improve nimbus note conversion (by adding html preprocessors)

src/formats/nimbus_note.py

@@ -3,12 +3,33 @@
 import base64
 from pathlib import Path
 
+from bs4 import BeautifulSoup
+
 import common
 import converter
 import intermediate_format as imf
 import markdown_lib.common
 
 
+def clean_tables(soup):
+    for table in soup.find_all("table"):
+        tags_to_remove = ["div", "span"]
+        for tag in tags_to_remove:
+            for element in table.find_all(tag):
+                element.unwrap()
+
+
+def assign_lists(_soup):
+    # TODO:
+    # - all lists are unnumbered lists (ul)
+    #   - type is in the class attr (list-item-number, -bullet, -checkbox)
+    # - indentation is in the class attr (indent-0)
+
+    # for task_list in soup.find_all("ul", class_="checklist"):
+    #     ...
+    pass
+
+
 class Converter(converter.BaseConverter):
     accept_folder = True
 
@@ -55,8 +76,15 @@ def convert(self, file_or_folder: Path):
             common.extract_zip(file_, temp_folder=temp_folder_note)
 
             # HTML note seems to have the name "note.html" always
-            note_body_html = (temp_folder_note / "note.html").read_text("utf-8")
-            note_body_markdown = markdown_lib.common.markup_to_markdown(note_body_html)
+            note_body_html = (temp_folder_note / "note.html").read_text(
+                encoding="utf-8"
+            )
+
+            soup = BeautifulSoup(note_body_html, "html.parser")
+            clean_tables(soup)
+            assign_lists(soup)
+
+            note_body_markdown = markdown_lib.common.markup_to_markdown(str(soup))
             resources = self.handle_markdown_links(note_body_markdown, temp_folder_note)
             note_imf = imf.Note(
                 title,