knife documentation

knife.lazy.mapknife — Lazier evaluated mapping knife

«  knife.lazy.filterknife — Lazier evaluated filtering knife   ::   Contents   ::   knife.lazy.mathknife — Lazier evaluated mathing knife  »

knife.lazy.mapknife — Lazier evaluated mapping knife

class knife.lazy.mapknife(*things, **kw)

Lazier evaluated mapping knife. Provides mapping operations for incoming things.

>>> from knife.lazy import mapknife
__init__(*things, snapshots=5)
Parameters:
  • things – incoming things
  • snapshots (integer) – snapshots to keep (default: 5)
__len__()

Number of incoming things.

__iter__()

Iterate over outgoing things.

append(*things)

Insert things after other incoming things.

Parameters:things – incoming things
Return type:knife object 
>>> from knife import __
>>> __(3, 4, 5).append(1, 2, 3, 4, 5, 6).peek()
[3, 4, 5, 1, 2, 3, 4, 5, 6]
apply(worker, *args, **kw)

Assign callable used to work on incoming things plus any positional arguments and keyword arguments it will use.

Note

Global positional arguments and keyword arguments assigned with params are reset whenever apply is called.

Parameters:worker – a callable
Return type:knife object
argmap(merge=False)

Feed each incoming thing to worker as positional arguments.

Parameters:merge (bool) – merge global positional params with positional arguments derived from incoming things when passed to worker
Return type:knife object 
>>> from knife import __
>>> # default behavior
>>> test = __((1, 2), (2, 3), (3, 4))
>>> test.worker(lambda x, y: x * y).argmap().get()
[2, 6, 12]
>>> # merge global positional arguments with iterable arguments
>>> test.original().worker(
...   lambda x, y, z, a, b: x * y * z * a * b
... ).params(7, 8, 9).argmap(merge=True).get()
[1008, 3024, 6048]

See also

itertools.starmap
function in Python standard library
ascii(errors='strict')

encode outgoing things as bytes with the 'latin-1' codec.

Parameters:errors (str) – error handling for encoding
Return type:knife object 
>>> from stuf.six import u, b
>>> test = __([1], True, r't', b('i'), u('g'), None, (1,))
>>> test.ascii().oneach().peek()
['[1]', 'True', 't', 'i', 'g', 'None', '(1,)']
back()

Switch back to knife object that piped its incoming things through this knife object.

Return type:knife object
baseline()

Restore incoming things back to the baseline snapshot.

Return type:knife object 
>>> from knife import __
>>> undone = __(1, 2, 3).prepend(1, 2, 3, 4, 5, 6)
>>> undone.peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3]
>>> undone.snapshot().append(1).append(2).peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3, 1, 2]
>>> undone.baseline().peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3]
bytes(encoding='utf_8', errors='strict')

encode outgoing things as bytes.

Parameters:
Return type:

knife object

 
>>> test = __([1], True, r't', b('i'), u('g'), None, (1,))
>>> test.bytes().oneach().peek()
['[1]', 'True', 't', 'i', 'g', 'None', '(1,)']
clear()

Clear everything.

Return type:knife object
get()

Return outgoing things wrapped with wrap.

Note

With only one outgoing thing, only that one outgoing thing is returned. With multiple outgoing things, they are returned wrapped with the wrapper assigned with wrap (default wrapper is list).

invoke(name)

Feed global positional arguments and keyword arguments to each incoming thing’s name method.

Note

The original thing is returned if the return value of method name is None.

Parameters:name (str) – method name
Return type:knife object 
>>> # invoke list.index()
>>> __([5, 1, 7], [3, 2, 1]).params(1).invoke('index').get()
[1, 2]
>>> # invoke list.sort() but return sorted list instead of None
>>> __([5, 1, 7], [3, 2, 1]).invoke('sort').get()
[[1, 5, 7], [1, 2, 3]]

See also

invoke
function in Underscore.js
invoke
function in Underscore.lua
invoke
function in Underscore.perl
invoke
function in Underscore.php
kwargmap(merge=False)

Feed each incoming thing as a tuple of positional arguments and keyword arguments to worker.

Parameters:merge (bool) – merge global positional or keyword params with positional and keyword arguments derived from incoming things into a single tuple of wildcard positional and keyword arguments like (*iterable_args + global_args, **global_kwargs + iterable_kwargs) when passed to worker
Return type:knife object 
>>> # default behavior
>>> test = __(
...  ((1, 2), {'a': 2}), ((2, 3), {'a': 2}), ((3, 4), {'a': 2})
... )
>>> def tester(*args, **kw):
...    return sum(args) * sum(kw.values())
>>> test.worker(tester).kwargmap().get()
[6, 10, 14]
>>> # merging global and iterable derived positional and keyword args
>>> test.original().worker(tester).params(
...   1, 2, 3, b=5, w=10, y=13
... ).kwargmap(merge=True).get()
[270, 330, 390]
map()

Feed each incoming thing to worker.

Return type:knife object 
>>> __(1, 2, 3).worker(lambda x: x * 3).map().get()
[3, 6, 9]

See also

itertools.imap
function in Python standard library (replaces map in Python 3)
map
function in Underscore.js
map
function in Underscore.lua
map
function in Underscore.perl
map
function in Underscore.php
mapping(keys=False, values=False)

Run worker on incoming mapping things.

Parameters:
  • keys (bool) – collect mapping keys only
  • values (bool) – collect mapping values only
Return type:

knife object

 
>>> # filter items
>>> __(dict([(1, 2), (2, 3), (3, 4)]), dict([(1, 2), (2, 3), (3, 4)])
... ).worker(lambda x, y: x * y).mapping().get()
[2, 6, 12, 2, 6, 12]
>>> # mapping keys only
>>> __(dict([(1, 2), (2, 3), (3, 4)]), dict([(1, 2), (2, 3), (3, 4)])
... ).mapping(keys=True).get()
[1, 2, 3, 1, 2, 3]
>>> # mapping values only
>>> __(dict([(1, 2), (2, 3), (3, 4)]), dict([(1, 2), (2, 3), (3, 4)])
... ).mapping(values=True).get()
[2, 3, 4, 2, 3, 4]

See also

keys
function in Underscore.js
values
function in Underscore.js
keys
function in Underscore.lua
values
function in Underscore.lua
keys
function in Underscore.perl
values
function in Underscore.perl
keys
function in Underscore.php
values
function in Underscore.php
oneach()

Toggle whether each outgoing thing should be individually wrapped with the wrapper assigned with wrap (default wrapper is list ) or whether all outgoing things should be wrapped all at once.

Note

knife object default behavior is to wrap all outgoing things all at once. knife objects reset back to this behavior after get or peek is called.

Return type:knife object
original()

Restore incoming things back to the original snapshot.

Note

THe original snapshot of incoming things is taken following the first knife method call but before the second knife method call (if there is a second method call)

Return type:knife object 
>>> undone = __(1, 2, 3).prepend(1, 2, 3, 4, 5, 6)
>>> undone.peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3]
>>> undone.original().peek()
[1, 2, 3]
params(*args, **kw)

Assign positional arguments and keyword arguments to be used globally.

Return type:knife object
pattern(pattern, type='parse', flags=0)

Compile search pattern for use as worker.

Note

Global positional arguments and keyword arguments assigned with params are reset whenever a new pattern is compiled.

Parameters:
Return type:

knife object

 
>>> # using parse expression
>>> test = __('first test', 'second test', 'third test')
>>> test.pattern('first {}').filter().get()
'first test'
>>> # using glob pattern
>>> test.original().pattern('second*', type='glob').filter().get()
'second test'
>>> # using regular expression
>>> test.original().pattern('third .', type='regex').filter().get()
'third test'
peek()

Preview current incoming things wrapped with wrap.

Note

With only one outgoing thing, only that one thing is returned. With multiple outgoing things, everything is returned wrapped with the wrapper assigned with wrap (default wrapper is list).

pipe(knife)

Pipe incoming things from some other knife object through this knife object.

Parameters:knife – another knife object
Return type:knife object
prepend(*things)

Insert things before other incoming things.

Parameters:things – incoming things
Return type:knife object 
>>> __(3, 4, 5).prepend(1, 2, 3, 4, 5, 6).peek()
[1, 2, 3, 4, 5, 6, 3, 4, 5]
snapshot()

Take a baseline snapshot of current incoming things.

Return type:knife object
undo(snapshot=0)

Revert incoming things to a previous snapshot.

Note

A snapshot of current incoming things is taken when a knife method is called but before the main body of the method executes.

Parameters:snapshot (int) – number of steps ago 1, 2, 3, etc.
Return type:knife object 
>>> undone = __(1, 2, 3).prepend(1, 2, 3, 4, 5, 6)
>>> undone.peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3]
>>> # undo back one step
>>> undone.append(1).undo().peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3]
>>> # undo back one step
>>> undone.append(1).append(2).undo().peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3, 1]
>>> # undo back 2 steps
>>> undone.append(1).append(2).undo(2).peek()
[1, 2, 3, 4, 5, 6, 1, 2, 3, 1]
unicode(encoding='utf_8', errors='strict')

unicode (str under Python 3) decode outgoing things.

Parameters:
Return type:

knife object

 
>>> test = __([1], True, r't', b('i'), u('g'), None, (1,))
>>> test.unicode().oneach().peek()
[u'[1]', u'True', u't', u'i', u'g', u'None', u'(1,)']
worker(worker)

Assign callable used to work on incoming things.

Note

Global positional arguments and keyword arguments assigned with params are reset whenever a new worker is assigned.

Parameters:worker – a callable
Return type:knife object
wrap(wrapper)

Assign object, type, or class used to wrap outgoing things.

Note

A knife object resets back to its default wrapper (list) after get or peek is called.

Parameters:wrapper – an object, type, or class
Return type:knife object 
>>> __(1, 2, 3, 4, 5, 6).wrap(tuple).peek()
(1, 2, 3, 4, 5, 6)

«  knife.lazy.filterknife — Lazier evaluated filtering knife   ::   Contents   ::   knife.lazy.mathknife — Lazier evaluated mathing knife  »

© Copyright 2012, L. C. Rees. Created using Sphinx 1.1.3.