[pt] extract etymology section

src/wiktextract/extractor/pt/etymology.py

@@ -0,0 +1,43 @@
+from wikitextprocessor.parser import (
+    LEVEL_KIND_FLAGS,
+    LevelNode,
+    NodeKind,
+    WikiNode,
+)
+
+from ...page import clean_node
+from ...wxr_context import WiktextractContext
+from .models import WordEntry
+
+
+def extract_etymology_section(
+    wxr: WiktextractContext,
+    page_data: list[WordEntry],
+    level_node: LevelNode,
+) -> None:
+    cats = {}
+    e_nodes = []
+    e_texts = []
+    for node in level_node.children:
+        if isinstance(node, WikiNode) and node.kind in LEVEL_KIND_FLAGS:
+            break
+        elif isinstance(node, WikiNode) and node.kind == NodeKind.LIST:
+            e_text = clean_node(wxr, cats, e_nodes).lstrip(": ")
+            if e_text != "":
+                e_texts.append(e_text)
+            e_nodes.clear()
+            for list_item in node.find_child(NodeKind.LIST_ITEM):
+                e_text = clean_node(wxr, cats, list_item.children)
+                if e_text != "":
+                    e_texts.append(e_text)
+        else:
+            e_nodes.append(node)
+
+    if len(e_nodes) > 0:
+        e_text = clean_node(wxr, cats, e_nodes).lstrip(": ")
+        if e_text != "":
+            e_texts.append(e_text)
+    for data in page_data:
+        if data.lang_code == page_data[-1].lang_code:
+            data.etymology_texts.extend(e_texts)
+            data.categories.extend(cats.get("categories", []))

src/wiktextract/extractor/pt/models.py

@@ -68,3 +68,4 @@ class WordEntry(PortugueseBaseModel):
     antonyms: list[Linkage] = []
     synonyms: list[Linkage] = []
     derived: list[Linkage] = []
+    etymology_texts: list[str] = []

src/wiktextract/extractor/pt/page.py

@@ -8,6 +8,7 @@
 
 from ...page import clean_node
 from ...wxr_context import WiktextractContext
+from .etymology import extract_etymology_section
 from .linkage import extract_expression_section, extract_linkage_section
 from .models import Sense, WordEntry
 from .pos import extract_pos_section
@@ -47,6 +48,8 @@ def parse_section(
             level_node,
             LINKAGE_SECTIONS[title_text],
         )
+    elif title_text == "Etimologia":
+        extract_etymology_section(wxr, page_data, level_node)
 
     cats = {}
     for link_node in level_node.find_child(NodeKind.LINK):

src/wiktextract/extractor/pt/section_titles.py

@@ -11,6 +11,7 @@
     "Pronome": {"pos": "pron"},
     "Substantivo": {"pos": "noun"},
     "Verbo": {"pos": "verb"},
+    "Forma de substantivo": {"pos": "noun", "tags": ["form-of"]},
     "Forma verbal": {"pos": "verb", "tags": ["form-of"]},
     "Locução substantiva": {"pos": "phrase", "tags": ["substantive"]},
     "Locução adjetiva": {"pos": "phrase", "tags": ["adjectival"]},

src/wiktextract/extractor/pt/translation.py

@@ -125,16 +125,17 @@ def extract_trad_template(
         if arg not in t_node.template_parameters:
             break
         tr_str = clean_node(wxr, None, t_node.template_parameters.get(arg, ""))
-        translations.append(
-            Translation(
-                word=tr_str,
-                lang=lang_name,
-                lang_code=lang_code,
-                roman=roman,
-                sense=sense,
-                sense_index=sense_index,
+        if tr_str != "":
+            translations.append(
+                Translation(
+                    word=tr_str,
+                    lang=lang_name,
+                    lang_code=lang_code,
+                    roman=roman,
+                    sense=sense,
+                    sense_index=sense_index,
+                )
             )
-        )
     return translations
 
 

tests/test_pt_etymology.py

@@ -0,0 +1,49 @@
+from unittest import TestCase
+
+from wikitextprocessor import Wtp
+
+from wiktextract.config import WiktionaryConfig
+from wiktextract.extractor.pt.page import parse_page
+from wiktextract.wxr_context import WiktextractContext
+
+
+class TestPtEtymology(TestCase):
+    maxDiff = None
+
+    def setUp(self) -> None:
+        conf = WiktionaryConfig(
+            dump_file_lang_code="pt",
+            capture_language_codes=None,
+        )
+        self.wxr = WiktextractContext(
+            Wtp(
+                lang_code="pt",
+                parser_function_aliases=conf.parser_function_aliases,
+            ),
+            conf,
+        )
+
+    def test_list(self):
+        self.wxr.wtp.add_page("Predefinição:-pt-", 10, "Português")
+        self.wxr.wtp.add_page(
+            "Predefinição:etimo2",
+            10,
+            """:[[Categoria:Entrada de étimo latino (Português)]]Do [[latim]] ''[[oculus#la|oculus]]''<small><sup> ([[:la:oculus|<span title="ver no Wikcionário em latim">la</span>]])</sup></small>.""",
+        )
+        data = parse_page(
+            self.wxr,
+            "olho",
+            """={{-pt-}}=
+==Substantivo==
+# órgão
+==Etimologia==
+{{etimo2|la|oculus|pt}}
+:* '''Datação''': [[w:século XIII|século XIII]]""",
+        )
+        self.assertEqual(
+            data[0]["etymology_texts"],
+            ["Do latim oculus⁽ˡᵃ⁾.", "Datação: século XIII"],
+        )
+        self.assertEqual(
+            data[0]["categories"], ["Entrada de étimo latino (Português)"]
+        )