From d6317feb8c7278e39e95383921a9ab1a352ebb18 Mon Sep 17 00:00:00 2001
From: xxyzz <gitpull@protonmail.com>
Date: Tue, 19 Nov 2024 11:52:31 +0800
Subject: [PATCH] [nl] extract "-csadjc-comp-*" forms table templates

---
 src/wiktextract/extractor/nl/inflection.py | 27 ++++++++++++++++++++++
 src/wiktextract/extractor/nl/page.py       | 15 +++++++-----
 2 files changed, 36 insertions(+), 6 deletions(-)

diff --git a/src/wiktextract/extractor/nl/inflection.py b/src/wiktextract/extractor/nl/inflection.py
index 9c8804db..c8ad4e08 100644
--- a/src/wiktextract/extractor/nl/inflection.py
+++ b/src/wiktextract/extractor/nl/inflection.py
@@ -22,6 +22,8 @@ def extract_inflection_template(
         extract_noun_adj_table(wxr, word_entry, t_node)
     elif t_node.template_name == "-nlstam-":
         extract_nlstam_template(wxr, word_entry, t_node)
+    elif t_node.template_name.startswith("-csadjc-comp-"):
+        extract_csadjc_comp_template(wxr, word_entry, t_node)
 
 
 def extract_noun_adj_table(
@@ -241,3 +243,28 @@ def nlverb_table_cell_is_header(node: WikiNode) -> bool:
         node.kind == NodeKind.TABLE_HEADER_CELL
         or node.attrs.get("class", "") == "infoboxrijhoofding"
     )
+
+
+def extract_csadjc_comp_template(
+    wxr: WiktextractContext, word_entry: WordEntry, t_node: TemplateNode
+) -> None:
+    # https://nl.wiktionary.org/wiki/Sjabloon:-csadjc-comp-ý3-
+    expanded_node = wxr.wtp.parse(
+        wxr.wtp.node_to_wikitext(t_node), expand_all=True
+    )
+    for table in expanded_node.find_child(NodeKind.TABLE):
+        for row in table.find_child(NodeKind.TABLE_ROW):
+            row_header = ""
+            for cell_node in row.find_child(
+                NodeKind.TABLE_HEADER_CELL | NodeKind.TABLE_CELL
+            ):
+                if cell_node.kind == NodeKind.TABLE_HEADER_CELL:
+                    row_header = clean_node(wxr, None, cell_node)
+                elif cell_node.kind == NodeKind.TABLE_CELL:
+                    form_text = clean_node(wxr, None, cell_node)
+                    if form_text not in ["", wxr.wtp.title]:
+                        form = Form(form=form_text)
+                        if row_header != "":
+                            form.raw_tags.append(row_header)
+                            translate_raw_tags(form)
+                        word_entry.forms.append(form)
diff --git a/src/wiktextract/extractor/nl/page.py b/src/wiktextract/extractor/nl/page.py
index 296d8de6..24db9f64 100644
--- a/src/wiktextract/extractor/nl/page.py
+++ b/src/wiktextract/extractor/nl/page.py
@@ -92,10 +92,6 @@ def parse_section(
         extract_fixed_preposition_section(
             wxr, page_data[-1] if len(page_data) > 0 else base_data, level_node
         )
-    elif title_text == "Vervoeging":
-        pass  # conjugation
-    elif title_text == "Verbuiging":
-        pass  # inflection
     elif title_text in [
         "Gangbaarheid",
         "Meer informatie",
@@ -103,7 +99,7 @@ def parse_section(
         "Citaten",
     ]:
         pass  # ignore
-    else:
+    elif not title_text.startswith(("Vervoeging", "Verbuiging")):
         wxr.wtp.debug(f"unknown title: {title_text}", sortid="nl/page/60")
 
     for next_level in level_node.find_child(LEVEL_KIND_FLAGS):
@@ -112,7 +108,14 @@ def parse_section(
         wxr, page_data[-1] if len(page_data) > 0 else base_data, level_node
     )
     for t_node in level_node.find_child(NodeKind.TEMPLATE):
-        extract_inflection_template(wxr, forms_data, t_node)
+        extract_inflection_template(
+            wxr,
+            page_data[-1]
+            if title_text.startswith(("Vervoeging", "Verbuiging"))
+            and len(page_data) > 0
+            else forms_data,
+            t_node,
+        )
     return etymology_data