LonamiWebs/Telethon
1
1
Fork 0
mirror of https://github.com/LonamiWebs/Telethon.git synced 2025-01-24 08:14:14 +03:00
Code Issues Packages Projects Releases Wiki Activity

Clean up markdown parsing since tuples aren't used anymore

Browse Source
This commit is contained in:
Lonami Exo 2017-11-10 11:41:49 +01:00
parent 7d75eebdab
commit cb3f20db65
1 changed files with 36 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

68
telethon/extensions/markdown.py
View File

@ -48,12 +48,12 @@ def parse(message, delimiters=None, url_re=None):
i = 0 i = 0
result = [] result = []
current = None current = None
end_delimiter = None
# Work on byte level with the utf-16le encoding to get the offsets right. # Work on byte level with the utf-16le encoding to get the offsets right.
# The offset will just be half the index we're at. # The offset will just be half the index we're at.
message = message.encode('utf-16le') message = message.encode('utf-16le')
while i < len(message): while i < len(message):
url_match = None
if url_re and current is None: if url_re and current is None:
# If we're not inside a previous match since Telegram doesn't allow # If we're not inside a previous match since Telegram doesn't allow
# nested message entities, try matching the URL from the i'th pos. # nested message entities, try matching the URL from the i'th pos.
@ -70,42 +70,46 @@ def parse(message, delimiters=None, url_re=None):
offset=i // 2, length=len(url_match.group(1)) // 2, offset=i // 2, length=len(url_match.group(1)) // 2,
url=url_match.group(2).decode('utf-16le') url=url_match.group(2).decode('utf-16le')
)) ))
# We matched the delimiter which is now gone, and we'll add i += len(url_match.group(1))
# +2 before next iteration which will make us skip a character. # Next loop iteration, don't check delimiters, since
# Go back by one utf-16 encoded character (-2) to avoid it. # a new inline URL might be right after this one.
i += len(url_match.group(1)) - 2 continue
if not url_match: if end_delimiter is None:
# We're not expecting any delimiter, so check them all
for d, m in delimiters.items(): for d, m in delimiters.items():
# Slice the string at the current i'th position to see if # Slice the string at the current i'th position to see if
# it matches the current delimiter d. # it matches the current delimiter d, otherwise skip it.
if message[i:i + len(d)] == d: if message[i:i + len(d)] != d:
if current is not None and not isinstance(current, m): continue
# We were inside another delimiter/mode, ignore this.
continue
if message[i + len(d):i + 2 * len(d)] == d: if message[i + len(d):i + 2 * len(d)] == d:
# The same delimiter can't be right afterwards, if # The same delimiter can't be right afterwards, if
# this were the case we would match empty strings # this were the case we would match empty strings
# like `` which we don't want to. # like `` which we don't want to.
continue continue
# Get rid of the delimiter by slicing it away # Get rid of the delimiter by slicing it away
message = message[:i] + message[i + len(d):] message = message[:i] + message[i + len(d):]
if current is None: if m == MessageEntityPre:
if m == MessageEntityPre: # Special case, also has 'lang'
# Special case, also has 'lang' current = m(i // 2, None, '')
current = MessageEntityPre(i // 2, None, '') else:
else: current = m(i // 2, None)
current = m(i // 2, None)
# No need to i -= 2 here because it's been already end_delimiter = d # We expect the same delimiter.
# checked that next character won't be a delimiter. break
else:
current.length = (i // 2) - current.offset elif message[i:i + len(end_delimiter)] == end_delimiter:
result.append(current) message = message[:i] + message[i + len(end_delimiter):]
current = None current.length = (i // 2) - current.offset
i -= 2 # Delimiter matched and gone, go back 1 char result.append(current)
break current, end_delimiter = None, None
# Don't increment i here as we matched a delimiter,
# and there may be a new one right after. This is
# different than when encountering the first delimiter,
# as we already know there won't be the same right after.
continue
# Next iteration, utf-16 encoded characters need 2 bytes. # Next iteration, utf-16 encoded characters need 2 bytes.
i += 2 i += 2