benry 0.1.0

rx()

benry.rexp.rx() is a short cut to re.compile().

from benry.rexp import rx

## comping -- same as re.compile(r'pat') or re.compile(r'pat', rx.I|rx.S)
regexp = rx(r'pat')
regexp = rx(r'pat', rx.I|rx.S)

## matching -- same as re.compile(r'pat').search(string)
m = rx(r'pat').search(string)

## replacing -- same as re.compile(r'pat').sub('repl', string, count=1)
rx(r'pat').sub('repl', string, count=1)

You don’t need to use re.xxxx() functions because rx().xxxx() does same things, and has more features.

## For example you can't specify starting/ending position with re.match().
re.match(r'pat', string, re.I)

## But you can specify them with rx().match().
rx(r'pat', re.I).match(string, start_pos, end_pos)

rx.compile()

rx.compile() (or benry.rexp.compile()) is similar to re.compile(), but the former doesn’t cache compiled pattern while the latter caches it.

This is very useful when there are a lot of regexp pattern and they are no need to cache into library.

mappings = [
    (r'^/posts$',                      'app.PostsPage'),
    (r'^/posts/new$',                  'app.PostCreatePage'),
    (r'^/posts/(?P<id>\d+)$',          'app.PostPage'),
    (r'^/posts/(?P<id>\d+)/edit$',     'app.PostUpdatePage'),
    (r'^/posts/(?P<id>\d+)/comments$', 'app.PostCommentsPage'),
]

## no need to cache patterns, so calls rx.compile() instead of re.compile()
from benry.rexp import rx   # or: import compile
compiled_mappings = [ (rx.compile(pat), cls) for pat, cls in mappings ]

rx.matching()

rx.matching() (or benry.rexp.matching()) is a utility class to help you when matching to a lot of patterns.

Without rx.matching():

import re

m = re.match(r'^(\d\d\d\d)-(\d\d)-(\d\d)$', string)
if m:
    Y, M, D = m.groups()
else:
    m = re.match(r'^(\d\d)/(\d\d)/(\d\d\d\d)$', string)
    if m:
        M, D, Y = m.groups()
    else:
        m = re.match(r'^(\d\d\d\d)/(\d\d)/(\d\d)$', string)
        if m:
            Y, M, D = m.groups()

With rx.matching():

from benry.rexp import rx

m = rx.matching(string)
if   m.match(r'^(\d\d\d\d)-(\d\d)-(\d\d)$'):
    Y, M, D = m.groups()     # or Y, M, D = m
elif m.match(r'^(\d\d)/(\d\d)/(\d\d\d\d)$'):
    M, D, Y = m.groups()     # or M, D, Y = m
elif m.match(r'^(\d\d\d\d)/(\d\d)/(\d\d)$'):
    Y, M, D = m.groups()     # or Y, M, D = m

You can get SRE_Match object by m.matched.

m = rx.matching(string)
if m.match(r'pat'):
    print(m.matched.pos)
    print(m.matched.endpos)

type

rx.type represents class of regular expression.

import re
from benry.rexp import rx

assert rx.type is type(re.compile('x'))

class UTCDateTime

UTCDdateTime is a subclass of datetime.datetime representing UTC offset.

from benry.date_time import UTCDateTime

print(UTCDateTime.offset_minutes)      #=> 0
print(UTCDateTime.offset_timedelta)    #=> timedelta(seconds=0)
print(UTCDateTime.is_utc)              #=> True
print(UTCDateTime.is_local)            #=> False

## almost same as datetime.utcnow(), except returning UTCDateTime object.
utc_dt = UTCDateTime.now()

print(utc_dt.to_utc())                 # returns self.
print(utc_dt.to_local())               # returns LocalDateTime object.

class LocalDateTime

UTCDdateTime is a subclass of datetime.datetime representing local time. This class calculates offset between local time and UTC time.

from benry.date_time import LocalDateTime

print(LocalDateTime.offset_minutes)    #=> 9*60  (ex: JST timezone)
print(LocalDateTime.offset_timedelta)  #=> timedelta(seconds=9*60*60)
print(LocalDateTime.is_utc)            #=> False
print(LocalDateTime.is_local)          #=> True

## almost same as datetime.now(), except returning LocalDateTime object.
local_dt = LocalDateTime.now()

print(local_dt.to_utc())               # returns UTCDateTime object.
print(local_dt.to_local())             # returns self.