0

I am working on a scraping script for python. I don't want to scrape non-English letters and special characters.

I am using this code to get rid of most symbols/characters/flags that I don't need:

 emoji_pattern = re.compile("["
                                                                u"\U0001F600-\U0001F64F"  # emoticons
                                                                u"\U0001F300-\U0001F5FF"  # symbols & pictographs
                                                                u"\U0001F680-\U0001F6FF"  # transport & map symbols
                                                                u"\U0001F1E0-\U0001F1FF"  # flags (iOS)
                                                                u"\U00002500-\U00002BEF"  # chinese char
                                                                u"\U00002702-\U000027B0"
                                                                u"\U00002702-\U000027B0"
                                                                u"\U000024C2-\U0001F251"
                                                                u"\U0001f926-\U0001f937"
                                                                u"\U00010000-\U0010ffff"
                                                                u"\u2640-\u2642" 
                                                                u"\u2600-\u2B55"
                                                                u"\u200d"
                                                                u"\u23cf"
                                                                u"\u23e9"
                                                                u"\u231a"
                                                                u"\ufe0f"  # dingbats
                                                                u"\u3030"
                                                           "]+", re.UNICODE)

Unforunately this code still ignores text like this:

vɒs səˈvɑːnt
meɪhər ʃælæl ˈhæʃ bɑːz
מַהֵר שָׁלָל חָשׁ בַּז
Mahēr šālāl ḥāš baz

How can I get rid of these as well?

Markus Ham
  • 31
  • 4
  • You could maybe do a whitelist instead of a blacklist. For example, start by [matching all ASCII characters](https://stackoverflow.com/questions/3203190/regex-any-ascii-character). – totok Jun 27 '22 at 08:45

1 Answers1

3

Does it filter enough?

import re


string = '''English text? vɒs səˈvɑːnt

\U0001F600 \U0001F64F
meɪhər ʃælæl ˈhæʃ bɑːz
מַהֵר שָׁלָל חָשׁ בַּז
Mahēr šālāl ḥāš baz'''


print(re.sub('[^\sA-Za-z0-9.!?\\-]+','', string))

Output:

English text? vs svnt

 
mehr ll h bz
   
Mahr ll  baz

I was not sure if you need punctuation. If not - use this pattern [^\sA-Za-z0-9]

olegr
  • 1,999
  • 18
  • 23