Merge pull request #887 from xxyzz/nl

[nl] improve extract inflection and linakge data code
tatuylonen · Oct 25, 2024 · 03e0faf · 03e0faf
2 parents 7f16b89 + 04fb3f7
commit 03e0faf
Show file tree

Hide file tree

Showing 4 changed files with 50 additions and 10 deletions.
diff --git a/src/wiktextract/extractor/nl/inflection.py b/src/wiktextract/extractor/nl/inflection.py
@@ -40,15 +40,19 @@ def extract_noun_adj_table(
                 if col_index == 0:
                     row_header = clean_node(wxr, None, data_node)
                 else:
-                    form_str = clean_node(wxr, None, data_node)
-                    if form_str not in ["", "-", wxr.wtp.title]:
-                        form = Form(form=form_str)
-                        if row_header not in ["", "naamwoord"]:
-                            form.raw_tags.append(row_header)
-                        if col_index - 1 < len(column_headers):
-                            form.raw_tags.append(column_headers[col_index - 1])
-                        translate_raw_tags(form)
-                        word_entry.forms.append(form)
+                    for form_str in clean_node(
+                        wxr, None, data_node
+                    ).splitlines():
+                        if form_str not in ["", "-", wxr.wtp.title]:
+                            form = Form(form=form_str)
+                            if row_header not in ["", "naamwoord"]:
+                                form.raw_tags.append(row_header)
+                            if col_index - 1 < len(column_headers):
+                                form.raw_tags.append(
+                                    column_headers[col_index - 1]
+                                )
+                            translate_raw_tags(form)
+                            word_entry.forms.append(form)
 
     for link_node in expanded_node.find_child(NodeKind.LINK):
         clean_node(wxr, word_entry, link_node)

diff --git a/src/wiktextract/extractor/nl/linkage.py b/src/wiktextract/extractor/nl/linkage.py
@@ -89,6 +89,8 @@ def extract_linkage_list_item(
                 getattr(word_entry, linkage_type).append(
                     Linkage(word=word, sense=sense, sense_index=sense_index)
                 )
+        elif isinstance(node, TemplateNode) and node.template_name == "expr":
+            extract_expr_template(wxr, word_entry, node, linkage_type)
 
 
 def extract_nld_template(

diff --git a/src/wiktextract/extractor/nl/tags.py b/src/wiktextract/extractor/nl/tags.py
@@ -158,7 +158,7 @@
     # "familie": "family",
     "farmacologie": "pharmacology",
     # "feest": "party",
-    # "fietsen": "cycle",
+    "fietsen": "cycling",
     "filatelie": "philately",
     "filmkunst": "cinematography",
     "filosofie": "philosophy",

diff --git a/tests/test_nl_inflection.py b/tests/test_nl_inflection.py
@@ -196,3 +196,37 @@ def test_nlstam(self):
                 },
             ],
         )
+
+    def test_nlnoun_lines(self):
+        self.wxr.wtp.add_page(
+            "Sjabloon:-nlnoun-",
+            10,
+            """{| class="infobox"
+|-
+!
+! [[enkelvoud]]
+! [[meervoud]]
+|-
+| class="infoboxrijhoofding" | [[zelfstandig naamwoord|naamwoord]]
+| corpus
+| [[corpora]]<br>[[corpussen]]
+|}""",
+        )
+        data = parse_page(
+            self.wxr,
+            "corpus",
+            """==Nederlands==
+=====Woordherkomst en -opbouw=====
+*Leenwoord
+{{-nlnoun-|{{pn}}|[[corpora]]<br>[[{{pn}}sen]]|[[corpusje]]|[[corpusjes]]}}
+====Zelfstandig naamwoord====
+{{-l-|n}}
+# alle verzamelde""",
+        )
+        self.assertEqual(
+            data[0]["forms"],
+            [
+                {"form": "corpora", "tags": ["plural"]},
+                {"form": "corpussen", "tags": ["plural"]},
+            ],
+        )