From d46946868ce1801e4e4fb3e8e67f9d3d2e8f6304 Mon Sep 17 00:00:00 2001
From: xxyzz <gitpull@protonmail.com>
Date: Mon, 30 Oct 2023 16:32:10 +0800
Subject: [PATCH] Extract nested gloss and example lists in French Wiktionary

---
 src/wiktextract/extractor/fr/gloss.py      | 17 +++++----
 src/wiktextract/extractor/fr/inflection.py |  3 +-
 tests/test_fr_gloss.py                     | 41 ++++++++++++++++++++++
 tests/test_fr_linkage.py                   |  4 +--
 4 files changed, 54 insertions(+), 11 deletions(-)

diff --git a/src/wiktextract/extractor/fr/gloss.py b/src/wiktextract/extractor/fr/gloss.py
index 56ccea4b..c63a1abb 100644
--- a/src/wiktextract/extractor/fr/gloss.py
+++ b/src/wiktextract/extractor/fr/gloss.py
@@ -11,6 +11,7 @@ def extract_gloss(
     wxr: WiktextractContext,
     page_data: List[Dict],
     list_node: WikiNode,
+    parent_glosses: List[str] = [],
 ) -> None:
     for list_item_node in list_node.find_child(NodeKind.LIST_ITEM):
         gloss_nodes = list(
@@ -63,19 +64,23 @@ def extract_gloss(
             if index not in tag_indexes
         ]
         gloss_text = clean_node(wxr, gloss_data, gloss_only_nodes)
-        gloss_data["glosses"] = [gloss_text]
-        extract_examples(wxr, gloss_data, list_item_node)
+        gloss_data["glosses"] = parent_glosses + [gloss_text]
         page_data[-1]["senses"].append(gloss_data)
+        for nest_gloss_list in list_item_node.find_child(NodeKind.LIST):
+            if nest_gloss_list.sarg.endswith("#"):
+                extract_gloss(
+                    wxr, page_data, nest_gloss_list, gloss_data["glosses"]
+                )
+            elif nest_gloss_list.sarg.endswith("*"):
+                extract_examples(wxr, gloss_data, nest_gloss_list)
 
 
 def extract_examples(
     wxr: WiktextractContext,
     gloss_data: Dict,
-    gloss_list_node: WikiNode,
+    example_list_node: WikiNode,
 ) -> None:
-    for example_node in gloss_list_node.find_child_recursively(
-        NodeKind.LIST_ITEM
-    ):
+    for example_node in example_list_node.find_child(NodeKind.LIST_ITEM):
         example_node_children = list(example_node.filter_empty_str_child())
         if len(example_node_children) == 0:
             continue
diff --git a/src/wiktextract/extractor/fr/inflection.py b/src/wiktextract/extractor/fr/inflection.py
index 9479a222..8d48501c 100644
--- a/src/wiktextract/extractor/fr/inflection.py
+++ b/src/wiktextract/extractor/fr/inflection.py
@@ -79,8 +79,7 @@ def process_inflection_table(
             and "invisible" not in row_node_child.attrs.get("class", "")
         ]
         current_row_has_data_cell = any(
-            isinstance(cell, WikiNode)
-            and cell.kind == NodeKind.TABLE_CELL
+            isinstance(cell, WikiNode) and cell.kind == NodeKind.TABLE_CELL
             for cell in table_row_nodes
         )
         row_headers = []
diff --git a/tests/test_fr_gloss.py b/tests/test_fr_gloss.py
index ff6a1395..b559a4d6 100644
--- a/tests/test_fr_gloss.py
+++ b/tests/test_fr_gloss.py
@@ -257,3 +257,44 @@ def test_template_is_not_tag(self, mock_get_page):
                 }
             ],
         )
+
+    def test_nest_gloss(self):
+        self.maxDiff = None
+        self.wxr.wtp.start_page("eau")
+        root = self.wxr.wtp.parse(
+            """# [[fluide|Fluides]], [[sérosité]]s qui se trouvent ou qui se forment dans le [[corps]] de l’[[homme]] ou de l’[[animal]].
+#* example 1
+## [[salive|Salive]].
+##* nest example
+            """
+        )
+        page_data = [defaultdict(list)]
+        extract_gloss(self.wxr, page_data, root.children[0])
+        self.assertEqual(
+            page_data[-1]["senses"],
+            [
+                {
+                    "examples": [
+                        {
+                            "text": "example 1",
+                            "type": "example",
+                        }
+                    ],
+                    "glosses": [
+                        "Fluides, sérosités qui se trouvent ou qui se forment dans le corps de l’homme ou de l’animal."
+                    ],
+                },
+                {
+                    "examples": [
+                        {
+                            "text": "nest example",
+                            "type": "example",
+                        }
+                    ],
+                    "glosses": [
+                        "Fluides, sérosités qui se trouvent ou qui se forment dans le corps de l’homme ou de l’animal.",
+                        "Salive.",
+                    ],
+                },
+            ],
+        )
diff --git a/tests/test_fr_linkage.py b/tests/test_fr_linkage.py
index b4e2f1aa..498faf3f 100644
--- a/tests/test_fr_linkage.py
+++ b/tests/test_fr_linkage.py
@@ -183,9 +183,7 @@ def test_derives_autres_langues_section(self):
     def test_words_divided_by_slash(self):
         page_data = [defaultdict(list)]
         self.wxr.wtp.start_page("eau")
-        root = self.wxr.wtp.parse(
-            "* [[benoîte d’eau]] / [[benoite d’eau]]"
-        )
+        root = self.wxr.wtp.parse("* [[benoîte d’eau]] / [[benoite d’eau]]")
         extract_linkage(self.wxr, page_data, root, "dérivés")
         self.assertEqual(
             page_data,