61 lines
2.2 KiB
Python
61 lines
2.2 KiB
Python
import datetime
|
|
import decimal
|
|
import functools
|
|
import re
|
|
import unicodedata
|
|
|
|
import babel.numbers
|
|
|
|
CURRENCY_SPEC_PATTERN = r'^{space}(?:|{symbol}{space}{code}|{code}{space}{symbol}){space}$'.format(
|
|
code=r'[A-Za-z]{,3}',
|
|
space=r'\s*',
|
|
symbol=r'(\W?)',
|
|
)
|
|
|
|
@functools.lru_cache()
|
|
def _currency_amount_pattern(locale):
|
|
minus = babel.numbers.get_minus_sign_symbol(locale)
|
|
plus = babel.numbers.get_plus_sign_symbol(locale)
|
|
dec_sym = babel.numbers.get_decimal_symbol(locale)
|
|
sep_sym = '.' if dec_sym == ',' else ','
|
|
return r'([{}{}]?)\s*(\W?)\s*(\d+(?:{}\d+)*(?:{}\d*)?)'.format(
|
|
minus,
|
|
plus,
|
|
re.escape(sep_sym),
|
|
re.escape(dec_sym),
|
|
)
|
|
|
|
def currency_decimal(s, locale='en_US_POSIX'):
|
|
try:
|
|
match = re.search(_currency_amount_pattern(locale), s)
|
|
except TypeError:
|
|
return decimal.Decimal(s)
|
|
if not match:
|
|
raise ValueError("no decimal found in {!r}".format(s))
|
|
extra_s = ''.join([s[:match.start()], match.group(2), s[match.end():]])
|
|
# The only extra text allowed is currency specifiers:
|
|
# '€', 'A$', 'US$', 'CAD', '$USD', etc.
|
|
extra_match = re.match(CURRENCY_SPEC_PATTERN, extra_s)
|
|
if not extra_match:
|
|
extra_ok = False
|
|
else:
|
|
symbol = extra_match.group(1) or extra_match.group(2)
|
|
extra_ok = (not symbol) or (unicodedata.category(symbol) == 'Sc')
|
|
if not extra_ok:
|
|
raise ValueError("non-currency text in {!r}: {!r}".format(s, extra_s))
|
|
return babel.numbers.parse_decimal(match.group(1) + match.group(3), locale)
|
|
|
|
def date(date_s, date_fmt):
|
|
return datetime.datetime.strptime(date_s, date_fmt).date()
|
|
|
|
def _rejoin_slice_words(method_name, source, wordslice, sep=None, limit=-1, joiner=None):
|
|
if joiner is None:
|
|
joiner = ' ' if sep is None else sep
|
|
return joiner.join(_slice_words(method_name, source, wordslice, sep, limit))
|
|
rejoin_slice_words = functools.partial(_rejoin_slice_words, 'split')
|
|
rejoin_rslice_words = functools.partial(_rejoin_slice_words, 'rsplit')
|
|
|
|
def _slice_words(method_name, source, wordslice, sep=None, limit=-1):
|
|
return getattr(source, method_name)(sep, limit)[wordslice]
|
|
slice_words = functools.partial(_slice_words, 'split')
|
|
rslice_words = functools.partial(_slice_words, 'rsplit')
|