Test German parse_section() with at least some sense data

tatuylonen · Oct 5, 2023 · e3af868 · e3af868
1 parent e6d3bfa
commit e3af868
Showing 1 changed file with 8 additions and 85 deletions.
diff --git a/tests/test_de_page.py b/tests/test_de_page.py
@@ -1,7 +1,6 @@
 # Tests for parsing a page from the German Wiktionary
 
 import unittest
-from unittest.mock import patch
 
 from collections import defaultdict
 
@@ -81,57 +80,10 @@ def test_de_parse_page_skipping_head_templates(self):
             ],
         )
 
-    def mock_append_base_data_side_effects(
-        self, page_data, field: str, value, base_data
-    ) -> None:
-        import copy
-
-        if page_data[-1].get(field) is not None:
-            if len(page_data[-1]["senses"]) > 0:
-                # append new dictionary if the last dictionary has sense data and
-                # also has the same key
-                page_data.append(copy.deepcopy(base_data))
-            elif isinstance(page_data[-1].get(field), list):
-                page_data[-1][field] += value
-            else:
-                page_data.append(copy.deepcopy(base_data))
-
-        else:
-            page_data[-1][field] = value
-
-    @patch("wiktextract.extractor.de.page.append_base_data")
-    def test_de_parse_section_with_mock(self, mock_append_base_data):
-        mock_append_base_data.side_effect = (
-            self.mock_append_base_data_side_effects
-        )
-
-        self.wxr.wtp.add_page("Vorlage:Wortart", 10, "")
-        page_text = """
-=== {{Wortart|Adjektiv|Englisch}}, {{Wortart|Adverb|Englisch}} ===
-=== {{Wortart|Verb|Englisch}} ===
-=== {{Wortart|Substantiv|Englisch}} ===
-"""
-        self.wxr.wtp.start_page("")
-        root = self.wxr.wtp.parse(
-            page_text,
-            pre_expand=True,
-        )
-
-        base_data = defaultdict(list, {"lang_code": "de"})
-        page_data = [defaultdict(list, {"lang_code": "de"})]
-        parse_section(self.wxr, page_data, base_data, root.children)
-
-        self.assertEqual(
-            page_data,
-            [
-                {"lang_code": "de", "pos": "adj", "senses": []},
-                {"lang_code": "de", "pos": "adv", "senses": []},
-                {"lang_code": "de", "pos": "verb", "senses": []},
-                {"lang_code": "de", "pos": "noun"},
-            ],
-        )
-
-    def test_de_parse_section_with_senses(self):
+    # The way append_base_data() works requires the presence of a sense
+    # dictionary before starting a new pos section. Therefore, we need to add
+    # at least one sense data point to the test case.
+    def test_de_parse_section(self):
         self.wxr.wtp.add_page("Vorlage:Wortart", 10, "")
         self.wxr.wtp.add_page("Vorlage:Bedeutungen", 10, "")
         page_text = """
@@ -140,10 +92,10 @@ def test_de_parse_section_with_senses(self):
 :[1] gloss1
 === {{Wortart|Verb|Englisch}} ===
 {{Bedeutungen}}
-:[1] gloss1
+:[1] gloss2
 === {{Wortart|Substantiv|Englisch}} ===
 {{Bedeutungen}}
-:[1] gloss1
+:[1] gloss3
 
 """
         self.wxr.wtp.start_page("")
@@ -182,7 +134,7 @@ def test_de_parse_section_with_senses(self):
                     "pos": "verb",
                     "senses": [
                         {
-                            "glosses": ["gloss1"],
+                            "glosses": ["gloss2"],
                         },
                     ],
                 },
@@ -191,42 +143,13 @@ def test_de_parse_section_with_senses(self):
                     "pos": "noun",
                     "senses": [
                         {
-                            "glosses": ["gloss1"],
+                            "glosses": ["gloss3"],
                         },
                     ],
                 },
             ],
         )
 
-    def test_de_parse_section_without_mock(self):
-        self.wxr.wtp.add_page("Vorlage:Wortart", 10, "")
-        page_text = """
-=== {{Wortart|Adjektiv|Englisch}}, {{Wortart|Adverb|Englisch}} ===
-=== {{Wortart|Verb|Englisch}} ===
-=== {{Wortart|Substantiv|Englisch}} ===
-"""
-        self.wxr.wtp.start_page("")
-        root = self.wxr.wtp.parse(
-            page_text,
-            pre_expand=True,
-        )
-
-        base_data = defaultdict(list, {"lang_code": "de"})
-        page_data = [defaultdict(list, {"lang_code": "de"})]
-        parse_section(self.wxr, page_data, base_data, root.children)
-
-        self.assertEqual(
-            page_data,
-            [
-                {"lang_code": "de", "pos": "adj", "senses": []},
-                {"lang_code": "de", "pos": "adv", "senses": []},
-                {"lang_code": "de", "pos": "verb", "senses": []},
-                {"lang_code": "de", "pos": "noun"},
-            ],
-        )
-        # Assertion fails. The resulting page_data is just:
-        # [{'lang_code': 'de', 'pos': 'adj', 'senses': []}]
-
     def test_de_fix_level_hierarchy_of_subsections(self):
         self.wxr.wtp.add_page("Vorlage:Englisch Substantiv Übersicht", 10, "")
         self.wxr.wtp.add_page("Vorlage:Worttrennung", 10, "")