I have such strings:
ye<V><vn:inf><N><0><V><cpl:pres><3s>
çok<Postp:adv:ablC><0><N><0><V><cpl:pres><3s>
yağ<N><li><Adv><0><N><0><V><cpl:evid><3s>
And I want to extract;
ye, V, 3s
çok, Postp:adv:ablC, 3s
yağ, N, 3s
I have hundreds millions of such strings. What can be the best, efficient, and fastest way to do it? Can you show an example?
Thanks,
Try this:
l = s.split('<')
'{}, {}, {}'.format(l[0], l[1][:-1], l[-1][:-1])
Example of output:
>>> s = 'ye<V><vn:inf><N><0><V><cpl:pres><3s>'
>>> l = s.split('<')
>>> '{}, {}, {}'.format(l[0], l[1][:-1], l[-1][:-1])
'ye, V, 3s'
format instead of string concatenation with +You could try using using the findall. For example,
import re
regex = re.compile(r'(?P<g1>3s)|(?P<g2>ye)')
regex.findall(test_string)
This will return a list of tuples for the matches like the following:
# Output
# [('3s', ''), ('', 'ye'), ('3s', ''), ('', 'ye')]
The regular expression that I compiled does not have all of the named groups that you desire, but you can add those easily enough.
s1 = 'ye<V><vn:inf><N><0><V><cpl:pres><3s>'
s2 = 'çok<Postp:adv:ablC><0><N><0><V><cpl:pres><3s>'
s3 = 'yağ<N><li><Adv><0><N><0><V><cpl:evid><3s>'
if __name__ == '__main__':
for s in (s1,s2,s3):
print('{0}, {1}, {2}'.format(s.split('<')[0], s.split('<')[1].split('>')[0], s.split('<')[-1].split('>')[0]))
