞", "口语词汇"],
+                },
             ],
         )