Remove images and URLs without any text when converting HTML to Markdown

discord_rss_bot/markdown.py

@@ -10,6 +10,9 @@ def convert_html_to_md(html: str) -> str:
     Returns:
         Our markdown.
     """
+    if not html:
+        return html
+
     soup: BeautifulSoup = BeautifulSoup(html, features="lxml")
 
     for bold in soup.find_all("b") + soup.find_all("strong"):
@@ -24,9 +27,15 @@ def convert_html_to_md(html: str) -> str:
     for code in soup.find_all("code") + soup.find_all("pre"):
         code.replace_with(f"`{code.text}`")
 
+    for image in soup.find_all("img"):
+        image.decompose()
+
     for link in soup.find_all("a") + soup.find_all("link"):
-        link_text: str = link.text or link.get("href") or "Link"
-        link.replace_with(f"[{link_text}]({link.get('href')})")
+        if not link.get_text().strip():
+            link.decompose()
+        else:
+            link_text: str = link.text or link.get("href")
+            link.replace_with(f"[{link_text}]({link.get('href')})")
 
     for strikethrough in soup.find_all("s") + soup.find_all("del") + soup.find_all("strike"):
         strikethrough.replace_with(f"~~{strikethrough.text}~~")