I want to delete certain words from a paragraph, such as "and", "as", and "like". Is there an easier way to delete words from a string than doing it via replace --
new_str = str.replace(' and ', '').replace(' as ', '').replace(' like ', '')
For example, is there a method similar to the following?
str.remove([' and ', ' like ', ' as '])
Yes, you could use the sub function from the re module:
>>> import re
>>> s = 'I like this as much as that'
>>> re.sub('and|as|like', '', s)
'I this much that'
str.replace is known to be oodles faster than re.sub.)
replace, so I was forced to look for another (worse performance) solutionYou could use regular expressions:
>>> import re
>>> test = "I like many words but replace some occasionally"
>>> to_substitute = "many|words|occasionally"
>>> re.sub(to_substitute, '', test)
'I like but replace some '
You may also do without regex. See the following example
def StringRemove(st,lst):
return ' '.join(x for x in st.split(' ') if x not in lst)
>>> StringRemove("Python string Java is immutable, unlike C or C++ that would give you a performance benefit. So you can't change them in-place",['like', 'as', 'and'])
"Python string Java is immutable, unlike C or C++ that would give you a performance benefit. So you can't change them in-place"
>>> st="Python string Java is immutable, unlike C or C++ that would give you a performance benefit. So you can't change them in-place"
>>> StringRemove(st,['like', 'as', 'and'])==st
True
>>>
\r, \n and \t into space as well. If you care about spaces, use st.split(' ') instead of st.split(). Also, the square brackets around the join() body aren't neat. I'd scrap them and make it a generator expression (which for larger inputs will use less memory, also) instead of a list comprehension.
Note that if all you care about is readability and not necessarily performance, you could do something like this:
new_str = str
for word_to_remove in [' and ', ' as ', ' like ']:
new_str = new_str.replace(word_to_remove, '')
