fix BoboTiG#237: [FR] Sub sub list not handled correctly

lasconic · Nov 21, 2020 · 880ea51 · 880ea51
1 parent 4c582e5
commit 880ea51
Show file tree

Hide file tree

Showing 5 changed files with 42 additions and 9 deletions.
diff --git a/scripts/convert.py b/scripts/convert.py
@@ -213,7 +213,15 @@ def save_html(
                         definitions += f"<li>{definition}</li>"
                     else:
                         definitions += '<ol style="list-style-type:lower-alpha">'
-                        definitions += "".join(f"<li>{d}</li>" for d in definition)
+                        for subdef in definition:
+                            if isinstance(subdef, str):
+                                definitions += f"<li>{subdef}</li>"
+                            else:
+                                definitions += (
+                                    '<ol style="list-style-type:lower-roman">'
+                                )
+                                definitions += "".join(f"<li>{d}</li>" for d in subdef)
+                                definitions += "</ol>"
                         definitions += "</ol>"
 
                 pronunciation = convert_pronunciation(details.pronunciations)

diff --git a/scripts/get.py b/scripts/get.py
@@ -37,7 +37,7 @@
     sublist_patterns,
     words_to_keep,
 )
-from .stubs import Definitions, Word, Words
+from .stubs import Definitions, SubDefinitions, Word, Words
 from .utils import clean, convert_pronunciation, convert_genre
 
 if TYPE_CHECKING:  # pragma: nocover
@@ -183,10 +183,21 @@ def find_section_definitions(
                 definitions.append(definition)
 
                 # ... And its eventual sub-definitions
-                subdefinitions: List[str] = []
+                subdefinitions: List[SubDefinitions] = []
                 for sublist in a_list.sublists(i=idx, pattern=sublist_patterns[locale]):
-                    for subcode in sublist.items:
-                        subdefinitions.append(clean(word, subcode, locale))
+                    for idx2, subcode in enumerate(sublist.items):
+                        subdefinition = clean(word, subcode, locale)
+                        subdefinitions.append(subdefinition)
+                        subsubdefinitions: List[str] = []
+                        for subsublist in sublist.sublists(
+                            i=idx2, pattern=sublist_patterns[locale]
+                        ):
+                            for subsubcode in subsublist.items:
+                                subsubdefinitions.append(
+                                    clean(word, subsubcode, locale)
+                                )
+                        if subsubdefinitions:
+                            subdefinitions.append(tuple(subsubdefinitions))
                 if subdefinitions:
                     definitions.append(tuple(subdefinitions))
 
@@ -356,12 +367,20 @@ def strip_html(text: str) -> str:
     if details.etymology:
         print(strip_html(details.etymology), "\n")
 
-    for i, definition in enumerate(details.definitions, start=1):
+    index = 1
+    for definition in details.definitions:
         if isinstance(definition, tuple):
             for a, subdef in zip("abcdefghijklmopqrstuvwxz", definition):
-                print(f"{a}.".rjust(8), strip_html(subdef))
+                if isinstance(subdef, tuple):
+                    for rn, subsubdef in zip(
+                        ["i", "ii", "iii", "iv", "v", "vi", "vii"], subdef
+                    ):
+                        print(f"{rn}.".rjust(12), strip_html(subsubdef))
+                else:
+                    print(f"{a}.".rjust(8), strip_html(subdef))
         else:
-            print(f"{i}.".rjust(4), strip_html(definition))
+            print(f"{index}.".rjust(4), strip_html(definition))
+            index = index + 1
 
     if details.variants:
         print("[variants]", ", ".join(iter(details.variants)))

diff --git a/scripts/stubs.py b/scripts/stubs.py
@@ -2,7 +2,8 @@
 from collections import namedtuple
 from typing import Dict, Tuple, Union
 
-Definitions = Union[str, Tuple[str, ...]]
+SubDefinitions = Union[str, Tuple[str, ...]]
+Definitions = Union[str, Tuple[str, ...], Tuple[SubDefinitions, ...]]
 Parts = Tuple[str, ...]
 Word = namedtuple("Word", "pronunciations, genre, etymology, definitions, variants")
 Words = Dict[str, Word]
diff --git a/tests/test_0_get.py b/tests/test_0_get.py
@@ -480,4 +480,5 @@ def test_xml_parse_word_special_character(tmp_path):
 
 def test_get_and_parse_word():
     get.get_and_parse_word("fondation", "fr")
+    get.get_and_parse_word("base", "fr")
     get.get_and_parse_word("suis", "fr", raw=True)
diff --git a/tests/test_fr.py b/tests/test_fr.py
@@ -100,6 +100,10 @@
                 "<i>(Mathématiques)</i>",
                 (
                     "<i>(Géométrie)</i> Surface sur laquelle on conçoit que certains corps solides sont appuyés.",
+                    (
+                        "<i>(Par extension)</i> Côté du triangle opposé à l’angle qui est regardé comme le sommet.",
+                        "Côté d’une figure géométrique naturellement choisi comme côté principal.",
+                    ),
                     "<i>(Algèbre)</i> Famille libre de vecteurs, génératrice d’un espace vectoriel.",
                     "<i>(Analyse réelle)</i> Nombre réel élevé à une puissance variable.",
                     "<i>(Arithmétique)</i> Nombre de chiffres utilisé pour dénombrer.",