fluquid / find_job_titles Goto Github PK

Using the API,

1. some titles are in title case and some of the titles like .NET are in upper case so we can not change input case into any particular case.

2. If the file is accessible then we can add some new titles if we want.

Please you should add this if possible

I always receive an StopIteration Error, when there's no match.
To stop the program from throwing StopIteration and instead continue the program, update these lines of code:

File find_job_titles\__init__.py
FROM

def longest_match(matches):
    """

    find respective longest matches from all overlapping aho corasick matches

    """

    longest = next(matches)

    if longest is None:

        return

TO

def longest_match(matches):

    """

    find respective longest matches from all overlapping aho corasick matches

    """

    longest = next(matches, 0)

    if longest is None or longest == 0:

        return

In analytics/analytic_service.py on line 15: TODO fix if datetine None use current time

If a new timeseries entry is created without specifying a timestamp, the timestamp should default to datetime.utcnow

finder=Finder()
Traceback (most recent call last):
File "", line 1, in
File "/Users/fanglinchen/anaconda2/envs/py2Env/lib/python2.7/site-packages/find_job_titles/init.py", line 145, in init
autom.add_word(title, title)
TypeError: string1 required

I can use the other constructor without a problem though!

Cannot be used to detect any intern/internship position.

from find_job_titles import FinderPyaho
finder=FinderPyaho()
finder.findall('I am the retailer')
[Match(start=11, end=16, match='tailer')]

Note: It is fetching tailor from 'retailer'
finder.findall('SBBSBPresidentAAAAA')
[Match(start=5, end=13, match='President')]
Please have a look. It should fetch the complete word not sub-string.