How to extract a float from a string

If your float is always expressed in decimal notation something like

>>> import re
>>> re.findall("\d+\.\d+", "2017.5 is halfway through the year")
['2017.5']

may suffice.

For parse int and float (point separator) values:

re.findall( r'\d+\.*\d*', '2017.5 is halfway through the year' )

result:

['2017.5']

here is an example using re.match:

>>> import re
>>> ok = '2017.5 and some stuff'
>>> hmm = re.match(r'[0-9\.]+', ok)
>>> hmm.group()
'2017.5'

r'[0-9\.]+' is a regex to extract the group that matches numbers and periods

just a dirty implementation without regex and with the single digit/float per sentence:

s = "new 2017.5 is halfway through the year1"

def process(all,new):

    def is_float_try(str):
        try:
            float(str)
            return True
        except ValueError:
            return False

    return new if new.isdigit() or is_float_try(new) else all

print reduce(process,s.split(),'')
# 2017.5