Skip to content

Add support for processing language parameter in message entities #2466

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Badiboy merged 3 commits into from
Apr 6, 2025
Merged

Add support for processing language parameter in message entities #2466

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
19bf138
Add support for processing language parameter in message entities
pplulee Apr 5, 2025
e270371
Add test for HTML parsing
pplulee Apr 6, 2025
d25dae8
Fix special character conversion
pplulee Apr 6, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 9 additions & 4 deletions telebot/formatting.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -418,7 +418,7 @@ def apply_html_entities(text: str, entities: Optional[List], custom_subs: Option
utf16_text = text.encode("utf-16-le")
html_text = ""

def func(upd_text, subst_type=None, url=None, user=None, custom_emoji_id=None):
def func(upd_text, subst_type=None, url=None, user=None, custom_emoji_id=None, language=None):
upd_text = upd_text.decode("utf-16-le")
if subst_type == "text_mention":
subst_type = "text_link"
Expand All @@ -431,6 +431,8 @@ def func(upd_text, subst_type=None, url=None, user=None, custom_emoji_id=None):
subs = _subs.get(subst_type)
if subst_type == "custom_emoji":
return subs.format(text=upd_text, custom_emoji_id=custom_emoji_id)
elif (subst_type == "pre") and language:
return "<pre><code class=\"language-{0}\">{1}</code></pre>".format(language, upd_text)
return subs.format(text=upd_text, url=url)

offset = 0
Expand All @@ -444,14 +446,16 @@ def func(upd_text, subst_type=None, url=None, user=None, custom_emoji_id=None):
offset = entity.offset

new_string = func(utf16_text[offset * 2: (offset + entity.length) * 2], subst_type=entity.type,
url=entity.url, user=entity.user, custom_emoji_id=entity.custom_emoji_id)
url=entity.url, user=entity.user, custom_emoji_id=entity.custom_emoji_id,
language=entity.language)
start_index = len(html_text)
html_text += new_string
offset += entity.length
end_index = len(html_text)
elif entity.offset == offset:
new_string = func(utf16_text[offset * 2: (offset + entity.length) * 2], subst_type=entity.type,
url=entity.url, user=entity.user, custom_emoji_id=entity.custom_emoji_id)
url=entity.url, user=entity.user, custom_emoji_id=entity.custom_emoji_id,
language=entity.language)
start_index = len(html_text)
html_text += new_string
end_index = len(html_text)
Expand All @@ -463,7 +467,8 @@ def func(upd_text, subst_type=None, url=None, user=None, custom_emoji_id=None):
# And we don't change it).
entity_string = html_text[start_index: end_index].encode("utf-16-le")
formatted_string = func(entity_string, subst_type=entity.type, url=entity.url, user=entity.user,
custom_emoji_id=entity.custom_emoji_id). \
custom_emoji_id=entity.custom_emoji_id,
language=entity.language). \
replace("&amp;", "&").replace("&lt;", "<").replace("&gt;", ">")
html_text = html_text[:start_index] + formatted_string + html_text[end_index:]
end_index = len(html_text)
Expand Down
81 changes: 51 additions & 30 deletions tests/test_telebot.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,11 +44,11 @@ def test_message_listener(self):
def listener(messages):
assert len(messages) == 100

tb = telebot.TeleBot('')
tb = telebot.TeleBot(TOKEN)
tb.set_update_listener(listener)

def test_message_handler(self):
tb = telebot.TeleBot('')
tb = telebot.TeleBot(TOKEN)
msg = self.create_text_message('/help')

@tb.message_handler(commands=['help', 'start'])
Expand All @@ -60,7 +60,7 @@ def command_handler(message):
assert msg.text == 'got'

def test_message_handler_reg(self):
bot = telebot.TeleBot('')
bot = telebot.TeleBot(TOKEN)
msg = self.create_text_message(r'https://web.telegram.org/')

# noinspection PyUnusedLocal
Expand All @@ -73,7 +73,7 @@ def command_url(message):
assert msg.text == 'got'

def test_message_handler_lambda(self):
bot = telebot.TeleBot('')
bot = telebot.TeleBot(TOKEN)
msg = self.create_text_message(r'lambda_text')

# noinspection PyUnusedLocal
Expand All @@ -86,7 +86,7 @@ def command_url(message):
assert msg.text == 'got'

def test_message_handler_lambda_fail(self):
bot = telebot.TeleBot('')
bot = telebot.TeleBot(TOKEN)
msg = self.create_text_message(r'text')

# noinspection PyUnusedLocal
Expand All @@ -99,7 +99,7 @@ def command_url(message):
assert not msg.text == 'got'

def test_message_handler_reg_fail(self):
bot = telebot.TeleBot('')
bot = telebot.TeleBot(TOKEN)
msg = self.create_text_message(r'web.telegram.org/')

# noinspection PyUnusedLocal
Expand Down Expand Up @@ -147,6 +147,7 @@ def test_send_file_with_filename(self):
ret_msg = tb.send_document(CHAT_ID, file_data)
assert ret_msg.message_id

file_data.seek(0)
ret_msg = tb.send_document(CHAT_ID, file_data, visible_file_name="test.jpg")
assert ret_msg.message_id

Expand Down Expand Up @@ -529,28 +530,31 @@ def create_message_update(text):
params = {'text': text}
chat = types.User(11, False, 'test')
message = types.Message(1, None, None, chat, 'text', params, "")
edited_message = None
channel_post = None
edited_channel_post = None
inline_query = None
chosen_inline_result = None
callback_query = None
shipping_query = None
pre_checkout_query = None
poll = None
poll_answer = None
my_chat_member = None
chat_member = None
chat_join_request = None
message_reaction = None
message_reaction_count = None
chat_boost = None
chat_boost_removed = None
purchased_paid_media = None
return types.Update(-1001234038283, message, edited_message, channel_post, edited_channel_post, inline_query,
chosen_inline_result, callback_query, shipping_query, pre_checkout_query, poll, poll_answer,
my_chat_member, chat_member, chat_join_request, message_reaction, message_reaction_count, chat_boost, chat_boost_removed,
purchased_paid_media)
return types.Update(
update_id=-1001234038283,
message=message,
edited_message=None,
channel_post=None,
edited_channel_post=None,
inline_query=None,
chosen_inline_result=None,
callback_query=None,
shipping_query=None,
pre_checkout_query=None,
poll=None,
poll_answer=None,
my_chat_member=None,
chat_member=None,
chat_join_request=None,
message_reaction=None,
message_reaction_count=None,
chat_boost=None,
removed_chat_boost=None,
purchased_paid_media=None,
business_message=None,
business_connection=None,
edited_business_message=None,
deleted_business_messages=None, )

def test_is_string_unicode(self):
s1 = u'string'
Expand Down Expand Up @@ -657,7 +661,7 @@ def test_typed_middleware_handler(self):

apihelper.ENABLE_MIDDLEWARE = True

tb = telebot.TeleBot('')
tb = telebot.TeleBot(TOKEN)
update = self.create_message_update('/help')

# noinspection PyUnusedLocal
Expand All @@ -678,7 +682,7 @@ def test_default_middleware_handler(self):

apihelper.ENABLE_MIDDLEWARE = True

tb = telebot.TeleBot('')
tb = telebot.TeleBot(TOKEN)
update = self.create_message_update('/help')

# noinspection PyUnusedLocal
Expand All @@ -703,3 +707,20 @@ def test_chat_permissions(self):
#tb = telebot.TeleBot(TOKEN)
#permissions = types.ChatPermissions(can_send_messages=True, can_send_polls=False)
#msg = tb.set_chat_permissions(CHAT_ID, permissions)

def test_apply_html_entities(self):
text = {
"*bold*": "<b>bold</b>",
"__italic__": "<u>italic</u>",
"~strikethrough~": "<s>strikethrough</s>",
"`inline code`": "<code>inline code</code>",
"```\ncode block```": "<pre>code block</pre>",
"```python\nprint('Hello, world!')\n```": "<pre><code class=\"language-python\">print('Hello, world!')</code></pre>",
"```python\nprint(1 < 2)\n```": "<pre><code class=\"language-python\">print(1 &lt; 2)</code></pre>",
"[link](http://example.com/)": "<a href=\"http://example.com/\">link</a>",
">blockquote": "<blockquote>blockquote</blockquote>",
}
tb = telebot.TeleBot(TOKEN)
for key, value in text.items():
ret_msg = tb.send_message(CHAT_ID, text=key, parse_mode='MarkdownV2')
assert telebot.formatting.apply_html_entities(ret_msg.text, ret_msg.entities, None) == value