foc 0.4.6

A collection of python functions for somebody's sanity

foc

fun oriented code or francis' odd collection.

Functions from the Python standard library are great. But some notations are a bit painful and confusing for personal use, so I created this odd collection of functions.

Tl;dr

• foc provides a collection of higher-order functions, some helpful (pure) functions.
• foc provides an easy way to compose functions with symbol. (. and |)
• foc respects the python standard library. Never reinvented the wheel.

How to use

# install
$ pip install -U foc

# import
>>> from foc import *

Ground rules

• No dependencies except for the python standard library

• No unnessary wrapping objects.

• Most function implementations should be less than 5-lines.

• Followed haskell-like function names and arguments order

• Used python generator first if possible. (lazy-evaluation)

  map, filter, zip, range, flat ...

• Provide the functions that unpack generators in list as well.

• Function names that end in l indicate the result will be unpacked in a list.

  mapl, filterl, zipl, rangel, flatl, takewhilel, dropwhilel, ...

• Function names that end in _ indicate that the function is a partial application builder.

  cf_, f_, ff_, c_, cc_, u_, ...

Quickstart

Let's pick lottery numbers. That is to pick 6 numbers from 1 to 45. The Lotto (korean lottery)

>>> range(1, 46) | choice(size=6) | sort
[2, 8, 22, 24, 37, 39]

# This is one game. People usually buy five games at once.

>>> [ range(1, 46) | choice(size=6) | sort for _ in range(5) ]
[[4, 6, 11, 38, 41, 45],
 [5, 8, 23, 25, 26, 40],
 [13, 18, 23, 25, 37, 44],
 [17, 21, 24, 32, 41, 43],
 [5, 9, 13, 25, 30, 38]]

>>> gumballs = replicate(5, range(1, 46))     # 5-set of gumballs

# in Unix pipelines manner
>>> gumballs | map(choice(size=6)) | map(sort) | collect
[[1, 3, 5, 10, 23, 41],
 [14, 18, 28, 33, 37, 39],
 [13, 15, 19, 23, 32, 45],
 [4, 11, 19, 27, 30, 39],
 [8, 33, 35, 39, 40, 41]]

# with Haskell-like mathematical symbol
>>> (collect . map(sort . fx(choice(size=6))))(gumballs)
[[4, 14, 15, 28, 42, 44],
 [12, 34, 37, 40, 41, 42],
 [7, 10, 21, 26, 31, 39],
 [6, 11, 12, 14, 25, 32],
 [2, 13, 15, 26, 27, 41]]

The functions foc provides are not particularly different. Exactly, it look like normal functions.

>>> id("francis")
'francis'

>>> even(3)
False

>>> take(3, range(5, 10))
[5, 6, 7]

foc just adds ways to compose functions with symbols

symbol	description	evaluation order	Available functions
. (dot)	same as the mathematical symbol	backwards	all globals, all built-ins
| (pipeline)	in Unix pipeline manner	in order	@fx-decorated functions

Composition of Functions with .

>>> (length . range)(10)
10

>>> (collect . filter(even) . range)(10)  # 'collect' unpacks generators
[0, 2, 4, 6, 8]

>>> (sum . map(f_("+", 5)) . range)(10)   # f_("+", 5) == lambda x: x+5
95

>>> (last . sort . shuffle . collect . range)(11)
10

all functions in globals() including all built-ins can be direcly composed by ., except for two.

• lambda
• partial application like: partial(map, lambda x: x+5)

  the same as map(f_("+", 5)). They are interchangeable.

In those case, just wrap them in fx.

>>> (fx(lambda x: x+2) . fx(lambda x: x*8))(5)    # don't. fx(lambda *args, **kwargs: ...)
42
>>> (id . f_("+", 2) . f_("*", 8))(5)             # isn't it better?
42

>>> (sum . fx(partial(map, lambda x: x+5)))(range(5))   # don't partial(map, lambda ...)
37
>>> (sum . map(f_("+", 5)))(range(5))                   # `map(f_("+", 5))` is enough
37

>>> (unchars . map(chr))(range(73, 82))
'IJKLMNOPQ'

>>> (collect . map(pred . succ) . range)(5)
[0, 1, 2, 3, 4]

But, it's very tedious work wrapping partial map in fx every time. Thus, f_, ff_, curry, uncurry, map, and filter have been processed so that they can be used without fx.

See also: f_, ff_, c_, cc_, u_, curry, uncurry, map, and filter

Composition of Functions with |

>>> range(10) | length
10

>>> range(10) | filter(even) | collect    # 'collect' unpacks generators
[0, 2, 4, 6, 8]

>>> range(10) | map(f_("+", 5)) | sum     # f_("+", 5) == lambda x: x+5
95

>>> rangel(11) | shuffle | sort | last
10

Unlike the case of ., composing functions with | is allowed only for composable functions (or fx function). But don't worry. Most functions foc provides are the fx function.

fx functions (or Function eXtension) are @fx-decorated functions.

To list/get all available fx functions, call catalog(fx=True) or lsfx().

If you want to make a function the fx function on the fly, just wrap the function in fx.

>>> 7 | fx(lambda x: x * 6)
42

Try binding a function to a new reference:

>>> foo = fx(func)

or use fx decorator. All the same.

>>> @fx               # from now on, arg | func == func(arg)
... def func(arg):     
...    ...

Examples

These are part of the symbol-composable functions foc provides.

To list all available functions, call catalog().

Basic (pure) functions

>>> id("francis")
'francis'

>>> const(5, "no-matther-what-comes-here")
5

>>> seq("only-returns-the-following-arg", 5)
5

>>> void(randbytes(256))

>>> fst(["sofia", "maria", "claire"])
'sofia'

>>> snd(("sofia", "maria", "claire"))
'maria'

>>> nth(3, ["sofia", "maria", "claire"]) 
'claire'

>>> take(3, range(5, 10))
[5, 6, 7]

>>> drop(3, "github") | collect 
['h', 'u', 'b']

>>> head(range(1,5))              # returns 'None' when []
1

>>> last(range(1,5))              # returns 'None' when []
4

>>> init(range(1,5)) | collect    # returns [] when []
[1, 2, 3]

>>> tail(range(1,5)) | collect    # returns [] when []
[2, 3, 4]

>>> pair("sofia", "maria")
('sofia', 'maria')

>>> pred(3)
2

>>> succ(3)
4

>>> odd(3)
True

>>> even(3)
False

>>> null([]) == null(()) == null({}) == null("")
True

>>> elem(5, range(10))
True

>>> not_elem("fun", "functions")
False

>>> nub("3333-13-1111111")
['3', '-', '1']

>>> chars("sofimarie")
['s', 'o', 'f', 'i', 'm', 'a', 'r', 'i', 'e']

>>> unchars(['s', 'o', 'f', 'i', 'm', 'a', 'r', 'i', 'e'])
'sofimarie'

>>> words("fun on functions")
['fun', 'on', 'functions']

>>> unwords(['fun', 'on', 'functions'])
'fun on functions'

>>> lines("fun\non\nfunctions")
['fun', 'on', 'functions']

>>> unlines(['fun', 'on', 'functions'])
"fun\non\nfunctions"

>>> take(3, repeat(5))        # repeat(5) = [5, 5, ...]
[5, 5, 5]

>>> take(5, cycle("fun"))     # cycle("fun") = ['f', 'u', 'n', 'f', 'u', 'n', ...]
['f', 'u', 'n', 'f', 'u']

>>> replicate(3, 5)           # the same as 'take(3, repeat(5))'
[5, 5, 5]

>>> take(3, count(2))         # count(2) = [2, 3, 4, 5, ...]
[2, 3, 4]

>>> take(3, count(2, 3))      # count(2, 3) = [2, 5, 8, 11, ...]
[2, 5, 8]

Higher-order functions

>>> flip(pow)(7, 3)                             # the same as `pow(3, 7) = 3 ** 7`
2187

>>> bimap(f_("+", 3), f_("*", 7), (5, 7))       # bimap (3+) (7*) (5, 7)
(8, 49)                                         # (3+5, 7*7)

>>> first(f_("+", 3), (5, 7))                   # first (3+) (5, 7)
(8, 7)                                          # (3+5, 7)

>>> second(f_("*", 7), (5, 7))                  # second (7*) (5, 7)
(5, 49)                                         # (5, 7*7)

>>> take(5, iterate(lambda x: x**2, 2))         # [2, 2**2, (2**2)**2, ((2**2)**2)**2, ...]
[2, 4, 16, 256, 65536]

>>> [* takewhile(even, [2, 4, 6, 1, 3, 5]) ]    
[2, 4, 6]

>>> takewhilel(even, [2, 4, 6, 1, 3, 5])
[2, 4, 6]

>>> [* dropwhile(even, [2, 4, 6, 1, 3, 5]) ]    
[1, 3, 5]

>>> dropwhilel(even, [2, 4, 6, 1, 3, 5])
[1, 3, 5]

# fold with a given initial value from the left
>>> foldl("-", 10, range(1, 5))                 # foldl (-) 10 [1..4]
0

# fold with a given initial value from the right
>>> foldr("-", 10, range(1, 5))                 # foldr (-) 10 [1..4]
8

# `foldl` without an initial value (used first item instead)
>>> foldl1("-", range(1, 5))                    # foldl1 (-) [1..4]
-8

# `foldr` without an initial value (used first item instead)
>>> foldr1("-", range(1, 5))                    # foldr1 (-) [1..4]
-2

# accumulate reduced values from the left
>>> scanl("-", 10, range(1, 5))                 # scanl (-) 10 [1..4]
[10, 9, 7, 4, 0]

# accumulate reduced values from the right
>>> scanr("-", 10, range(1, 5))                 # scanr (-) 10 [1..4]
[8, -7, 9, -6, 10]

# `scanl` but no starting value
>>> scanl1("-", range(1, 5))                    # scanl1 (-) [1..4]
[1, -1, -4, -8]

# `scanr` but no starting value
>>> scanr1("-", range(1, 5))                    # scanr1 (-) [1..4]
[-2, 3, -1, 4]

>>> concatl(["sofia", "maria"])
['s', 'o', 'f', 'i', 'a', 'm', 'a', 'r', 'i', 'a']
# Note that ["sofia", "maria"] = [['s','o','f','i','a'], ['m','a','r','i','a']]

>>> concatmapl(str.upper, ["sofia", "maria"])   
['S', 'O', 'F', 'I', 'A', 'M', 'A', 'R', 'I', 'A']

Real-World Example

A causal self-attention of the transformer model based on pytorch can be described as follows.
Somebody insists that this helps to follow the process flow without distraction.

    def forward(self, x):
        B, S, E = x.size()  # size_batch, size_block (sequence length), size_embed
        N, H = self.config.num_heads, E // self.config.num_heads  # E == (N * H)

        q, k, v = self.c_attn(x).split(self.config.size_embed, dim=2)
        q = q.view(B, S, N, H).transpose(1, 2)  # (B, N, S, H)
        k = k.view(B, S, N, H).transpose(1, 2)  # (B, N, S, H)
        v = v.view(B, S, N, H).transpose(1, 2)  # (B, N, S, H)

        # Attention(Q, K, V)
        #   = softmax( Q*K^T / sqrt(d_k) ) * V
        #         // q*k^T: (B, N, S, H) x (B, N, H, S) -> (B, N, S, S)
        #   = attention-prob-matrix * V
        #         // prob @ v: (B, N, S, S) x (B, N, S, H) -> (B, N, S, H)
        #   = attention-weighted value (attention score)

        return cf_(
            self.dropout,  # dropout of layer's output
            self.c_proj,  # linear projection
            ff_(torch.Tensor.view, *rev(B, S, E)),  # (B, S, N, H) -> (B, S, E)
            torch.Tensor.contiguous,  # contiguos in-memory tensor
            ff_(torch.transpose, *rev(1, 2)),  # (B, S, N, H)
            ff_(torch.matmul, v),  # (B, N, S, S) x (B, N, S, H) -> (B, N, S, H)
            self.dropout_attn,  # attention dropout
            ff_(torch.masked_fill, *rev(mask == 0, 0.0)),  # double-check masking
            f_(F.softmax, dim=-1),  # softmax
            ff_(torch.masked_fill, *rev(mask == 0, float("-inf"))),  # no-look-ahead
            ff_("/", math.sqrt(k.size(-1))),  # / sqrt(d_k)
            ff_(torch.matmul, k.transpose(-2, -1)),  # Q @ K^T -> (B, N, S, S)
        )(q)

In Detail

Get binary functions from python operators: sym

sym(OP) converts python's symbolic operators into binary functions.
The string forms of operators like +, -, /, *, **, ==, !=, .. represent the corresponding binary functions.

To list all available symbol operators, call sym().

>>> sym("+")(5, 2)                 # 5 + 2
7

>>> sym("==")("sofia", "maria")    # "sofia" == "maria"
False

>>> sym("%")(123456, 83)           # 123456 % 83
35

Build partial application: f_ and ff_

• f_ build left-associative partial application,
  where the given function's arguments partially evaluation from the left.
• ff_ build right-associative partial application,
  where the given function's arguments partially evaluation from the right.

f_(fn, *args, **kwargs)
ff_(fn, *args, **kwargs) == f_(flip(fn), *args, **kwargs)

>>> f_("+", 5)(2)    # the same as `(5+) 2` in Haskell
7                    # 5 + 2

>>> ff_("+", 5)(2)   # the same as `(+5) 2 in Haskell`
7                    # 2 + 5

>>> f_("-", 5)(2)    # the same as `(5-) 2`
3                    # 5 - 2

>>> ff_("-", 5)(2)   # the same as `(subtract 5) 2`
-3                   # 2 - 5

Build curried functions: c_ (curry) and cc_

• c_ is an alias for curry
• c_ takes the function's arguments from the left
• while cc_ takes them from the right.

c_(fn) == curry(fn)
cc_(fn) == c_(flip(fn))

See also uncurry

# currying from the left args
>>> c_("+")(5)(2)    # 5 + 2
7

>>> c_("-")(5)(2)    # 5 - 2
3

# currying from the right args
>>> cc_("+")(5)(2)   # 2 + 5
7

>>> cc_("-")(5)(2)   # 2 - 5
-3

Build unary functions on a tuple: u_ (uncurry)

• u_ is an alias for uncurry
• u_ converts a normal function to a unary function that takes a tuple of arguments only
• uncurry :: (a -> ... -> b -> o) -> (a, ..., b) -> o

>>> uncurry(pow)((2, 10))    # pow(2, 10)
1024

>>> (2, 3) | uncurry("+")    # 2 + 3 or (+) 2 3 
5

>>> ([1, 3], [2, 4]) | uncurry(zip) | collect    # collect(zip([1, 3], [2, 4]))
[(1, 2), (3, 4)]

>>> (collect . uncurry(zip))(([1,3], [2,4],))     # the same
[(1, 2), (3, 4)]

Build composition of functions: cf_ and cfd

• cf_ (composition of function) composes functions using the given list of functions.
• cfd (composing-function decorator) decorates a function with the given list of functions.

cf_(*fn, rep=None)
cfd(*fn, rep=None)

>>> square = ff_("**", 2)        # the same as (^2) in Haskell
>>> add5 = ff_("+", 5)           # the same as (+5) in Haskell
>>> mul7 = ff_("*", 7)           # the same as (*7) in Haskell

>>> cf_(mul7, add5, square)(3)   # (*7) . (+5) . (^2) $ 3
98                               # mul7(add5(square(3))) = ((3 ^ 2) + 5) * 7

>>> cf_(square, rep=3)(2)        # cf_(square, square, square)(2) == ((2 ^ 2) ^ 2) ^ 2 = 256
256


>>> @cfd(mul7, add5, square)
... def foo(x):
...    return len(x)

>>> foo([1,2,3])
98

# compare `cf_` with `cfd`
cf_(a, b, c, d, f)(x)    # (a . b . c . d . f)(x) 

cfd(a, b, c, d)(f)(x)    # (a . b . c . d)(f(x))

cfd is very handy and useful to recreate previously defined functions by composing functions. All you need is to write a basic functions to do fundamental things.

Seamlessly extends: map, filter and zip

• Extend usability while maintaining full compatibility
• No harm to existing usage. Just added ways to compose function with symbols

map(fn, *xs)
mapl(fn, *xs)

>>> (collect . map(abs))(range(-2, 3)) 
[2, 1, 0, 1, 2]
>>> map(abs)(range(-2, 3)) | collect
[2, 1, 0, 1, 2]

>>> (collect . map(lambda x: x*8))(range(1, 6))
[8, 16, 24, 32, 40]
>>> range(1, 6) | map(lambda x: x*8) | collect
[8, 16, 24, 32, 40]

>>> (collect . map("*", [1, 2, 3]))([4, 5, 6])
[4, 10, 18]
>>> [4, 5, 6] | map("*", [1, 2, 3]) | collect
[4, 10, 18]

filter(p, xs)
filterl(p, xs)

>>> (collect . filter(f_("==", "f")))("fun-on-functions")
['f', 'f']
>>> filter(f_("==", "f"))("fun-on-functions") | collect
['f', 'f']

>>> primes = [2, 3, 5, 7, 11, 13, 17, 19]
>>> (collect . filter(lambda x: x % 3 == 2))(primes)
[2, 5, 11, 17]
>>> primes | filter(cf_(ff_("==", 2), ff_("%", 3))) | collect
[2, 5, 11, 17]

zip(*xs, strict=False)
zipl(*xs, strict=False)

>>> (collect . f_(zip, "LOVE") . range)(3)
[('L', 0), ('O', 1), ('V', 2)]
>>> zip("LOVE", range(3)) | collect
[('L', 0), ('O', 1), ('V', 2)]

>>> (collect . uncurry(zip))(("LOVE", range(3),))
[('L', 0), ('O', 1), ('V', 2)]
>>> ("LOVE", range(3)) | uncurry(zip) | collect
[('L', 0), ('O', 1), ('V', 2)]

Lazy Evaluation: lazy and force

• lazy defers the evaluation of a function(or expression) and returns the deferred expression.
• force forces the deferred-expression to be fully evaluated when needed.

  it reminds Haskell's force x = deepseq x x.

lazy(fn, *args, **kwargs)
force(EXPR)
mforce([EXPR])

# strictly generate a random integer between [1, 10)
>>> randint(1, 10)

# generate a lazy expression for the above
>>> deferred = lazy(randint, 1, 10)

# evaluate it when it need
>>> force(deferred)

# the same as above
>>> deferred()

Are those evaluations with lazy really deferred?

>>> long_list = randint(1, 100000, 100000)    # a list of one million random integers

>>> %timeit sort(long_list)
142 ms ± 245 µs per loop (mean ± std. dev. of 7 runs, 10 loops each)

# See the evaluation was deferred
>>> %timeit lazy(sort, long_list)
1.03 µs ± 2.68 ns per loop (mean ± std. dev. of 7 runs, 1,000,000 loops each

when to use

For given a function randint(low, high), how can we generate a list of random integers?

[ randint(1, 10) for _ in range(5) ]    # exactly the same as 'randint(1, 10, 5)'

It's the simplest way but what about using replicate?

# generate a list of random integers using 'replicate'
>>> replicate(5, randint(1, 10))
[7, 7, 7, 7, 7]        # ouch, duplication of the first evaluated item.

Wrong! This result is definitely not what we want. We need to defer the function evaluation till it is replicated.

Just use lazy(randint, 1, 10) instead of randint(1, 10)

# replicate 'deferred expression'
>>> randos = replicate(5, lazy(randint, 1, 10))

# evaluate when needed
>>> mforce(randos)      # mforce = map(force), map 'force' over deferred expressions
[6, 2, 5, 1, 9]         # exactly what we wanted

Here is the simple secret: if you complete f_ or ff_ with a function name and its arguments, and leave it unevaluated (not called), they will act as a deferred expression.

Not related to lazy operation, but you do the same thing with uncurry

# replicate the tuple of arguments (1, 10) and then apply to uncurried function
>>> map(u_(randint))(replicate(5, (1,10)))    # u_ == uncurry
[7, 6, 1, 7, 2]

Raise and assert with expressions: error and guard

Raise any kinds of exception in lambda expression as well.

error(MESSAGE, e=EXCEPTION_TO_RAISE)

>>> error("Error, used wrong type", e=TypeError)

>>> error("out of range", e=IndexError)

>>> (lambda x: x if x is not None else error("Error, got None", e=ValueError))(None)

Likewise, use guard if there need assertion not as a statement, but as an expression.

guard(PREDICATE, MESSAGE, e=EXCEPTION_TO_RAISE)

>>> guard("Almost" == "enough", "'Almost' is never 'enough'")

>>> guard(rand() > 0.5, "Assertion error occurs with a 0.5 probability")

>>> guard(len(x := range(11)) == 10, f"length is not 10: {len(x)}")

Exception catcher builder: trap

trap is a decorator factory that creates exception catchers. e indicates error types you want to catch, callback is a callback function to invoke with the catched error.
This is very useful when handling exceptions with a functional approach on a function-by-function basis

trap(callback, e=None)

This will catch ValueError and then print the error message.

>>> trap(print, e=ValueError)(error)(msg="occured a value-error", e=ValueError)
Occured a value-error

This will catch all kinds of errors, then count the length of the error message when calling func(*args, **kwargs).

trap(cf(len, str), e=None)(func)(*args, **kwargs)

This function will never throw errors. It return only None instead of raising exceptions.

@trap(callback=void, e=None)
def func(*args, **kwargs):
    ...

Utilities

Flatten iterables: flat and flatten

flat completely removes all nesting levels. (deep flatten)
flatten, on the other hand, reduces the nesting depth by the given level. (swallow flatten)
String-like iterables such as str, bytes, and bytearray are not flattened.

flat(*args)
flatl(*args)
flatten(ITERABLE, d=LEVEL)

>>> data = [1,2,[3,4,[[[5],6],7,{8},((9),10)],range(11,13)], (x for x in [13,14,15])]

>>> flat(data) | collect    
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]

>>> data = [1, [[2,{3}]], [[[[4]],5]], (('sofia','maria'),)]  
>>> flatten(data)  # by default, d=1
[1, [2, {3}], [[[4]], 5], ('sofia', 'maria')]

>>> flatten(data d=2) 
[1, 2, {3}, [[4]], 5, 'sofia', 'maria']

>>> flatl(data)    # flatl(data) == flat(data) | collect  
[1, 2, 3, 4, 5, 'sofia', 'maria']

Shell Command: shell

shell executes shell commands synchronosly and asynchronously and capture their outputs.

shell(CMD, sync=True, o=True, *, executable="/bin/bash")

  --------------------------------------------------------------------
    o-value  |  return |  meaning
  --------------------------------------------------------------------
     o =  1  |  [str]  |  captures stdout/stderr (2>&1)
     o = -1  |  None   |  discard (&>/dev/null)
  otherwise  |  None   |  do nothing or redirection (2>&1 or &>FILE)
  --------------------------------------------------------------------

shell performs the same operation as ipython's magic command !!. However, it can also be used within a python script.

>>> output = shell("ls -1 ~")    
>>> output = "ls -1 ~" | shell           # the same

>>> shell("find . | sort" o=-1)          # run and discard the result  
>>> "find . | sort" | shell(o=-1)    

>>> shell("cat *.md", o=writer(FILE))    # redirect to FILE
>>> "cat *.md" | shell(o=writer(FILE))   # redirect to FILE

Neatify data structures: neatly and nprint

neatly generates neatly formatted string of the complex data structures of dict and list.

nprint (neatly-print) prints data structures to stdout using neatly formatter.
nprint(...) = print(neatly(...))

nprint(DICT, _cols=INDENT, _width=WRAP, _repr=BOOL, **kwargs)

>>> o = {
...   "$id": "https://example.com/enumerated-values.schema.json",
...   "$schema": "https://json-schema.org/draft/2020-12/schema",
...   "title": "Enumerated Values",
...   "type": "object",
...   "properties": {
...     "data": {
...       "enum": [42, True, "hello", None, [1, 2, 3]]
...     }
...   }
... }

>>> nprint(o)
       $id  |  'https://example.com/enumerated-values.schema.json'
   $schema  |  'https://json-schema.org/draft/2020-12/schema'
properties  |  data  |  enum  +  42
            :        :        -  True
            :        :        -  'hello'
            :        :        -  None
            :        :        -  +  1
            :        :           -  2
            :        :           -  3
     title  |  'Enumerated Values'
      type  |  'object'

Dot-accessible dictionary: dmap

dmap is a yet another dict. It's exactly the same as dict but it enables to access its nested structure with 'dot notations'.

dmap(DICT, **kwargs)

>>> d = dmap()    # empty dict

>>> o = dict(name="yunchan lim", age=19)
>>> d = dmap(o, profession="pianist")    

>>> d = dmap(name="yunchan lim", age=19, profession="pianist")    # the same

# just put the value in the desired keypath
>>> d.cliburn.semifinal.mozart = "piano concerto no.22"
>>> d.cliburn.semifinal.liszt = "12 transcendental etudes"
>>> d.cliburn.final.beethoven = "piano concerto no.3"
>>> d.cliburn.final.rachmaninoff = "piano concerto no.3"
>>> nprint(d)
       age  |  19
   cliburn  |      final  |     beethoven  |  'piano concerto no.3'
            :             :  rachmaninoff  |  'piano concerto no.3'
            :  semifinal  |   liszt  |  '12 transcendental etudes'
            :             :  mozart  |  'piano concerto no.22'
      name  |  'yunchan lim'
profession  |  'pianist'

>>> del d.cliburn.semifinal
>>> d.profession = "one-in-a-million talent"
>>> nprint(d)
       age  |  19
   cliburn  |  final  |     beethoven  |  'piano concerto no.3'
            :         :  rachmaninoff  |  'piano concerto no.3'
      name  |  'yunchan lim'
profession  |  'one-in-a-million talent'

# No such keypath
>>> d.bach.chopin.beethoven
{}

Handy File Tools: ls and grep

Use ls and grep in the same way you use in your terminal every day.
This is just a more intuitive alternative to os.listdir and os.walk. When applicable, use shell instead.

ls(*paths, grep=REGEX, i=BOOL, r=BOOL, f=BOOL, d=BOOL, g=BOOL)

# couldn't be simpler!
>>> ls()       # the same as ls("."): get contents of the curruent dir

# expands "~" automatically
>>> ls("~")    # the same as `ls -a1 ~`: returns a list of $HOME

# support glob patterns (*, ?, [)
>>> ls("./*/*.py")

# with multiple filepaths
>>> ls(FILE, DIR, ...)

# list up recursively and filter hidden files out
>>> ls(".git", r=True, grep="^[^\.]")

# only files in '.git' directory
>>> ls(".git", r=True, f=True)

# only directories in '.git' directory
>>> ls(".git", r=True, d=True)

# search recursivley and matching a pattern with `grep`
>>> ls(".", r=True, i=True, grep=".Py")    # 'i=True' for case-insensitive grep pattern

[ ..
 '.pytest_cache/v/cache/stepwise',
 'foc/__init__.py',
 'foc/__pycache__/__init__.cpython-310.pyc',
 'tests/__init__.py',
.. ]

# regex patterns come in
>>> ls(".", r=True, grep=".py$")

['foc/__init__.py', 'setup.py', 'tests/__init__.py', 'tests/test_foc.py']

# that's it!
>>> ls(".", r=True, grep="^(foc).*py$")

# the same as above
>>> ls("foc/*.py")

['foc/__init__.py']

grep build a filter to select items matching REGEX pattern from iterables.

grep(REGEX, i=BOOL)

# 'grep' builds filter with regex patterns
>>> grep(r"^(foc).*py$")(ls(".", r=True))

['foc/__init__.py']

See also: HOME, cd, pwd, mkdir, rmdir, exists, dirname, and basename.

Flexible Progress Bar: taskbar

taskbar makes it easy to do progress bar related tasks. Acutally taskbar is the same as the rich.progress except for below:

• No install required

  taskbar use pip's bundle. pip is already installed almost everywhere.

• Fixed to default tqdm-like bar style
• Simplified further the rich.progress's usage

taskbar(x=None, desc="working", *, start=0, total=None, barcolor="white", **kwargs)
See also: rich.progress.Progress(.., **kwargs)

# simply with iterables (or generators with 'total=LENGTH')
>>> for _ in taskbar(range(100), "[cyan] training model"):
...    ...   
 training model  100%  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  100/100  0:00:20 < 0:00:00  4.94 it/s

# when staring in the middle of a progress
>>> for _ in taskbar(range(100), "[cyan] training model", start=30):
...     ... 

# manual update with multiple tasks
>>> with taskbar() as tb:
...     task1 = tb.add_task("[red] fine-tuning", total=1000)
...     task2 = tb.add_task("[green] train-critic", total=1000)
...     task3 = tb.add_task("[cyan] reinforce", total=1000)
...     while not tb.finished:
...         ...
...         tb.update(task1, advance=0.9)
...         tb.update(task2, advance=0.5)
...         tb.update(task3, advance=0.1)
...
 fine-tuning   18%  ━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━   178/1000  0:00:20 < 0:01:34  8.79 it/s
 train-critic  10%  ━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━    99/1000  0:00:20 < 0:03:05  4.88 it/s
 reinforce      6%  ━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━    59/1000  0:00:20 < 0:05:22  2.93 it/s