Add strikethrough/underline <-> unicode converter to formatter

mautrix_telegram/formatter/from_matrix.py

@@ -25,7 +25,8 @@ from telethon_aio.tl.types import *
 
 from .. import user as u, puppet as pu, portal as po
 from ..db import Message as DBMessage
-from .util import (add_surrogates, remove_surrogates, trim_reply_fallback_html, trim_reply_fallback_text)
+from .util import (add_surrogates, remove_surrogates, trim_reply_fallback_html,
+                   trim_reply_fallback_text, html_to_unicode)
 
 log = logging.getLogger("mau.fmt.mx")
 
@@ -35,7 +36,7 @@ class MatrixParser(HTMLParser):
     room_regex = re.compile("https://matrix.to/#/(#.+:.+)")
     block_tags = ("br", "p", "pre", "blockquote",
                   "ol", "ul", "li",
-                  "h1", "h2", "h3", "h4", "h5", "h6"
+                  "h1", "h2", "h3", "h4", "h5", "h6",
                   "div", "hr", "table")
 
     def __init__(self):
@@ -159,6 +160,14 @@ class MatrixParser(HTMLParser):
                 text = url
         elif previous_tag == "command":
             text = f"/{text}"
+
+        # Strikethrough
+        if "del" in self._open_tags:
+            text = html_to_unicode(text, "\u0336")
+        # Underline
+        if "u" in self._open_tags:
+            text = html_to_unicode(text, "\u0332")
+
         list_entry_handled_once = False
         # In order to maintain order of things like blockquotes in lists or lists in blockquotes,
         # we can't just have ifs/elses and we need to actually loop through the open tags in order.

mautrix_telegram/formatter/from_telegram.py

@@ -23,7 +23,7 @@ from mautrix_appservice import MatrixRequestError
 from .. import user as u, puppet as pu, portal as po
 from ..db import Message as DBMessage
 from .util import (add_surrogates, remove_surrogates, trim_reply_fallback_html,
-                   trim_reply_fallback_text)
+                   trim_reply_fallback_text, unicode_to_html)
 
 log = logging.getLogger("mau.fmt.tg")
 
@@ -138,6 +138,9 @@ async def telegram_to_matrix(evt, source, main_intent=None, is_edit=False):
         text += f"\n- {evt.post_author}"
         html += f"<br/><i>- <u>{evt.post_author}</u></i>"
 
+    html = unicode_to_html(text, html, "\u0336", "del")
+    html = unicode_to_html(text, html, "\u0332", "u")
+
     if html:
         html = html.replace("\n", "<br/>")
 

mautrix_telegram/formatter/util.py

@@ -1,3 +1,4 @@
+from html import escape
 import struct
 import re
 
@@ -31,3 +32,33 @@ HTML_REPLY_FALLBACK_REGEX = re.compile(r"^<blockquote data-mx-reply>[\s\S]+?</bl
 
 def trim_reply_fallback_html(html):
     return HTML_REPLY_FALLBACK_REGEX.sub("", html)
+
+
+def unicode_to_html(text, html, ctrl, tag):
+    if "\u0336" not in text and "\u0332" not in text:
+        return html
+    if not html:
+        html = escape(text)
+    tag_start = f"<{tag}>"
+    tag_end = f"</{tag}>"
+    characters = html.split(ctrl)
+    html = ""
+    in_del = False
+    for char in characters:
+        if not in_del:
+            if len(char) > 1:
+                html += char[0:-1]
+                char = char[-1]
+            html += tag_start
+            in_del = True
+            html += char
+        else:
+            if len(char) > 1:
+                html += tag_end
+                in_del = False
+            html += char
+    return html
+
+
+def html_to_unicode(text, ctrl):
+    return ctrl.join(text) + ctrl