Trafaret
Ultimate transformation library that supports validation, contexts and aiohttp
.
Trafaret is rigid and powerful lib to work with foreign data, configs etc. It provides simple way to check anything, and convert it accordingly to your needs.
It has shortcut syntax and ability to express anything that you can code:
>>> from trafaret.constructor import construct >>> validator = construct({'a': int, 'b': [str]}) >>> validator({'a': 5, 'b': ['lorem', 'ipsum']}) {'a': 5, 'b': ['lorem', 'ipsum']} >>> validator({'a': 5, 'b': ['gorky', 9]}) Traceback (most recent call last): File "" , line 1, in <module> File "/Users/mkrivushin/w/trafaret/trafaret/__init__.py", line 204, in __call__ return self.check(val) File "/Users/mkrivushin/w/trafaret/trafaret/__init__.py", line 144, in check return self._convert(self.check_and_return(value)) File "/Users/mkrivushin/w/trafaret/trafaret/__init__.py", line 1105, in check_and_return raise DataError(error=errors, trafaret=self) trafaret.DataError: {'b': DataError({1: DataError(value is not a string)})}
Read The Docs hosted documentation http://trafaret.readthedocs.org/en/latest/ or look to the docs/intro.rst for start.
Trafaret can even generate Trafarets instances to build transformators from json, like in json schema implementation Trafaret Schema
New
2.0.2
- construct for int and float will use ToInt and ToFloat
2.0.0
WithRepr
– use it to return custom representation, like- Strip a lot from dict, like
keys()
- Trafarets are not mutable
- DataError has new
code
attribute, self.failure hascode
argument - OnError has
code
argument too - New
DataError.to_struct
method that returns errors in more consistent way String
,AnyString
,Bytes
,FromBytes(encoding=utf-8)
Int
,ToInt
,Float
,ToFloat
ToDecimal
Iterable
that acts like aList
, but works with any iterable- New
Date
,ToDate
andDateTime
,ToDateTime
trafarets StrBool
trafaret renamed toToBool
Visitor
trafaret was deleted- Test coverage
1.x.x
- converters and
convert=False
are deleted in favor ofAnd
and&
String
parameterregex
deleted in favor ofRegexp
andRegexpRaw
usage- new
OnError
to customize error message context=something
argument for__call__
andcheck
Trafaret methods. Supported byOr
,And
,Forward
etc.- new customizable method
transform
likechange_and_return
but takescontext=
arg - new
trafaret_instance.async_check
method that works withawait
Doc
For simple example what can be done:
import datetime
import trafaret as t
date = t.Dict(year=t.Int, month=t.Int, day=t.Int) >> (lambda d: datetime.datetime(**d))
assert date.check({'year': 2012, 'month': 1, 'day': 12}) == datetime.datetime(2012, 1, 12)
Work with regex:
>>> c = t.RegexpRaw(r'^name=(\w+)$') >> (lambda m: m.group(1))
>>> c.check('name=Jeff')
'Jeff'
Rename dict keys:
>>> c = t.Dict({(t.Key('uNJ') >> 'user_name'): t.String})
>>> c.check({'uNJ': 'Adam'})
{'user_name': 'Adam'}
Arrow
date checking:
import arrow
def check_datetime(str):
try:
return arrow.get(str).naive
except arrow.parser.ParserError:
return t.DataError('value is not in proper date/time format')
Yes, you can write trafarets that simple.