StringMethods should have the same methods as standard str

[sinhrks]

Should StringMethods be compatible with standard str, or have only useful methods? Sometimes I have to adjust text formats before converting other types, and I've chosen additionals which makes formatting / adjustments easier. Any ideas?

• str.find (ENH: Add StringMethod.find and rfind #9386)
• str.rfind (ENH: Add StringMethod.find and rfind #9386)
• str.isalnum (ENH: StringMethods supports is_xxx methods #9282)
• str.isalpha (ENH: StringMethods supports is_xxx methods #9282)
• str.isdigit (ENH: StringMethods supports is_xxx methods #9282)
• str.isspace (ENH: StringMethods supports is_xxx methods #9282)
• str.islower (ENH: StringMethods supports is_xxx methods #9282)
• str.isupper (ENH: StringMethods supports is_xxx methods #9282)
• str.istitle (ENH: StringMethods supports is_xxx methods #9282)
• str.ljust (ENH: StringMethods now supports ljust and rjust #9352)
• str.rjust (ENH: StringMethods now supports ljust and rjust #9352): StringMethods.pad works as the same as ljust, rjust and center by switching side option. And StringMethods.center exists, ljust and rjust don't.
• str.rsplit: split also supports regex, and there seems no straightforward way to rsplit using regex. (ENH: added rsplit to StringMethods #10303)
• str.zfill (ENH: Add StringMethods.zfill #9387)
• StringMethods.center and StringMethods.pad should have 'fillchar` kw to specify padding character. (ENH: StringMethods now supports ljust and rjust #9352)
• str.capitalize (ENH: Add StringMethods.capitalize and swapcase #9766)
• str.swapcase (ENH: Add StringMethods.capitalize and swapcase #9766)
• str.translate (ENH: support str translate for StringMethods #10052)
• str.index (ENH: support StringMethods index and rindex #10045)
• str.rindex (ENH: support StringMethods index and rindex #10045)
• str.partition (ENH: Add StringMethods.partition and rpartition #9773)
• str.rpartition (ENH: Add StringMethods.partition and rpartition #9773)
• unicode.isnumeric (ENH: Add StringMethods.isnumeric and isdecimal #9439)
• unicode.isdecimal (ENH: Add StringMethods.isnumeric and isdecimal #9439)
• unicodedata.normalize (ENH: Added str.normalize to use unicodedata.normalize #10031)

NOTE For python 2.x, once convert target to unicode internally then call the func.

Which may not be very userful:

• expandtabs
• splitlines

I think there is an option to implement these to say "support the same methods as standard str".

[sinhrks]

Added unicode.isnumeric and unicode.isdecimal to the list.

• https://docs.python.org/2/library/stdtypes.html#unicode.isnumeric

[sinhrks]

Would like to discuss an additional:

• Unicode normalization

  unicodedata.normalize is quite useful to standardize multi-bytes characters. I think it is nice if StringMethods.normalize can perform this.

  import pandas as pd
  s = pd.Series([u'ＡＢＣＤＥ', u'１２３４５'])
  s
  # 0    ＡＢＣＤＥ
  # 1    １２３４５
  # dtype: object
  
  s.str.normalize()
  # 0    ABCDE
  # 1    12345
  # dtype: object
  

[sinhrks]

The last remaining is str.rsplit. Even though split supports both str and regex, I think there is no easy way (and use case) to rsplit with regex. How about implement str.rsplit only for str and closig this.

CC @mortada

[mortada]

@sinhrks that sounds reasonable. I'm happy to do a PR for this.

[jreback]

looks like this is finished!

awesome job @sinhrks, cc @mortada

[sinhrks]

Thanks, @mortada !