[th] extract "syn", "ant" linkage templates

src/wiktextract/extractor/th/example.py

@@ -5,16 +5,19 @@
 from ...page import clean_node
 from ...wxr_context import WiktextractContext
 from ..ruby import extract_ruby
-from .models import Example, Sense
+from .models import Example, Sense, WordEntry
 from .tags import translate_raw_tags
 
 
 def extract_example_list_item(
     wxr: WiktextractContext,
+    word_entry: WordEntry,
     sense: Sense,
     list_item: WikiNode,
     ref: str = "",
 ) -> None:
+    from .linkage import LINKAGE_TEMPLATES, extract_syn_template
+
     for node in list_item.children:
         if isinstance(node, TemplateNode):
             if node.template_name in ["ux", "usex", "ko-usex"]:
@@ -25,9 +28,15 @@ def extract_example_list_item(
                 extract_template_ja_usex(wxr, sense, node, ref)
             elif node.template_name.startswith("quote-"):
                 ref = extract_quote_template(wxr, sense, node)
+            elif node.template_name in LINKAGE_TEMPLATES:
+                extract_syn_template(
+                    wxr, word_entry, node, LINKAGE_TEMPLATES[node.template_name]
+                )
         elif isinstance(node, WikiNode) and node.kind == NodeKind.LIST:
             for child_list_item in node.find_child(NodeKind.LIST_ITEM):
-                extract_example_list_item(wxr, sense, child_list_item, ref)
+                extract_example_list_item(
+                    wxr, word_entry, sense, child_list_item, ref
+                )
 
 
 def extract_ux_template(

src/wiktextract/extractor/th/linkage.py

@@ -1,3 +1,5 @@
+from itertools import count
+
 from wikitextprocessor.parser import (
     LEVEL_KIND_FLAGS,
     LevelNode,
@@ -129,3 +131,34 @@ def extract_ws_template(
     if word != "":
         l_data = Linkage(word=word, source=source)
         getattr(word_entry, linkage_type).append(l_data)
+
+
+LINKAGE_TEMPLATES = {
+    "syn": "synonyms",
+    "synonyms": "synonyms",
+    "synsee": "synonyms",
+    "ant": "antonyms",
+    "antonyms": "antonyms",
+    "cot": "coordinate_terms",
+    "coordinate terms": "coordinate_terms",
+    "hyper": "hypernyms",
+    "hypernyms": "hypernyms",
+    "hypo": "hyponyms",
+    "hyponyms": "hyponyms",
+}
+
+
+def extract_syn_template(
+    wxr: WiktextractContext,
+    word_entry: WordEntry,
+    t_node: TemplateNode,
+    linkage_type: str,
+) -> None:
+    for arg_name in count(2):
+        if arg_name not in t_node.template_parameters:
+            break
+        arg_value = clean_node(wxr, None, t_node.template_parameters[arg_name])
+        if arg_value.startswith("อรรถาภิธาน:"):
+            extract_thesaurus_page(wxr, word_entry, linkage_type, arg_value)
+        elif arg_value != "":
+            getattr(word_entry, linkage_type).append(Linkage(word=arg_value))

src/wiktextract/extractor/th/models.py

@@ -102,3 +102,4 @@ class WordEntry(ThaiBaseModel):
     hyponyms: list[Linkage] = []
     hypernyms: list[Linkage] = []
     idioms: list[Linkage] = []
+    coordinate_terms: list[Linkage] = []

src/wiktextract/extractor/th/pos.py

@@ -66,7 +66,7 @@ def extract_gloss_list_item(
             (":", "*")
         ):
             for e_list_item in child_list.find_child(NodeKind.LIST_ITEM):
-                extract_example_list_item(wxr, sense, e_list_item)
+                extract_example_list_item(wxr, word_entry, sense, e_list_item)
 
     if gloss_str != "":
         sense.glosses.append(gloss_str)

tests/test_th_linkage.py

@@ -83,3 +83,17 @@ def test_theasurus_page(self):
             page_data[0]["synonyms"],
             [{"word": "ต่อมโลหิต", "source": "อรรถาภิธาน:ระดู"}],
         )
+
+    def test_syn_template(self):
+        page_data = parse_page(
+            self.wxr,
+            "โทรทัศน์",
+            """== ภาษาไทย ==
+=== คำนาม ===
+# กระบวนการถ่ายทอด
+#: {{syn|th|ทีวี|โทรภาพ}}""",
+        )
+        self.assertEqual(
+            page_data[0]["synonyms"],
+            [{"word": "ทีวี"}, {"word": "โทรภาพ"}],
+        )