[it] extract "It-decl-agg*" table templates

src/wiktextract/extractor/it/inflection.py

@@ -1,8 +1,9 @@
-from wikitextprocessor import TemplateNode
+from wikitextprocessor import NodeKind, TemplateNode
 
 from ...page import clean_node
 from ...wxr_context import WiktextractContext
 from .models import Form, WordEntry
+from .tags import translate_raw_tags
 
 
 def extract_tabs_template(
@@ -22,3 +23,56 @@ def extract_tabs_template(
         if arg_value not in ["", wxr.wtp.title]:
             form = Form(form=arg_value, tags=tags[arg_name - 1])
             word_entry.forms.append(form)
+
+
+def extract_it_decl_agg_template(
+    wxr: WiktextractContext, word_entry: WordEntry, t_node: TemplateNode
+) -> None:
+    # https://it.wiktionary.org/wiki/Template:It-decl-agg4
+    # https://it.wiktionary.org/wiki/Template:It-decl-agg2
+    expanded_node = wxr.wtp.parse(
+        wxr.wtp.node_to_wikitext(t_node), expand_all=True
+    )
+    for table in expanded_node.find_child(NodeKind.TABLE):
+        raw_tag = ""
+        col_tags = []
+        for row in table.find_child(NodeKind.TABLE_ROW):
+            row_tag = ""
+            col_index = 0
+            for cell in row.find_child(
+                NodeKind.TABLE_HEADER_CELL | NodeKind.TABLE_CELL
+            ):
+                match cell.kind:
+                    case NodeKind.TABLE_HEADER_CELL:
+                        col_span = cell.attrs.get("colspan", "")
+                        if col_span != "":
+                            raw_tag = clean_node(wxr, None, cell)
+                        elif (
+                            len(
+                                [
+                                    n
+                                    for n in row.find_child(
+                                        NodeKind.TABLE_HEADER_CELL
+                                    )
+                                ]
+                            )
+                            == 1
+                        ):
+                            row_tag = clean_node(wxr, None, cell)
+                        else:
+                            col_header = clean_node(wxr, None, cell)
+                            if col_header != "":
+                                col_tags.append(col_header)
+                    case NodeKind.TABLE_CELL:
+                        word = clean_node(wxr, None, cell)
+                        if word not in ["", wxr.wtp.title]:
+                            form = Form(form=word)
+                            if raw_tag != "":
+                                form.raw_tags.append(raw_tag)
+                            if row_tag != "":
+                                form.raw_tags.append(row_tag)
+                            if col_index < len(col_tags):
+                                form.raw_tags.append(col_tags[col_index])
+                            translate_raw_tags(form)
+                            word_entry.forms.append(form)
+                        col_index += 1

src/wiktextract/extractor/it/page.py

@@ -53,6 +53,7 @@ def parse_page(
 ) -> list[dict[str, Any]]:
     # page layout
     # https://it.wiktionary.org/wiki/Wikizionario:Manuale_di_stile
+    # https://it.wiktionary.org/wiki/Aiuto:Come_iniziare_una_pagina
     wxr.wtp.start_page(page_title)
     tree = wxr.wtp.parse(page_text, pre_expand=True)
     page_data: list[WordEntry] = []

src/wiktextract/extractor/it/tag_form_line.py

@@ -2,7 +2,7 @@
 
 from ...page import clean_node
 from ...wxr_context import WiktextractContext
-from .inflection import extract_tabs_template
+from .inflection import extract_it_decl_agg_template, extract_tabs_template
 from .models import Form, WordEntry
 
 
@@ -18,6 +18,8 @@ def extract_tag_form_line_nodes(
                 extract_tabs_template(wxr, word_entry, node)
             elif node.template_name.lower() in FORM_LINK_TEMPLATES.keys():
                 extract_form_link_template(wxr, word_entry, node)
+            elif node.template_name.lower().startswith("it-decl-agg"):
+                extract_it_decl_agg_template(wxr, word_entry, node)
 
 
 ITALIC_TAGS = {

src/wiktextract/extractor/it/tags.py

@@ -0,0 +1,30 @@
+from .models import WordEntry
+
+TABLE_TAGS = {
+    # https://it.wiktionary.org/wiki/Template:It-decl-agg4
+    "singolare": "singular",
+    "plurale": "plural",
+    "positivo": "positive",
+    "superlativo assoluto": ["absolute", "superlative"],
+    "maschile": "masculine",
+    "femminile": "feminine",
+    # https://it.wiktionary.org/wiki/Template:It-decl-agg2
+    "m e f": ["masculine", "feminine"],
+}
+
+
+TAGS = {**TABLE_TAGS}
+
+
+def translate_raw_tags(data: WordEntry) -> None:
+    raw_tags = []
+    for raw_tag in data.raw_tags:
+        if raw_tag in TAGS:
+            tr_tag = TAGS[raw_tag]
+            if isinstance(tr_tag, str):
+                data.tags.append(tr_tag)
+            elif isinstance(tr_tag, list):
+                data.tags.extend(tr_tag)
+        else:
+            raw_tags.append(raw_tag)
+    data.raw_tags = raw_tags

tests/test_it_forms.py

@@ -55,3 +55,31 @@ def test_linkp_template(self):
             [{"form": "cagne", "tags": ["plural"]}],
         )
         self.assertEqual(data[0]["tags"], ["feminine", "singular"])
+
+    def test_it_decl_agg(self):
+        self.wxr.wtp.add_page("Template:-it-", 10, "Italiano")
+        self.wxr.wtp.add_page("Template:It-decl-agg4", 10, """{|
+|- align="center"
+|  
+!bgcolor="#FFFFE0" color="#000"| ''[[singolare]]'' 
+!bgcolor="#FFFFE0" color="#000"| ''[[plurale]]'' 
+|- align="center"
+!bgcolor="#FFFFE0" color="#000" colspan="3"| ''[[positivo]]'' 
+|- align="center"
+!bgcolor="#FFFFE0" color="#000"| ''[[maschile]]'' 
+|  [[libero]]  
+|  [[liberi]]  
+|}""")
+        data = parse_page(
+            self.wxr,
+            "libero",
+            """== {{-it-}} ==
+===Aggettivo===
+{{It-decl-agg4|liber}}
+{{Pn|w}} ''m sing''
+# non [[imprigionato]] o in [[schiavitù]]""",
+        )
+        self.assertEqual(
+            data[0]["forms"],
+            [{"form": "liberi", "tags": ["positive", "masculine", "plural"]}],
+        )