[th] extract form-of templates

tatuylonen · Jan 17, 2025 · 42b3fdb · 42b3fdb
1 parent d4b786b
commit 42b3fdb
Show file tree

Hide file tree

Showing 4 changed files with 68 additions and 3 deletions.
diff --git a/src/wiktextract/extractor/th/models.py b/src/wiktextract/extractor/th/models.py
@@ -28,13 +28,19 @@ class Example(ThaiBaseModel):
     raw_tags: list[str] = []
 
 
+class AltForm(ThaiBaseModel):
+    word: str
+    roman: str = ""
+
+
 class Sense(ThaiBaseModel):
     glosses: list[str] = []
     tags: list[str] = []
     raw_tags: list[str] = []
     categories: list[str] = []
     examples: list[Example] = []
     classifiers: list[str] = []
+    form_of: list[AltForm] = []
 
 
 class Form(ThaiBaseModel):

diff --git a/src/wiktextract/extractor/th/pos.py b/src/wiktextract/extractor/th/pos.py
@@ -1,11 +1,16 @@
 import itertools
 
-from wikitextprocessor import LevelNode, NodeKind, TemplateNode, WikiNode
+from wikitextprocessor import (
+    LevelNode,
+    NodeKind,
+    TemplateNode,
+    WikiNode,
+)
 
 from ...page import clean_node
 from ...wxr_context import WiktextractContext
 from .example import extract_example_list_item
-from .models import Form, Sense, WordEntry
+from .models import AltForm, Form, Sense, WordEntry
 from .section_titles import POS_DATA
 from .tags import translate_raw_tags
 
@@ -57,6 +62,15 @@ def extract_gloss_list_item(
             extract_label_template(wxr, sense, node)
         elif isinstance(node, TemplateNode) and node.template_name == "cls":
             extract_cls_template(wxr, sense, node)
+        elif isinstance(node, TemplateNode) and (
+            node.template_name.endswith(" of")
+            or node.template_name == "altform"
+        ):
+            expanded_node = wxr.wtp.parse(
+                wxr.wtp.node_to_wikitext(node), expand_all=True
+            )
+            extract_form_of_template(wxr, sense, expanded_node)
+            gloss_nodes.append(expanded_node)
         elif not (isinstance(node, WikiNode) and node.kind == NodeKind.LIST):
             gloss_nodes.append(node)
 
@@ -167,3 +181,22 @@ def extract_note_section(
             )
             if note_str != "":
                 word_entry.notes.append(note_str)
+
+
+def extract_form_of_template(
+    wxr: WiktextractContext,
+    sense: Sense,
+    expanded_node: WikiNode,
+) -> None:
+    form = AltForm(word="")
+    for i_tag in expanded_node.find_html_recursively("i"):
+        form.word = clean_node(wxr, None, i_tag)
+        break
+    for span_tag in expanded_node.find_html_recursively("span"):
+        if "mention-tr" in span_tag.attrs.get("class", ""):
+            form.roman = clean_node(wxr, None, span_tag)
+            break
+    if form.word != "":
+        sense.form_of.append(form)
+        if "form-of" not in sense.tags:
+            sense.tags.append("form-of")
diff --git a/tests/test_th_example.py b/tests/test_th_example.py
@@ -187,7 +187,7 @@ def test_ja_x(self):
         self.wxr.wtp.add_page(
             "แม่แบบ:syn of",
             10,
-            "คำพ้องความของ 北極熊 (ฮกเกียวกุงุมะ, “หมีขั้วโลก”)",
+            """<span class='form-of-definition use-with-mention'>คำพ้องความของ <span class='form-of-definition-link'><i class="Jpan mention" lang="ja">[[北極熊#ภาษาญี่ปุ่น|北極熊]]</i> <span class="mention-gloss-paren annotation-paren">(</span><span class="mention-tr tr">ฮกเกียวกุงุมะ</span>, <span class="mention-gloss-double-quote">“</span><span class="mention-gloss">หมีขั้วโลก</span><span class="mention-gloss-double-quote">”</span><span class="mention-gloss-paren annotation-paren">)</span></span></span>""",
         )
         page_data = parse_page(
             self.wxr,
@@ -236,5 +236,7 @@ def test_ja_x(self):
                         "ref": "1990 มิถุนายน 15, Takahashi, Rumiko, “:แม่แบบ:jaru [PART.5 Snatching the Scroll of Secrets]”, in :แม่แบบ:wj [Ranma ½], volume 11 (fiction), Tokyo: Shogakukan, →ISBN, page 72:",
                     }
                 ],
+                "form_of": [{"word": "北極熊", "roman": "ฮกเกียวกุงุมะ"}],
+                "tags": ["form-of"],
             },
         )
diff --git a/tests/test_th_gloss.py b/tests/test_th_gloss.py
@@ -234,3 +234,27 @@ def test_lo_alt(self):
                 {"form": "ທຸຣຽນ", "raw_tags": ["ล้าสมัย"], "roman": "ทุรย̂น"},
             ],
         )
+
+    def test_alt_form_template(self):
+        self.wxr.wtp.add_page(
+            "แม่แบบ:altform",
+            10,
+            """<span class='form-of-definition use-with-mention'>อีกรูปหนึ่งของ <span class='form-of-definition-link'><i class="Lana mention" lang="nod">[[ᨸᩣ᩠ᨠ#ภาษาคำเมือง|ᨸᩣ᩠ᨠ]]</i> <span class="mention-gloss-paren annotation-paren">(</span><span lang="nod-Latn" class="mention-tr tr Latn">ปาก</span><span class="mention-gloss-paren annotation-paren">)</span></span></span>""",
+        )
+        page_data = parse_page(
+            self.wxr,
+            "ปาก",
+            """== ภาษาคำเมือง ==
+=== คำนาม ===
+{{nod-noun}}
+
+# {{altform|nod|ᨸᩣ᩠ᨠ}}""",
+        )
+        self.assertEqual(
+            page_data[0]["senses"][0],
+            {
+                "glosses": ["อีกรูปหนึ่งของ ᨸᩣ᩠ᨠ (ปาก)"],
+                "form_of": [{"word": "ᨸᩣ᩠ᨠ", "roman": "ปาก"}],
+                "tags": ["form-of"],
+            },
+        )