[de] extract article data to Form.article field (#1545)

only do this for the "Deutsch Substantiv Übersicht" table template

Author: xxyzz

src/wiktextract/extractor/de/inflection.py

@@ -54,11 +54,11 @@ class RowspanHeader:
 
 
 def process_verb_table(
-    wxr: WiktextractContext, word_entry: WordEntry, template_node: TemplateNode
+    wxr: WiktextractContext, word_entry: WordEntry, t_node: TemplateNode
 ) -> None:
     # Vorlage:Deutsch Verb Übersicht
     expanded_template = wxr.wtp.parse(
-        wxr.wtp.node_to_wikitext(template_node), expand_all=True
+        wxr.wtp.node_to_wikitext(t_node), expand_all=True
     )
     table_nodes = list(expanded_template.find_child(NodeKind.TABLE))
     if len(table_nodes) == 0:
@@ -137,20 +137,22 @@ def process_verb_table(
 
 
 def process_noun_table(
-    wxr: WiktextractContext, word_entry: WordEntry, template_node: TemplateNode
+    wxr: WiktextractContext, word_entry: WordEntry, t_node: TemplateNode
 ) -> None:
     # Vorlage:Deutsch Substantiv Übersicht
     from .page import extract_note_section
 
     expanded_template = wxr.wtp.parse(
-        wxr.wtp.node_to_wikitext(template_node), expand_all=True
+        wxr.wtp.node_to_wikitext(t_node), expand_all=True
     )
     table_nodes = list(expanded_template.find_child(NodeKind.TABLE))
     if len(table_nodes) == 0:
         return
     table_node = table_nodes[0]
     column_headers = []
     table_header = ""
+    forms = []
+    flexion_pages = []
     for table_row in table_node.find_child(NodeKind.TABLE_ROW):
         row_header = ""
         is_header_row = not table_row.contain_node(NodeKind.TABLE_CELL)
@@ -188,7 +190,7 @@ def process_noun_table(
                 for link_node in table_cell.find_child(NodeKind.LINK):
                     link_text = clean_node(wxr, None, link_node)
                     if link_text.startswith("Flexion:"):
-                        parse_flexion_page(wxr, word_entry, link_text)
+                        flexion_pages.append(link_text)
             else:
                 for form_text in cell_text.splitlines():
                     form_text = form_text.strip()
@@ -209,10 +211,15 @@ def process_noun_table(
                         ):
                             form.raw_tags.append(col_header.text)
                     translate_raw_tags(form)
-                    word_entry.forms.append(form)
+                    forms.append(form)
                 col_index += 1
 
+    if t_node.template_name == "Deutsch Substantiv Übersicht":
+        forms = seprarte_de_article(wxr, forms)
+    word_entry.forms.extend(forms)
     clean_node(wxr, word_entry, expanded_template)  # category links
+    for flexion_page in flexion_pages:
+        parse_flexion_page(wxr, word_entry, flexion_page)
     # Vorlage:Deutsch Nachname Übersicht
     for level_node in expanded_template.find_child(NodeKind.LEVEL4):
         section_text = clean_node(wxr, None, level_node.largs)
@@ -326,3 +333,19 @@ def extract_pronoun_table(
                         word_entry.forms.append(form)
                     article = ""
                 col_index += 1
+
+
+def seprarte_de_article(
+    wxr: WiktextractContext, forms: list[Form]
+) -> list[Form]:
+    # https://de.wiktionary.org/wiki/Vorlage:Deutsch_Substantiv_Übersicht
+    # https://en.wikipedia.org/wiki/German_articles
+    new_forms = []
+    for form in forms:
+        m = re.match(r"(der|die|das|den|dem|des)\s+", form.form)
+        if m is not None:
+            form.form = form.form[m.end() :]
+            form.article = m.group(1)
+            if form.form != wxr.wtp.title:
+                new_forms.append(form)
+    return new_forms

src/wiktextract/extractor/de/models.py

@@ -101,6 +101,7 @@ class Form(GermanBaseModel):
     sense_index: str = ""
     topics: list[str] = []
     pronouns: list[str] = []
+    article: str = ""
 
 
 class Descendant(GermanBaseModel):

tests/test_de_forms.py

@@ -47,9 +47,21 @@ def test_noun_table(self):
         self.assertEqual(
             word_entry.model_dump(exclude_defaults=True)["forms"],
             [
-                {"form": "des Wörterbuches", "tags": ["genitive", "singular"]},
-                {"form": "des Wörterbuchs", "tags": ["genitive", "singular"]},
-                {"form": "der Wörterbücher", "tags": ["genitive", "plural"]},
+                {
+                    "article": "des",
+                    "form": "Wörterbuches",
+                    "tags": ["genitive", "singular"],
+                },
+                {
+                    "article": "des",
+                    "form": "Wörterbuchs",
+                    "tags": ["genitive", "singular"],
+                },
+                {
+                    "article": "der",
+                    "form": "Wörterbücher",
+                    "tags": ["genitive", "plural"],
+                },
             ],
         )