PyContracts is a Python package that allows to declare constraints on function parameters and return values. Contracts can be specified using Python3 annotations, or inside a docstring. PyContracts supports a basic type system, variables binding, arithmetic constraints, and has several specialized contracts and an extension API.
Home Page: http://andreacensi.github.io/contracts/
License: Other
This small example should be all that is needed to explain the problem.
from contracts import contract
@contract(url=str, returns=int)
def validate_url(url=""):
""" returns True if the input is a valid url """
return False
print validate_url("google.com")
# FalseI also noticed, and maybe I am wrong, that I cannot easily apply a custom validation function over all objects of a collection, be they a dict or list. This may be something that others will find useful as well. For example, maybe I want to validate that all items in a list are only alpha-numeric, have at most 2 numerals and say are no longer than 30 chars.
Anyway, sounds esoteric, but I have use for this today.
Hello. I propose the following syntax to implement
@contract(arg="notNone")
def function(arg):
pass
Action: perform the following python code
assert(arg is not None)
Hi Andrea!
I only quickly skimmed through the docs, so I might missed something, but it seems that it's not possible to (out of the box) declare a contract on the function that takes function as an argument and returns another function. I'm dealing with a lot of functional-style code lately and I thought that this could be very handy in certain situations. I'm guessing it's possible to add support for this via custom contracts easily (haven't tried yet, but I'm going to) and what I'd like to know is if has a chance of becoming built in if implemented. (For example of this feature check here: http://docs.racket-lang.org/guide/contract-func.html#(part._.Contracts_on_.Higher-order_.Functions) )
Also, in some contracts library for CoffeeScript (which wasn't actually working IIRC :)) I saw an idea of embedding object's invariants into its methods' contracts. This should be easy enough, given the fact that self is explicitly passed in Python, but it would require having a syntax for making contracts for objects' attributes, which I didn't see built in either? I think I'll be exploring some possibilities in the course of next few weeks, I just wanted to know what do you, as an author and maintainer, think about these ideas.
Best regards,
Piotr Klibert
You are now at version 1.7.12, but changelog.rst only goes up to 1.7.7. The docs on http://andreacensi.github.io/contracts/ say they are for 1.7.2, and the changelog goes up to 1.7.5.
Ideally, you'd keep a CHANGELOG.rst in the root, so that sites like requires.io could display the changes more easily.
The exception messages on contract failures can be very long, which makes it more difficult to find the context of the code where the contract failed. I wonder if the standard contract message should try to re-embed some of the lines surrounding the failure location towards the bottom of the exception message?
Hi, I've just faced a problem that seq fails to match a set:
Expected a sequence, got 'set'.
checking: seq for value: Instance of set: set(['\xa3\x9a\x05\x10\xa7l\xc61\x8e', '\xd... [clip]
Is it by design?
I was curious for the numpy part of the library if you had considered anything like abstracting over lists of dimensions?
Would love to be able to write something like this:
@contract
def batch_matrix_multiply(a, b):
''' Multiplies two batches fo matrices together.
:param a: The first tensor, must end in MxN.
:type a: array[*xMxN],M>0,N>0
:param b: The second tensor, must end in NxP
:type b: array[*xNxP],P>0
:rtype: array[*xMxP]
'''Hello, guys.
I've found issue #23. Can we implement this in docstring validation?
Is there a way to specify that a parameter's type is file in the docstring?
The expression ":type in_file: file" does not work, nor do other formats I could think of.
(Note that the decorator "@contract(in_file=file)" works, but I'd like to use the docstring.)
Hello,
If I try using this
@contract(txt_pn='str, != ""', returns='list(str, != ""')
def load_txt2lst(txt_pn):
# type: (str) -> List[str]
...
or this
@contract(txt_pn='str, != \'\'', returns='list(str, != \'\'')
def load_txt2lst(txt_pn):
# type: (str) -> List[str]
...
this error is returned
...
return parse_contract_string_actual(string)
File "C:\Users\JMatos\Envs\solve_sudoku\lib\site-packages\contracts\main_actua
l.py", line 41, in parse_contract_string_actual
raise ContractSyntaxError(msg, where=where)
contracts.interface.ContractSyntaxError: Expected operation with ["!"] (at char
13), (line:1, col:14)
line 1 |list(str, != ""
\u2191
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "solve_sudoku.py", line 58, in <module>
@contract(txt_pn='str, != ""', returns='list(str, != ""')
File "C:\Users\JMatos\Envs\solve_sudoku\lib\site-packages\contracts\main.py",
line 149, in tmp_wrap
raise e.copy()
contracts.interface.ContractSyntaxError: Expected operation with ["!"] (at char
13), (line:1, col:14)
line 1 |list(str, != ""
\u2191
I am doing something wrong or is it a bug?
I'm using Python 3.5.4 32b and pycontracts 1.8.3 under Windows 7 Pro x64.
Thanks,
JM
write now it is possible to do
new_contract('alistofstrings', 'list(string)')but not
new_contract('aniteratorofstrings', 'Iterator(string)')or
new_contract('aniteratorofstrings', 'Sequence(string)')This really stifles polymorphism. I want it to accept a tuple, or a list, etc. Maybe I'm
missing something?
NB: (On a related note, I'd like something that would accept generators too, but that has it's own problems).
This does not parse correctly:
@contract(x='int,M', returns='array[(2*M+1),(2*M+1)](int)'))
and neither this:
@contract(x='int,M', returns='array[2*M+1),2*M+1](int)'))
With the following code:
from contracts import contract, new_contract
new_contract('vector', 'list[N]')
@contract
def foo(X: 'vector', y: 'vector'):
pass
foo([1, 2, 3], [1, 2])I would've expected it to fail since the two lists have different lengths, but it doesn't throw an error. However, if I don't use new_contract, it works. Is this the intended behaviour? If so, how would I get the behaviour that I want? I'm using version 1.7.15.
Hello,
I'm just trying out pycontracts and found this strange behavior when using different contract notations.
I import my module and run the function with an invalid argument and the contract's checks work fine when using this notation:
square_size = 9 # type: int
nr_rows = square_size # type: int
nr_cols = square_size # type: int
@contract(row_nr='int, >= 0, < $nr_rows', col_nr='int, >= 0, < $nr_cols',
returns='int,>= 0, < $square_size')
def loc2square_nr(row_nr, col_nr):
# type: (int, int) -> int
"""Convert location (row, column) to square number.
(int, int) -> int
:param row_nr: Row number to convert to square number.
:param col_nr: Column number to convert to square number.
:return: Square number.
"""
square_nr = (int(row_nr / square_side) * square_side
+ int(col_nr / square_side)) # type: int
return square_nr
But if I change the contract notation to inside the docstring, like this:
square_size = 9 # type: int
nr_rows = square_size # type: int
nr_cols = square_size # type: int
@contract
def loc2square_nr(row_nr, col_nr):
# type: (int, int) -> int
"""Convert location (row, column) to square number.
:param row_nr: Row number to convert to square number.
:type row_nr: int, >= 0, < $nr_rows
:param col_nr: Column number to convert to square number.
:type col_nr: int, >= 0, < $nr_cols
:return: Square number.
:rtype: int, >= 0, < $square_size
"""
square_nr = (int(row_nr / square_side) * square_side
+ int(col_nr / square_side)) # type: int
return square_nr
Python returns a huge error when importing the module (I called it base.py).
This is the final part of the error:
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "C:\Users\JMatos\Envs\solve_sudoku\lib\site-packages\pyparsing.py", line
1405, in _parseNoCache
tokens = fn( instring, tokensStart, retTokens )
File "C:\Users\JMatos\Envs\solve_sudoku\lib\site-packages\pyparsing.py", line
1049, in wrapper
ret = func(*args[limit[0]:])
File "C:\Users\JMatos\Envs\solve_sudoku\lib\site-packages\contracts\utils.py",
line 247, in f2
return f(*args, **kwargs)
File "C:\Users\JMatos\Envs\solve_sudoku\lib\site-packages\contracts\library\sc
oped_variables.py", line 95, in scoped_parse_action
val = _lookup_from_calling_scope(tokens[0])
File "C:\Users\JMatos\Envs\solve_sudoku\lib\site-packages\contracts\library\sc
oped_variables.py", line 67, in _lookup_from_calling_scope
raise ExternalScopedVariableNotFound(token)
contracts.interface.ExternalScopedVariableNotFound: Token not found: 'square_siz
e'.
>>>
If I understand correctly, pycontracts doesn't recognize the variable square_size in the 2nd notation. But it does on the 1st.
I'm using Python 32b 3.5.4 with pycontracts 1.8.3. I'm running on a Windows 7 Pro x64 PC.
Thanks,
JM
Thanks for a great project that I use everyday. It would be SO GREAT if you could find the time to complete the language reference at https://andreacensi.github.io/contracts/reference.html#contracts-language-reference!
@contract(handle='io.TextIOBase', returns='None')
def read_file_handle(self, handle: io.TextIOBase) -> 'None':
this errors:
contracts.interface.ContractSyntaxError: Unknown identifier 'io'. Did you mean 'np_zeroshape_array'? (at char 0), (line:1, col:1)
It works if I use pyparsing 2.1.5. Here's the traceback:
Python 3.5.1 (default, Jul 12 2016, 22:16:50)
[GCC 4.9.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import contracts
Traceback (most recent call last):
File "/usr/lib/python3.5/site-packages/contracts/useful_contracts/numbers.py", line 6, in <module>
import numpy # @UnusedImport
ImportError: No module named 'numpy'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1392, in _parseNoCache
tokens = fn( instring, tokensStart, retTokens )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1036, in wrapper
ret = func(*args[limit[0]:])
File "/usr/lib/python3.5/site-packages/contracts/library/suggester.py", line 108, in parse_action
raise ParseSyntaxException(pe)
pyparsing.ParseSyntaxException: Unknown identifier 'float'. Did you mean 'Float'? (at char 0), (line:1, col:1)
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/lib/python3.5/site-packages/contracts/__init__.py", line 44, in <module>
from .useful_contracts import *
File "/usr/lib/python3.5/site-packages/contracts/useful_contracts/__init__.py", line 8, in <module>
from .numbers import *
File "/usr/lib/python3.5/site-packages/contracts/useful_contracts/numbers.py", line 8, in <module>
new_contract('float', 'Float')
File "/usr/lib/python3.5/site-packages/contracts/main.py", line 535, in new_contract
return new_contract_impl(*args)
File "/usr/lib/python3.5/site-packages/contracts/main.py", line 562, in new_contract_impl
c = parse_contract_string(identifier)
File "/usr/lib/python3.5/site-packages/contracts/main.py", line 649, in parse_contract_string
return parse_contract_string_actual(string)
File "/usr/lib/python3.5/site-packages/contracts/main_actual.py", line 30, in parse_contract_string_actual
parseAll=True)[0]
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1594, in parseString
loc, tokens = self._parse( instring, 0 )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3567, in parseImpl
return self.expr._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3373, in parseImpl
ret = e._parse( instring, loc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3567, in parseImpl
return self.expr._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3567, in parseImpl
return self.expr._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3373, in parseImpl
ret = e._parse( instring, loc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3567, in parseImpl
return self.expr._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3221, in parseImpl
loc, resultlist = self.exprs[0]._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3221, in parseImpl
loc, resultlist = self.exprs[0]._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3567, in parseImpl
return self.expr._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3373, in parseImpl
ret = e._parse( instring, loc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3567, in parseImpl
return self.expr._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3221, in parseImpl
loc, resultlist = self.exprs[0]._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3221, in parseImpl
loc, resultlist = self.exprs[0]._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3567, in parseImpl
return self.expr._parse( instring, loc, doActions, callPreParse=False )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1501, in _parseCache
value = self._parseNoCache(instring, loc, doActions, callPreParse)
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1366, in _parseNoCache
loc,tokens = self.parseImpl( instring, preloc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 3373, in parseImpl
ret = e._parse( instring, loc, doActions )
File "/usr/lib/python3.5/site-packages/pyparsing.py", line 1504, in _parseCache
cache.set(lookup, pe.__class__(*pe.args))
TypeError: __init__() takes 2 positional arguments but 4 were given
>>>
Consider a file foo.py (note, this is condensed from the real test case; apologies for typos, etc.):
from contracts import contract, new_contract
@new_contract
def myContract(inst):
return isinstance(inst, object)
class Foo(object):
@contract
def __init__(self, a):
"""
:type a: number,$$%^^&
"""
passIf foo gets imported in multiple test cases (we're running them via python setup.py nosetests), then the following occurs:
======================================================================
ERROR: Failure: ContractSyntaxError (Expected "(" (at char 8), (line:1, col:9)
line 1 >number,$$%^^&
^
|
here or nearby)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/nose/loader.py", line 390, in loadTestsFromName
addr.filename, addr.module)
File "/usr/local/lib/python2.7/dist-packages/nose/importer.py", line 39, in importFromPath
return self.importFromDir(dir_path, fqname)
File "/usr/local/lib/python2.7/dist-packages/nose/importer.py", line 86, in importFromDir
mod = load_module(part_fqname, fh, filename, desc)
...
File "/usr/local/lib/python2.7/dist-packages/contracts/main.py", line 159, in contract_decorator
return contracts_decorate(function, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/contracts/main.py", line 224, in contracts_decorate
for x in accepts_dict])
File "/usr/local/lib/python2.7/dist-packages/contracts/main.py", line 306, in parse_flexible_spec
return parse_contract_string(spec)
File "/usr/local/lib/python2.7/dist-packages/contracts/main.py", line 80, in parse_contract_string
raise ContractSyntaxError(msg, where=where)
ContractSyntaxError: Expected "(" (at char 8), (line:1, col:9)
line 1 >number,$$%^^&
^
|
here or nearby
======================================================================
ERROR: Failure: ValueError (Tried to redefine 'myContract' with a definition that looks different to me.
- old: CheckCallable(<function myContract at 0x3198410>)
- new: CheckCallable(<function myContract at 0x39427d0>)
)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/nose/loader.py", line 390, in loadTestsFromName
addr.filename, addr.module)
File "/usr/local/lib/python2.7/dist-packages/nose/importer.py", line 39, in importFromPath
return self.importFromDir(dir_path, fqname)
File "/usr/local/lib/python2.7/dist-packages/nose/importer.py", line 86, in importFromDir
mod = load_module(part_fqname, fh, filename, desc)
...
File "/usr/local/lib/python2.7/dist-packages/contracts/main.py", line 493, in new_contract
new_contract_impl(identifier, function)
File "/usr/local/lib/python2.7/dist-packages/contracts/main.py", line 577, in new_contract_impl
raise ValueError(msg)
ValueError: Tried to redefine 'myContract' with a definition that looks different to me.
- old: CheckCallable(<function myContract at 0x3198410>)
- new: CheckCallable(<function myContract at 0x39427d0>)
The ValueError is repeated for every subsequent test case that imports foo, even if the class with the broken definition is not used. This can make it difficult to find the real error when there are a large number of test cases.
Hello,
Starting from a very simple integer-related contract:
int,(-1 | 1)
I've tried to write the same kind of contract, but for string values, e.g.:
string,(a | b)
Unfortunately, that syntax is not accepted by PyContracts.
It would also be nice to introduce regexp in contracts, to be matched against string values.
Thanks.
setup.py fails with the following error, and the code doesn't seem to be here :/
Traceback (most recent call last):
File "setup.py", line 35, in <module>
version = get_version(filename='src/contracts/__init__.py')
File "setup.py", line 24, in get_version
with file(filename) as f:
NameError: name 'file' is not defined
Full tracebak here
In [13]: @contract
def fn(x):
''':type x: $SeqRecord'''
return x.seq
.....
.....
raise RuntimeError("Cound not find a scope to lookup %s" % token)
ExternalScopedVariableNotFound: Token not found: 'SeqRecord'.But it works fine outside of docstring:
In [13]: @contract(x='$SeqRecord')
def fn(x):
return x.seq
In[14] # no errorHi,
I have example
@contract
def double(x:List[int]) -> List[int]:
return list(map(lambda x:x*2, x))
double([1.245]) # doesn't complain about type problem
double(False) #complains about type problem (correctly)
that currently doesn't work but it should (and I believe it's already implemented because same functionality works using contracts-native string annotation). Is there any plan to support fully Python 3 types?
new_contract('valid_name', lambda s: isinstance(s, str) and len(s)>0)
@contract(names='dict(int: (valid_name, int))')
def process_accounting(records):
...
should be
new_contract('valid_name', lambda s: isinstance(s, str) and len(s)>0)
@contract(records='dict(int: (valid_name, int))')
def process_accounting(records):
...
In [32]: @contract(a='number')
...: def foo(a):
...: return a
In [33]: foo(9999999L)
. . . . . . . .
.................. # long error outptut here
------- (end clauses) -------
checking: float|int|uint for value: Instance of <type 'long'>: 9999999L
checking: $(float|int|uint) for value: Instance of <type 'long'>: 9999999L
checking: number for value: Instance of <type 'long'>: 9999999L
Variables bound in inner context:Hi,
first of all, thanks for your awesome library!
Second, I added your library in a project of mine where I include unicode_literals (from __ future __) in all the modules, that is causing me some issues with your library because the contract decorator expects a lot of things but not a unicode object.
I was wondering if you could check on contracts/main.py:337 if the spec is an instance of basestring instead of str so that unicode will also match.
Thanks!
Hi Andrea,
I'm trying to define a custom contract using the new_contract function and the ist function from contracts.library.miscellaneous_aliases as follows:
new_contract('virConnect', ist(libvirt.virConnect))The actual checking of the argument type works, i.e. fails when the argument is not an instance of libvirt.virConnect. However, when it fails, the error message is not properly displayed. I'm getting the following exception:
Traceback (most recent call last):
File "/usr/lib/python2.7/site-packages/pyqcy/integration.py", line 60, in test
run_tests([prop], verbosity=0, propagate_exc=True)
File "/usr/lib/python2.7/site-packages/pyqcy/runner.py", line 68, in run_tests
failure.propagate_failure()
File "/usr/lib/python2.7/site-packages/pyqcy/properties.py", line 127, in test_one
result.tags = self.__execute_test(coroutine)
File "/usr/lib/python2.7/site-packages/pyqcy/properties.py", line 148, in __execute_test
obj = next(coroutine)
File "/usr/lib/python2.7/site-packages/pyqcy/properties.py", line 81, in generator_func
func(*args, **kwargs)
File "/home/anton/repos/openstack-neat/tests/test_collector.py", line 58, in get_current_vms
collector.get_current_vms(1)
File "<string>", line 2, in get_current_vms
File "/usr/lib/python2.7/site-packages/contracts/main.py", line 261, in contracts_checker
accepts_parsed[arg]._check_contract(context, bound[arg])
File "/usr/lib/python2.7/site-packages/contracts/interface.py", line 219, in _check_contract
self.check_contract(context, value)
File "/usr/lib/python2.7/site-packages/contracts/library/extensions.py", line 23, in check_contract
self.contract._check_contract(context, value)
File "/usr/lib/python2.7/site-packages/contracts/interface.py", line 219, in _check_contract
self.check_contract(context, value)
File "/usr/lib/python2.7/site-packages/contracts/library/extensions.py", line 69, in check_contract
raise ValueError(msg)
CheckError: <unprintable CheckError object>
I suppose instead of , there should be a message like 'Value is not an instance of virConnect'. Do you know what the problem is?
Thanks,
Anton
I am looking for a method to add type + value checks to function arguments
eg.
I tested this on my machine
import time
from contracts import contract
@contract(a='int,>0', b='list[N],N>0', returns='list[N]')
def c(a, b):
return [a * i for i in b]
def d(a, b):
assert isinstance(a, int)
assert a > 0
assert isinstance(b, list)
assert len(b) > 0
out = [a * i for i in b]
assert len(out) == len(b)
return out
tic = time.time()
x = list(range(100))
for i in range(1000):
x = c(5, x)
tic2 = time.time()
print(tic2-tic)
x = list(range(100))
for i in range(1000):
x = d(5, x)
tic3 = time.time()
print(tic3-tic2)Found the result to be
$ python ttest.py
0.5048160552978516
0.07861089706420898
Are there any additional checks that are missing in my manual implementation?
@AndreaCensi, contracts is an awesome library, but I think one thing that would make it better, at least to me, is ability to define custom exceptions. For example, say I define a custom Type and then I want a specific error to be raised, ideally being able to pass parameters to that error, which would then allow me to do try/except and have a typical error object with .args and .message elements.
I have not really looked at implementation details, but seems it should be doable.
Suggested by @doctoryes:
One point we've agreed upon here: It'd be nice to not have to specify new_contract('Foo', Foo) for our own classes. Instead, it'd be nice if PyContracts could just search the namespace for the classname when it's specified in a contract.
There is a similar undocumented feature that uses the expression isinstance(classname):
@contract(x="isinstance(BlockKey)")
def f(x):
pass
This checks that the name of the type of x or any of its base classes is BlockKey.
(String matching is good because this way PyContracts doesn't need to import modules or search namespaces.)
The current syntax isinstance(name) is too verbose, and this this is a common enough use case that it deserves a more intuitive syntax. The main reason that this feature is undocumented because I couldn't decide on a good compact syntax.
Some possibilities:
@contract(x="^BlockKey")
@contract(x="$BlockKey")
@contract(x="@BlockKey")
or, in general, <glyph>ClassName.
Some non-possibilities:
This one is intuitive enough but it clashes with "<" used as numerical comparison:
@contract(x="<BlockKey")
We cannot use simply:
@contract(x="BlockKey")
because it makes many expressions ambiguous; for example lisst becomes a valid contract (a class named lisst) instead of a mispelling of list.
Currently if you use the @contract decorator with a function that has something like
"""
:param myparam: something
"""
without specifying a :type myparam: you receive a traceback that looks like
File "[snip]/python2.6/site-packages/contracts/main.py", line 345, in remove_quotes
if x.startswith('') and x.endswith('') and len(x) > 3:
AttributeError: 'NoneType' object has no attribute 'startswith'
This is because on line 363, params[name].type is None, which is then passed into the remove_quotes function.
The following patch filters out malparsed params, thus stopping the error.
Patch follows:
@@ -356,6 +356,11 @@
# These are the annotations
params = annotations.params
+ #ignore non parsable params (TODO, perhaps warn?)
+ for name in params.keys():
+ if params[name].type is None:
+ del params[name]
+
name2type = dict([(name, remove_quotes(params[name].type))
for name in params])
Hello. I propose the following syntax to implement
@contract(arg="str,in('a','b','c')")
def function(arg):
pass
Action: perform the following python code
assert(arg in {'a', 'b', 'c'})
Hi.
I think I have stumbled upon a bug or if not a bug then atleast an counterintuitive feature.
Consider the following code:
#!/usr/bin/python3
# -*- coding: utf8 -*-
from contracts import contract
class ObjectThatWillSeeDeathBeforeBeingBorn(object):
@contract
def __init__(self, TestingString: 'str[>0]'):
print('String given as parameter: %s.' % TestingString)
print('ID of ObjectThatWillSeeDeathBeforeBeingBorn: %s.' % id(self))
def __del__(self):
print('ID of object being killed: %s.' % id(self))
NewObjectForTestingTheBug = ObjectThatWillSeeDeathBeforeBeingBorn()When executed as listed above we will get this as output:
$ python3 contracts_possible_bug.py
ID of object being killed: 139899159500840.
Traceback (most recent call last):
File "contracts_possible_bug.py", line 17, in <module>
NewObjectForTestingTheBug = ObjectThatWillSeeDeathBeforeBeingBorn()
TypeError: __init__() missing 1 required positional argument: 'TestingString'
The notable thing here is that we don't get any output from init but we do get output from del.
If we modify the last line so it looks like this:
NewObjectForTestingTheBug = ObjectThatWillSeeDeathBeforeBeingBorn(42)
the output looks like this:
$ python3 contracts_possible_bug.py
Traceback (most recent call last):
File "contracts_possible_bug.py", line 17, in <module>
NewObjectForTestingTheBug = ObjectThatWillSeeDeathBeforeBeingBorn(42)
File "<decorator-gen-1>", line 2, in __init__
File "/usr/local/lib/python3.5/dist-packages/contracts/main.py", line 260, in contracts_checker
raise e
File "/usr/local/lib/python3.5/dist-packages/contracts/main.py", line 255, in contracts_checker
accepts_parsed[arg]._check_contract(context, bound[arg], silent=False)
File "/usr/local/lib/python3.5/dist-packages/contracts/interface.py", line 439, in _check_contract
self.check_contract(context, value, silent)
File "/usr/local/lib/python3.5/dist-packages/contracts/library/strings.py", line 19, in check_contract
value=value, context=context)
contracts.interface.ContractNotRespected: Breach for argument 'TestingString' to ObjectThatWillSeeDeathBeforeBeingBorn:__init__().
Expected a string, got 'int'.
checking: str[>0] for value: Instance of <class 'int'>: 42
Variables bound in inner context:
- self: Instance of <class '__main__.ObjectThatWillSeeDeathBeforeBeingBorn'>: <__main__.ObjectThatWillSeeDeathBeforeBeingBorn object at 0x7fe... [clip]
ID of object being killed: 140638653347880.
I had expected that when a breach of contract has been found everything grinds to a halt but we can see that in both cases code still gets executed.
Last but not least, thank You for Your great work on contracts!
I'm attempting to match a list of a newly defined contract (using a function), however contracts.check is running the function with some odd arguments.
import contracts
import types
def cheese(tmp):
print 'seeing value', tmp
return False
contracts.check("list[int]", [1])
contracts.new_contract("cheese", cheese)
contracts.check("cheese|list[cheese]|None", [ZeroDivisionError])
When run it returns
seeing value [<type 'exceptions.ZeroDivisionError'>]
seeing value 1
Traceback (most recent call last):
File "failtest.py", line 10, in <module>
contracts.check("cheese|list[cheese]|None", [ZeroDivisionError])
File "/var/envs/node/lib/python2.6/site-packages/contracts/main.py", line 422, in check
raise e
contracts.interface.ContractNotRespected: Could not satisfy any of the 3 clauses in cheese|list[cheese]|None.
---- Clause #1: cheese
| Value does not pass criteria of cheese() (module: __main__).
| checking: function cheese() for value: Instance of list: [<type 'exceptions.ZeroDivisionError'>]
| checking: cheese for value: Instance of list: [<type 'exceptions.ZeroDivisionError'>]
---- Clause #2: list[cheese]
| Value does not pass criteria of cheese() (module: __main__).
| checking: function cheese() for value: Instance of int: 1
| checking: cheese for value: Instance of int: 1
| checking: list[cheese] for value: Instance of list: [<type 'exceptions.ZeroDivisionError'>]
---- Clause #3: None
| Value does not pass criteria of is_None() (module: contracts.library.miscellaneous_aliases).
| checking: function is_None() for value: Instance of list: [<type 'exceptions.ZeroDivisionError'>]
| checking: None for value: Instance of list: [<type 'exceptions.ZeroDivisionError'>]
------- (end clauses) -------
checking: cheese|list[cheese]|None for value: Instance of list: [<type 'exceptions.ZeroDivisionError'>]
Which, is odd. Specificity i'm wondering why the contract handler for cheese gets run with "1" as an argument. I'm thinking that it might be because the parser is getting confused about it being a variable?
Hello,
I trying to doing a contract that return a post-condition as:
@contract(x=object)
def method(x)->x.operation(5):
return x.operation(5)
I don't know if I'm doing something wrong or if not supportable, in the second case i want to request your help to do it, in order to extend the library.
Also i want to know if exist someway to construct an invariant into an object.
Thanks.
The following script will produce an error:
#!/usr/bin/env python
from contracts import contract
@contract
def foo(bar, baz):
'''
:param bar: A thing.
:type bar: str
:param baz: Another thing.
'''
print bar, baz
if __name__ == '__main__':
foo('hello', 'world')The traceback is:
Traceback (most recent call last):
File "foo.py", line 5, in <module>
@contract
File "/Users/geordan/src/contracts/src/contracts/main.py", line 166, in contract_decorator
return contracts_decorate(function, **kwargs)
File "/Users/geordan/src/contracts/src/contracts/main.py", line 228, in contracts_decorate
accepts_dict, returns = parse_contracts_from_docstring(function)
File "/Users/geordan/src/contracts/src/contracts/main.py", line 362, in parse_contracts_from_docstring
for name in params])
File "/Users/geordan/src/contracts/src/contracts/main.py", line 347, in remove_quotes
if x.startswith('``') and x.endswith('``') and len(x) > 3:
AttributeError: 'NoneType' object has no attribute 'startswith'
The following code doesn't work:
class CommodityType(object):
"""
A generic commodity type.
"""
def __init__(self, symbol):
self.symbol = symbol
# @contract(other=CommodityType)
def __eq__(self, other):
return self.symbol == other.symbol
It complains that CommodityType is not defined. Is there any way around this problem? I was trying to avoid having an isinstance check inside.
Thanks,
Similar to an older issue regarding the "tried to redefine 'xxx' with a definition that looks different to me" error (#4 ), we can get the same problem when trying to redefine returns. Think of the case that we create two new contracts for different methods and we want in both cases to check their post condition, using the special name "returns". For example in case in one method we want the return type to be bool and in another int we will get the following error:
ValueError: Tried to redefine 'returns' with a definition that looks different to me.
Since checking what different function returns will always require "returns" redefinition wouldn't be better to add a check in main.py just before throwing the error and in case the identifier is "returns" just ignore it and continue? Is there any other way to do this postcondition checking other than using the "returns"?
Also, any plans for supporting invariants? I checked this issue #10 already and I am wondering if you plan to develop this any time soon...
It is often convention to have the fixed dimensions in a numpy array be leading dimensions. However it seems like this syntax is not supported. Would it be possible to do this?
@contract
def batch_matrix_multiply(a, b):
''' Multiplies two batches fo matrices together.
:param a: The first tensor, must end in MxN.
:type a: array[...xMxN],M>0,N>0
:param b: The second tensor, must end in NxP
:type b: array[...xNxP],P>0
:rtype: array[...xMxP]
'''It would be helpful to check custom types.
For example if an instance of a class is defined as such :
class CustomClass(object):
class_member = 42
def __init__(self):
inst_member = 23
cc_inst = CustomClass()It would be useful to have a contract for a function that require such an instance.
To keep python duck-typing semantic, we can check the instance members (just like we do for dict), but checking one by one :
@contract("object(class_member: int, inst_member: int)")
def process_custom(custom_class_instance):
[...]This way any object that has an int class_member and an int inst_member will satisfy the contract.
I am posting this issue because I haven't found any way to do this currently...
First of all, I want to thank you for this project. This is really great work and design. I'm using it to develop satellite data processing pipelines at NASA Goddard.
Currently, using 'object' as a data type is not supported. A NumPy array with a datatype 'object' holds references to Python objects. Sometimes, it is useful to have a numpy array of object: unlike a list, a numpy object array can support multiple dimensions and advanced slicing.
This is a feature request to add the 'object' datatype to the contract specification for numpy arrays.
Reference:
https://andreacensi.github.io/contracts/reference.html#numpy-specific-contracts
When I run my coverage.py test suite under Python 3.7, I see this:
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:19: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Container', ist(collections.Container))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:21: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Iterable', ist(collections.Iterable))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:23: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Hashable', ist(collections.Hashable))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:27: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Iterator', ist(collections.Iterator))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:28: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Sized', ist(collections.Sized))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:29: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Callable', ist(collections.Callable))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:30: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Sequence', ist(collections.Sequence))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:31: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Set', ist(collections.Set))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:32: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('MutableSequence', ist(collections.MutableSequence))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:33: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('MutableSet', ist(collections.MutableSet))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:34: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('Mapping', ist(collections.Mapping))
/Users/ned/coverage/v4.5.x/.tox/py37/lib/python3.7/site-packages/contracts/library/miscellaneous_aliases.py:35: DeprecationWarning: Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated, and in 3.8 it will stop working
m_new_contract('MutableMapping', ist(collections.MutableMapping))
This is related to #23
I'd like do define a custom contract using something like
def checker(value, x):
return value <= x # dumb example
new_contract('foo(x)', checker)
minimum_value = 5
@contract(value='foo(minimum_value)')
def bar(value):
sqrt(value - 5)
bar(6) # fine
bar(5) # fine
bar(4) # Contract violationThat is, I would like to put a parameter x in my contract definition, have the value of x extracted from the scope where the contract decorator is used, and pass both the value of the contract parameter and the value of the function input passed into my checker function.
This seems to be possible, given builtin contracts like isinstance(Foo). Is there a way to use new_contract function/decorator to set up similar contracts, or do I need to build a manual Contract subclass?
Hi Andrea,
I am working with numpy in high dimensional spaces for machine learning. I would like to use pycontracts to check dimension which have an exponential dependency e.g.:
arg1="array[N,M]", returns="array[N**M]"
Is there a way to check this? Is it alternatively possible to pass the dimension symbols (N,M,..) to contract defined by a function?
Best
Stefan
Currently, the 'number' contract does not allow instances of numpy.float32 (or numpy.number in general), which surprised me. I think it would be nice to either:
I think that it would be a change to https://github.com/AndreaCensi/contracts/blob/master/src/contracts/library/types_misc.py#L48 :
try:
import numpy
number_types = (Number, numpy.number)
except ImportError:
number_types = Number
add_contract(Keyword('number').setParseAction(CheckType.parse_action(number_types)))For reference, here is the MRO for numpy.float32:
>>> inspect.getmro(numpy.float32)
(<type 'numpy.float32'>, <type 'numpy.floating'>, <type 'numpy.inexact'>, <type 'numpy.number'>, <type 'numpy.generic'>, <type 'object'>)
Thanks!
The file README.rst (referenced in setup.py) is not included when the sdist distribution is built. The following patch will add a MANIFEST.in file so README.rst will be included:
diff --git a/MANIFEST.in b/MANIFEST.in
new file mode 100644
index 0000000..bb37a27
--- /dev/null
+++ b/MANIFEST.in
@@ -0,0 +1 @@
+include *.rst
Hello,
Please consider accepting as contract conditions these syntax example:
@contract(square_nr='0 <= int < $square_size',
returns='tuple(A, B), 0 <= A < $nr_rows, 0 <= B < $nr_cols')
Thanks,
JM
Hi pycontracts contributors,
Wonder is it possible to remove logging.basicConfig() in contracts/src/contracts/init.py:5.
Usually when importing pycontracts as a lib, we have our own logging handlers setup, and calling basicConfig will cause duplicated logging.
Thanks
Gengyu
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.