[nl] extract linkage sections

src/wiktextract/extractor/nl/linkage.py

@@ -0,0 +1,69 @@
+import re
+
+from wikitextprocessor import LevelNode, NodeKind, TemplateNode, WikiNode
+
+from ...page import clean_node
+from ...wxr_context import WiktextractContext
+from .models import Linkage, WordEntry
+
+
+def extract_linkage_section(
+    wxr: WiktextractContext,
+    word_entry: WordEntry,
+    level_node: LevelNode,
+    linkage_type: str,
+) -> None:
+    sense_index = 0
+    sense = ""
+    raw_tags = []
+    for node in level_node.children:
+        if isinstance(node, TemplateNode):
+            if node.template_name == "intens":
+                # https://nl.wiktionary.org/wiki/Sjabloon:intens
+                raw_tags = ["intensivering"]
+                s_index_str = node.template_parameters.get(2, "").strip()
+                if re.fullmatch(r"\d+", s_index_str):
+                    sense_index = int(s_index_str)
+        elif isinstance(node, WikiNode):
+            if node.kind == NodeKind.LINK:
+                word = clean_node(wxr, None, node)
+                if word != "":
+                    getattr(word_entry, linkage_type).append(
+                        Linkage(
+                            word=word,
+                            sense=sense,
+                            sense_index=sense_index,
+                            raw_tags=raw_tags,
+                        )
+                    )
+            elif node.kind == NodeKind.LIST:
+                for list_item in node.find_child(NodeKind.LIST_ITEM):
+                    extract_linkage_list_item(
+                        wxr,
+                        word_entry,
+                        list_item,
+                        linkage_type,
+                        sense,
+                        sense_index,
+                    )
+
+
+def extract_linkage_list_item(
+    wxr: WiktextractContext,
+    word_entry: WordEntry,
+    list_item: WordEntry,
+    linkage_type: str,
+    sense: str,
+    sense_index: str,
+) -> None:
+    for node in list_item.children:
+        if isinstance(node, str):
+            m = re.search(r"\[(\d+)\]", node)
+            if m is not None:
+                sense_index = int(m.group(1))
+        elif isinstance(node, WikiNode) and node.kind == NodeKind.LINK:
+            word = clean_node(wxr, None, node)
+            if word != "":
+                getattr(word_entry, linkage_type).append(
+                    Linkage(word=word, sense=sense, sense_index=sense_index)
+                )

src/wiktextract/extractor/nl/models.py

@@ -37,6 +37,17 @@ class Sound(DutchBaseModel):
     raw_tags: list[str] = []
 
 
+class Linkage(DutchBaseModel):
+    word: str
+    tags: list[str] = []
+    raw_tags: list[str] = []
+    roman: str = ""
+    sense: str = Field(default="", description="Definition of the word")
+    sense_index: int = Field(
+        default=0, ge=0, description="Number of the definition, start from 1"
+    )
+
+
 class WordEntry(DutchBaseModel):
     model_config = ConfigDict(title="Dutch Wiktionary")
     word: str = Field(description="Word string", min_length=1)
@@ -48,5 +59,18 @@ class WordEntry(DutchBaseModel):
     categories: list[str] = []
     tags: list[str] = []
     raw_tags: list[str] = []
-    sense_index: str = ""
+    etymology_index: str = Field(default="", exclude=True)
     sounds: list[Sound] = []
+    anagrams: list[Linkage] = []
+    antonyms: list[Linkage] = []
+    derived: list[Linkage] = []
+    proverbs: list[Linkage] = []
+    holonyms: list[Linkage] = []
+    homophones: list[Linkage] = []
+    hypernyms: list[Linkage] = []
+    hyponyms: list[Linkage] = []
+    metonyms: list[Linkage] = []
+    paronyms: list[Linkage] = []
+    related: list[Linkage] = []
+    rhymes: list[Linkage] = []
+    synonyms: list[Linkage] = []

src/wiktextract/extractor/nl/page.py

@@ -10,9 +10,10 @@
 
 from ...page import clean_node
 from ...wxr_context import WiktextractContext
+from .linkage import extract_linkage_section
 from .models import Sense, WordEntry
 from .pos import extract_pos_section
-from .section_titles import POS_DATA
+from .section_titles import LINKAGE_SECTIONS, POS_DATA
 from .sound import extract_sound_section
 
 
@@ -39,6 +40,13 @@ def parse_section(
         extract_sound_section(
             wxr, page_data[-1] if len(page_data) > 0 else base_data, level_node
         )
+    elif title_text in LINKAGE_SECTIONS:
+        extract_linkage_section(
+            wxr,
+            page_data[-1] if len(page_data) > 0 else base_data,
+            level_node,
+            LINKAGE_SECTIONS[title_text],
+        )
 
     for next_level in level_node.find_child(LEVEL_KIND_FLAGS):
         parse_section(wxr, page_data, base_data, next_level)

src/wiktextract/extractor/nl/pos.py

@@ -79,10 +79,10 @@ def extract_pos_header_line_nodes(
     wxr: WiktextractContext, word_entry: WordEntry, nodes: list[WikiNode | str]
 ) -> None:
     for node in nodes:
-        if isinstance(node, str) and word_entry.sense_index == "":
+        if isinstance(node, str) and word_entry.etymology_index == "":
             m = re.search(r"\[(.+)\]", node.strip())
             if m is not None:
-                word_entry.sense_index = m.group(1).strip()
+                word_entry.etymology_index = m.group(1).strip()
         elif isinstance(node, TemplateNode) and node.template_name == "-l-":
             extract_l_template(wxr, word_entry, node)
 

src/wiktextract/extractor/nl/section_titles.py

@@ -47,3 +47,23 @@
     "Symbool": {"pos": "symbol"},
     "Werkwoord": {"pos": "verb"},
 }
+
+
+LINKAGE_SECTIONS = {
+    "Anagrammen": "anagrams",
+    "Antoniemen": "antonyms",
+    "Afgeleide begrippen": "derived",
+    "Uitdrukkingen en gezegden": "proverbs",
+    "Holoniemen": "holonyms",
+    "Gelijkklinkende woorden": "homophones",
+    "Hyperoniemen": "hypernyms",
+    "Hyponiemen": "hyponyms",
+    "Hyponiemen (in taxonomische zin)": "hyponyms",
+    "Meroniemen": "metonyms",
+    "Paroniemen": "paronyms",
+    "Spreekwoorden": "proverbs",
+    "Verwante begrippen": "related",
+    "Rijmwoorden": "rhymes",
+    "Synoniemen": "synonyms",
+    "Typische woordcombinaties": "derived",
+}

tests/test_nl_gloss.py

@@ -79,7 +79,6 @@ def test_simple_case(self):
                             "raw_tags": ["roofdieren"],
                         }
                     ],
-                    "sense_index": "A",
                     "tags": ["masculine"],
                     "word": "hond",
                 }

tests/test_nl_linkage.py

@@ -0,0 +1,57 @@
+from unittest import TestCase
+
+from wikitextprocessor import Wtp
+
+from wiktextract.config import WiktionaryConfig
+from wiktextract.extractor.nl.page import parse_page
+from wiktextract.wxr_context import WiktextractContext
+
+
+class TestNlLinkage(TestCase):
+    maxDiff = None
+
+    def setUp(self) -> None:
+        self.wxr = WiktextractContext(
+            Wtp(lang_code="nl"),
+            WiktionaryConfig(
+                dump_file_lang_code="nl",
+                capture_language_codes=None,
+            ),
+        )
+
+    def tearDown(self) -> None:
+        self.wxr.wtp.close_db_conn()
+
+    def test_intens_template(self):
+        data = parse_page(
+            self.wxr,
+            "hond",
+            """==Nederlands==
+====Zelfstandig naamwoord====
+# zoogdier
+=====Hyponiemen=====
+{{intens|nld|1}} [[superhond]]
+{{intens|nld|2}} [[kankerhond]], [[tyfushond]]
+*[1] [[reu]]""",
+        )
+        self.assertEqual(
+            data[0]["hyponyms"],
+            [
+                {
+                    "word": "superhond",
+                    "sense_index": 1,
+                    "raw_tags": ["intensivering"],
+                },
+                {
+                    "word": "kankerhond",
+                    "sense_index": 2,
+                    "raw_tags": ["intensivering"],
+                },
+                {
+                    "word": "tyfushond",
+                    "sense_index": 2,
+                    "raw_tags": ["intensivering"],
+                },
+                {"word": "reu", "sense_index": 1},
+            ],
+        )