hyperboria/nexus/nlptools/utils.py
the-superpirate d51e5ab65d - fix: Various fixes for release
- fix: Translation fixes
- fix: Various fixes
- feat: PB translations, configuration changes
- fix: Bugfixes

GitOrigin-RevId: 55f8b148c42a296162fc707c36a5146ca0073b4b
2021-01-29 11:26:51 +03:00

110 lines
2.7 KiB
Python

import re
import struct
import unicodedata
from .regex import (
EMAIL_REGEX,
EMOJI_REGEX,
HASHTAG_REGEX,
MULTIWHITESPACE_REGEX,
NON_ALNUMWHITESPACE_REGEX,
TELEGRAM_LINK_REGEX,
URL_REGEX,
)
def add_surrogate(text):
return ''.join(
# SMP -> Surrogate Pairs (Telegram offsets are calculated with these).
# See https://en.wikipedia.org/wiki/Plane_(Unicode)#Overview for more.
''.join(chr(y) for y in struct.unpack('<HH', x.encode('utf-16le')))
if (0x10000 <= ord(x) <= 0x10FFFF) else x for x in text
)
def cast_string_to_single_string(s):
processed = MULTIWHITESPACE_REGEX.sub(' ', NON_ALNUMWHITESPACE_REGEX.sub(' ', s))
processed = processed.strip().replace(' ', '-')
return processed
def clean_text(text):
text = remove_markdown(remove_emoji(text))
text = remove_url(text)
text = despace_smart(text)
return text.strip()
def despace(text):
text = re.sub(r'\n+', '\n', text)
text = re.sub(r'[ \t]+', ' ', text)
text = re.sub(r'\n[ \t]+', '\n', text)
return text
def despace_full(text):
return re.sub(r'\s+', ' ', text).strip()
def despace_smart(text):
text = re.sub(r'\n\s*[-•]+\s*', r'\n', text)
text = re.sub(r'\n{2,}', r'\n', text).strip()
text = re.sub(r'\.?(\s+)?\n', r'. ', text)
text = re.sub(r'\s+', ' ', text)
return text
def escape_format(text):
text = text.replace("__", "_").replace("**", "*").replace("`", "'")
text = text.replace('[', r'`[`').replace(']', r'`]`')
return text
def remove_markdown(text):
text = re.sub('[*_~]{2,}', '', text)
text = re.sub('[`]+', '', text)
text = re.sub(r'\[\s*(.*?)(\s*)\]\(.*?\)', r'\g<1>\g<2>', text, flags=re.MULTILINE)
return text
def normalize_string(string):
string = re.sub('[^a-zA-Z0-9_\\-]+', '', string.lower().strip().replace(' ', '-'))
return unicodedata.normalize('NFKD', string).encode('ascii', 'ignore').decode('utf-8')
def remove_emails(text):
return re.sub(EMAIL_REGEX, '', text)
def remove_emoji(text):
text = re.sub(EMOJI_REGEX, '', text)
text = re.sub(u'\ufe0f', '', text)
return text
def remove_hashtags(text):
return re.sub(HASHTAG_REGEX, '', text)
def remove_url(text):
return re.sub(URL_REGEX, '', text)
def replace_telegram_link(text):
return re.sub(TELEGRAM_LINK_REGEX, r'@\1', text)
def split_at(s, pos):
if len(s) < pos:
return s
pos -= 10
pos = max(0, pos)
for p in range(pos, min(pos + 20, len(s) - 1)):
if s[p] in [' ', '\n', '.', ',', ':', ';', '-']:
return s[:p] + '...'
return s[:pos] + '...'
def unwind_hashtags(text):
return re.sub(HASHTAG_REGEX, r'\2', text)