dinamically altering a function

[vegetax]

I i need a decorator that adds a local variable in the function it
decorates, probably related with nested scopes, for example:

def dec(func):
def wrapper(obj = None):
if not obj : obj = Obj()
<bind obj to func>
return func()

return wrapper()

@dec()
def fun(b):
obj.desc = 'marked'
obj.b = b
return obj

so the call to fun : fun(obj = myobj,'b argument')
or fun('b argument')

So the function "fun" assumes it has an obj instance and other instance
objects computed by the decorator, this decorator will be like a generic
factory for this kind of functions,which depends on the decorator to work.

 

[Steven Bethard]

I i need a decorator that adds a local variable in the function it
decorates, probably related with nested scopes, for example:

def dec(func):
def wrapper(obj = None):
if not obj : obj = Obj()
<bind obj to func>
return func()

return wrapper()

@dec()
def fun(b):
obj.desc = 'marked'
obj.b = b
return obj

so the call to fun : fun(obj = myobj,'b argument')
or fun('b argument')

So the function "fun" assumes it has an obj instance and other instance
objects computed by the decorator, this decorator will be like a generic
factory for this kind of functions,which depends on the decorator to work.

For a byte-code hack that does something similar, see the recent thread:

http://mail.python.org/pipermail/python-list/2005-March/270324.html

It can do something like:

py> class Object(object):
.... pass
....
py> @presets.presets(obj=Object())
.... def fun(b):
.... obj.desc = "marked"
.... obj.b = b
.... return obj
....
py> fun(1)
<__main__.Object object at 0x01162BD0>
py> fun(1).b
1

But note that you then only have a single instance for all calls to the
function:

py> fun(1) is fun(2)
True

Have you considered using OO here? You might find that this is more
easily written as:

py> class Object(object):
.... pass
....
py> class fun(object):
.... def __new__(self, *args):
.... if len(args) == 2:
.... obj, b = args
.... elif len(args) == 1:
.... obj, = Object(), args
.... else:
.... raise TypeError
.... obj.desc = "marked"
.... obj.b = b
.... return obj
....
py> myobj = Object()
py> fun(myobj, 2)
<__main__.Object object at 0x01162E30>
py> myobj.b
2
py> obj = fun(1)
py> obj.b
1

This doesn't use any bytecode hacks, but I'm still not certain it's
really the way to go. What is it you're trying to write this way?

STeVe

 

[vegetax]

For a byte-code hack that does something similar, see the recent thread:

http://mail.python.org/pipermail/python-list/2005-March/270324.html

It can do something like:

py> class Object(object):
... pass
...
py> @presets.presets(obj=Object())
... def fun(b):
... obj.desc = "marked"
... obj.b = b
... return obj
...
py> fun(1)
<__main__.Object object at 0x01162BD0>
py> fun(1).b
1

But note that you then only have a single instance for all calls to the
function:

py> fun(1) is fun(2)
True

Interesting hack, but the functions must not share the same object.Maybe i
could make my version of that decorator,i will check it.

Have you considered using OO here? You might find that this is more
easily written as:

py> class Object(object):
... pass
...
py> class fun(object):
... def __new__(self, *args):
... if len(args) == 2:
... obj, b = args
... elif len(args) == 1:
... obj, = Object(), args
... else:
... raise TypeError
... obj.desc = "marked"
... obj.b = b
... return obj
...
py> myobj = Object()
py> fun(myobj, 2)
<__main__.Object object at 0x01162E30>
py> myobj.b
2
py> obj = fun(1)
py> obj.b
1

This doesn't use any bytecode hacks, but I'm still not certain it's
really the way to go. What is it you're trying to write this way?

OO doesnt work here,i have factored to classes with inheritance but it looks
clumsy and it is clumsy to use, this things are in nature,functions.

What i want is to declare in the decorator some code that is common to all
these functions, so the functions assume that the decorator will be there
and wont need to duplicate the code provided by it, and the functions are
not known ahead of time, it has to be dynamic.

 

[Steven Bethard]

Have you considered using OO here? You might find that this is more
easily written as:

py> class Object(object):
... pass
...
py> class fun(object):
... def __new__(self, *args):
... if len(args) == 2:
... obj, b = args
... elif len(args) == 1:
... obj, = Object(), args
... else:
... raise TypeError
... obj.desc = "marked"
... obj.b = b
... return obj
...
py> myobj = Object()
py> fun(myobj, 2)
<__main__.Object object at 0x01162E30>
py> myobj.b
2
py> obj = fun(1)
py> obj.b
1

This doesn't use any bytecode hacks, but I'm still not certain it's
really the way to go. What is it you're trying to write this way?

OO doesnt work here,i have factored to classes with inheritance but it looks
clumsy and it is clumsy to use, this things are in nature,functions.

What i want is to declare in the decorator some code that is common to all
these functions, so the functions assume that the decorator will be there
and wont need to duplicate the code provided by it, and the functions are
not known ahead of time, it has to be dynamic.

If you need to use it as a decorator, you could try something like:

py> class Object(object):
.... pass
....
py> class ObjectSupplier(object):
.... def __init__(self, func):
.... self.func = func
.... def __call__(self, *args):
.... if len(args) == 2:
.... obj, b = args
.... elif len(args) == 1:
.... obj, = Object(), args
.... else:
.... raise TypeError
.... return self.func(obj, b)
....
py> @ObjectSupplier
.... def fun(obj, b):
.... obj.desc = "marked"
.... obj.b = b
.... return obj
....
py> fun(1)
<__main__.Object object at 0x01162910>
py> fun(1).b
1
py> myobject = Object()
py> fun(myobject, 2)
<__main__.Object object at 0x01162770>
py> myobject.b
2

Basically, ObjectSupplier just guarantees that the function will be
called with an initialized Object argument if one isn't supplied. Note
that your functions must be declared with the initial argument now though.

STeVe

P.S. Making ObjectSupplier general enough to work with any number of
arguments is left as an exercise to the reader.

 

[Jeremy Bowers]

What i want is to declare in the decorator some code that is common to all

these functions, so the functions assume that the decorator will be there
and wont need to duplicate the code provided by it, and the functions are
not known ahead of time, it has to be dynamic.

This sounds like a call for a Pythonic varient on the Template pattern:

class Root(object):
def __init__(self):
self.dataChunk = 22 # or whatever

class Child(Root):
def __call__(self):
print self.dataChunk
22


Don't be put off by the "OO"-ness here, it acts just like a function
thanks to __call__, and behind the scenes you get full OO support for your
functions.

I strongly suspect this is the best solution to your problem, not a
decorator. Note whatever you are doing to create the functions can be done
in other ways, especially note that "class" statements are executed, not
declarations, for instance:
.... def __init__(self):
.... self.op1 = 22
.... self.op2 = 44
....

funcs = []
for op in operator.add, operator.sub, operator.pow:

.... def newfunc(self, basefunc = op):
.... print basefunc(self.op1, self.op2)
.... class New(Root):
.... __call__ = newfunc
.... funcs.append(New)
....

funcs # show the classes

[x() for x in funcs] # show the new "functions"

[x()() for x in funcs] # call each of the functions, note no return

[/QUOTE]
66
-22
116572995441436549620289361494791139391860487905922101805056
[None, None, None]

Upshot is, with a bit of creativity this can do whatever you want, in
conjection with dynamically-created classes, no bytecode hacks, no really
weird decorators, just standard OO and __call__.

 

[Bengt Richter]

Interesting hack, but the functions must not share the same object.Maybe i
could make my version of that decorator,i will check it.

What do you mean "function_s_" [plural] ? I see only one function named "fun".
An ordinary decorator will only modify that single function, so if you want
more than one function instance, IWT you would want a factory function that
can produce multiple function instances, tailored as you like. You could
use a decorator to modify a template function within the factory, and return
the result. But you might no need a decorator if you just make your template
function refer to stuff passed into the factory, i.e., as cell variables.

Will your template function know the names of dynamic things it will use?
If not, what kind of protocol will you use to discover what's available?
If it does know the names (like "obj") then you could just write something like

... if obj is None: obj = Obj()
... def fun(b):
... obj.desc = 'marked'
... obj.b = b
... return obj
... return fun
... Traceback (most recent call last):
File "<stdin>", line 1, in ?
File "<stdin>", line 2, in mkfun
NameError: global name 'Obj' is not defined

oops
True

The code of the function the factory produces is:
4 0 LOAD_CONST 1 ('marked')
3 LOAD_DEREF 0 (obj)
6 STORE_ATTR 1 (desc)

5 9 LOAD_FAST 0 (b)
12 LOAD_DEREF 0 (obj)
15 STORE_ATTR 2 (b)

6 18 LOAD_DEREF 0 (obj)
21 RETURN_VALUE

We could use the byte code hack to change the code not to use
LOAD_DEREF cell vars, but preset and use a local variable instead:
... if obj is None: obj = Obj()
... @presets(obj=obj)
... def fun(b):
... obj.desc = 'marked'
... obj.b = b
... return obj
... return fun
... 3 0 LOAD_CONST 2 (<__main__.Obj object at 0x02EF728C>)
3 STORE_FAST 1 (obj)

5 6 LOAD_CONST 1 ('marked')
9 LOAD_FAST 1 (obj)
12 STORE_ATTR 1 (desc)

6 15 LOAD_FAST 0 (b)
18 LOAD_FAST 1 (obj)
21 STORE_ATTR 2 (b)

7 24 LOAD_FAST 1 (obj)
27 RETURN_VALUE

Not sure that gains anything significant.

Have you considered using OO here? You might find that this is more
easily written as:

py> class Object(object):
... pass
...
py> class fun(object):
... def __new__(self, *args):
... if len(args) == 2:
... obj, b = args
... elif len(args) == 1:
... obj, = Object(), args
... else:
... raise TypeError
... obj.desc = "marked"
... obj.b = b
... return obj
...
py> myobj = Object()
py> fun(myobj, 2)
<__main__.Object object at 0x01162E30>
py> myobj.b
2
py> obj = fun(1)
py> obj.b
1

This doesn't use any bytecode hacks, but I'm still not certain it's
really the way to go. What is it you're trying to write this way?

Yes, that is a good question ;-)

OO doesnt work here,i have factored to classes with inheritance but it looks
clumsy and it is clumsy to use, this things are in nature,functions.

What i want is to declare in the decorator some code that is common to all
these functions, so the functions assume that the decorator will be there
and wont need to duplicate the code provided by it, and the functions are
not known ahead of time, it has to be dynamic.

Still not sure what you mean by "the functions are not known ahead of time."

Please pretend things worked the way you want, and post an example. Maybe we can
make it work.

Regards,
Bengt Richter

 

[vegetax]

Interesting hack, but the functions must not share the same object.Maybe i
could make my version of that decorator,i will check it.

What do you mean "function_s_" [plural] ? I see only one function named
"fun". An ordinary decorator will only modify that single function, so if
you want more than one function instance, IWT you would want a factory
function that can produce multiple function instances, tailored as you
like. You could use a decorator to modify a template function within the
factory, and return the result. But you might no need a decorator if you
just make your template function refer to stuff passed into the factory,
i.e., as cell variables.

Will your template function know the names of dynamic things it will use?
If not, what kind of protocol will you use to discover what's available?
If it does know the names (like "obj") then you could just write something
like

... if obj is None: obj = Obj()
... def fun(b):
... obj.desc = 'marked'
... obj.b = b
... return obj
... return fun
...Traceback (most recent call last):
File "<stdin>", line 1, in ?
File "<stdin>", line 2, in mkfun
NameError: global name 'Obj' is not defined

oops
True

The code of the function the factory produces is:
4 0 LOAD_CONST 1 ('marked')
3 LOAD_DEREF 0 (obj)
6 STORE_ATTR 1 (desc)

5 9 LOAD_FAST 0 (b)
12 LOAD_DEREF 0 (obj)
15 STORE_ATTR 2 (b)

6 18 LOAD_DEREF 0 (obj)
21 RETURN_VALUE

We could use the byte code hack to change the code not to use
LOAD_DEREF cell vars, but preset and use a local variable instead:
... if obj is None: obj = Obj()
... @presets(obj=obj)
... def fun(b):
... obj.desc = 'marked'
... obj.b = b
... return obj
... return fun
...3 0 LOAD_CONST 2 (<__main__.Obj object at
0x02EF728C>)
3 STORE_FAST 1 (obj)

5 6 LOAD_CONST 1 ('marked')
9 LOAD_FAST 1 (obj)
12 STORE_ATTR 1 (desc)

6 15 LOAD_FAST 0 (b)
18 LOAD_FAST 1 (obj)
21 STORE_ATTR 2 (b)

7 24 LOAD_FAST 1 (obj)
27 RETURN_VALUE

Not sure that gains anything significant.

Have you considered using OO here? You might find that this is more
easily written as:

py> class Object(object):
... pass
...
py> class fun(object):
... def __new__(self, *args):
... if len(args) == 2:
... obj, b = args
... elif len(args) == 1:
... obj, = Object(), args
... else:
... raise TypeError
... obj.desc = "marked"
... obj.b = b
... return obj
...
py> myobj = Object()
py> fun(myobj, 2)
<__main__.Object object at 0x01162E30>
py> myobj.b
2
py> obj = fun(1)
py> obj.b
1

This doesn't use any bytecode hacks, but I'm still not certain it's
really the way to go. What is it you're trying to write this way?

Yes, that is a good question ;-)

OO doesnt work here,i have factored to classes with inheritance but it
looks clumsy and it is clumsy to use, this things are in nature,functions.

What i want is to declare in the decorator some code that is common to all
these functions, so the functions assume that the decorator will be there
and wont need to duplicate the code provided by it, and the functions are
not known ahead of time, it has to be dynamic.

Still not sure what you mean by "the functions are not known ahead of
time."

Please pretend things worked the way you want, and post an example. Maybe
we can make it work.

Regards,
Bengt Richter

Ok , the complete use case is :

def fun1(a,b,obj = None):
isSuplied = 1
if not obj :
obj = generateObject()
isSuplied = 0

<do something with obj>

if not isSuplied : obj.dispose()

def fun2(c,obj = None):
isSuplied = 1
if not obj :
obj = generateObject()
isSuplied = 0

<do something else with obj>

if not isSuplied : obj.dispose()

So , maybe i will need to define fun3 later,i dont know , but fun3 also will
work with "obj".
So the thing is to try to factorize the code,so that it is in the decorator:

def dec(func):
def wrapper(*arg,**kw):
obj = kw.get('obj',None)
isSuplied = 1
if not obj :
obj = generateObject()
kw['obj'] = obj
isSuplied = 0

res = func(*arg,**kw)
if not isSuplied : obj.dispose()
return res

return wrapper

so the previous function declarations, will look like:

@dec
def fun1(a,b,obj = None):
<do something with obj>

@dec
def fun2(c,obj = None):
<do something with obj>

but i have to define obj = None in each function,but the functions should
know they will have an obj instance available, so the byte code hack would
be :

def dec(func):
def wrapper(*arg,**kw):
obj = kw.get('obj',None)
isSuplied = 1
if not obj :
obj = generateObject()
isSuplied = 0

<< BIND OBJ TO THE FUNCTION LOCAL NAMESPACE >>

res = func(*arg,**kw)
if not isSuplied : obj.dispose()
return res

return wrapper

so the functions knowing they will be altered(added an instance of obj),they
will just need to define their own arguments only.

@dec
fun1(a,b) : obj.result = a + b
@dec
fun2(c) : obj.result = c * 4

So the duplication would be removed.

But i realized the best aproach,is a callable class that provides that
infrastructure, or not? =P

 

[Bengt Richter]

Bengt Richter wrote: [...]

Please pretend things worked the way you want, and post an example. Maybe
we can make it work.

Regards,
Bengt Richter

Ok , the complete use case is :

def fun1(a,b,obj = None):
isSuplied = 1
if not obj :
obj = generateObject()
isSuplied = 0

<do something with obj>

if not isSuplied : obj.dispose()

def fun2(c,obj = None):
isSuplied = 1
if not obj :
obj = generateObject()
isSuplied = 0

<do something else with obj>

if not isSuplied : obj.dispose()

So , maybe i will need to define fun3 later,i dont know , but fun3 also will
work with "obj".
So the thing is to try to factorize the code,so that it is in the decorator:

def dec(func):
def wrapper(*arg,**kw):
obj = kw.get('obj',None)
isSuplied = 1
if not obj :
obj = generateObject()
kw['obj'] = obj
isSuplied = 0

res = func(*arg,**kw)
if not isSuplied : obj.dispose()
return res

return wrapper

so the previous function declarations, will look like:

@dec
def fun1(a,b,obj = None):
<do something with obj>

@dec
def fun2(c,obj = None):
<do something with obj>

but i have to define obj = None in each function,but the functions should
know they will have an obj instance available, so the byte code hack would
be :

def dec(func):
def wrapper(*arg,**kw):
obj = kw.get('obj',None)
isSuplied = 1
if not obj :
obj = generateObject()
isSuplied = 0

<< BIND OBJ TO THE FUNCTION LOCAL NAMESPACE >>

res = func(*arg,**kw)
if not isSuplied : obj.dispose()
return res

return wrapper

so the functions knowing they will be altered(added an instance of obj),they
will just need to define their own arguments only.

@dec
fun1(a,b) : obj.result = a + b
@dec
fun2(c) : obj.result = c * 4

So the duplication would be removed.

But i realized the best aproach,is a callable class that provides that
infrastructure, or not? =P

Maybe this implements what you are asking for?

----< vegetax.py >-------------------------
from ut.presets import presets

class generateObject(object):
def dispose(self): print 'disposing of %r: %r' %(self, vars(self))

def dec(func):
def getfun(f): return func # fake decorator to pass func to presets
def wrapper(*arg,**kw):
obj = kw.get('obj',None)
isSuplied = 1
if not obj :
obj = generateObject()
isSuplied = 0
else:
del kw['obj'] # don't pass unexpected kwarg to func
#<< BIND OBJ TO THE FUNCTION LOCAL NAMESPACE >>
@presets(obj=obj) # per comment above
@getfun
def funcalias(): pass # just for a local name to bind
res = funcalias(*arg, **kw)
if not isSuplied : obj.dispose()
return res
return wrapper

#so the functions knowing they will be altered(added an instance of obj),they
#will just need to define their own arguments only.
if __name__ == '__main__':
@dec
def fun1(a,b) : obj.result = a + b; return obj, vars(obj)
@dec
def fun2(c) : obj.result = c * 4; return obj, vars(obj)
o, v = fun1(111, 222) # bind obj to show that new obj is generated on next call
print o, v
print fun2(1111)
print 'supplying obj to fun2:', fun2(2222, obj=generateObject()) # supplied
print 'both ways:', (fun2(1111), fun2(2222, obj=generateObject()))
-------------------------------------------

[22:21] C:\pywk\clp>py24 vegetax.py
disposing of <__main__.generateObject object at 0x02EF9F0C>: {'result': 333}
<__main__.generateObject object at 0x02EF9F0C> {'result': 333}
disposing of <__main__.generateObject object at 0x02EF33CC>: {'result': 4444}
(<__main__.generateObject object at 0x02EF33CC>, {'result': 4444})
supplying obj to fun2: (<__main__.generateObject object at 0x02EF33CC>, {'result': 8888})
both ways: disposing of <__main__.generateObject object at 0x02EF33CC>: {'result': 4444}
((<__main__.generateObject object at 0x02EF33CC>, {'result': 4444}), (<__main__.generateObject o
bject at 0x02EF316C>, {'result': 8888}))


Regards,
Bengt Richter