cythonpowered 0.2.2

Cython-powered replacements for popular Python functions. And more.

Cython-powered replacements for popular Python functions. And more.

cythonpowered is a library containing replacements for various Python functions, that are generated with Cython and compiled at setup, intended to provide performance gains for developers.

Some functions are drop-in replacements, others are provided to enhance certain usages of the respective functions, or have slightly different implementations.

Installation

pip install cythonpowered

Usage

Simply import the desired function and use it in your Python code.

Run cythonpowered --list to view all available functions and their Python conunterparts.

Currently available functions:

               _   _                                                      _ 
     ___ _   _| |_| |__   ___  _ __  _ __   _____      _____ _ __ ___  __| |
    / __| | | | __| '_ \ / _ \| '_ \| '_ \ / _ \ \ /\ / / _ \ '__/ _ \/ _` |
   | (__| |_| | |_| | | | (_) | | | | |_) | (_) \ V  V /  __/ | |  __/ (_| |
    \___|\__, |\__|_| |_|\___/|_| |_| .__/ \___/ \_/\_/ \___|_|  \___|\__,_|
         |___/                      |_|                                     
                                                                  ver. 0.2.2

+----+------------------------------------------+-------------------------------------------+---------------------------------------------------------------+
|  # | Python function                          | Is replaced by                            | Usage / details                                               |
+----+------------------------------------------+-------------------------------------------+---------------------------------------------------------------+
|  1 | random.random()                          | cythonpowered.random.random()             | Drop-in replacement                                           |
|  2 | random.random()                          | cythonpowered.random.n_random()           | n_random(k) replaces [random() for i in range(k)]             |
|  3 | random.randint()                         | cythonpowered.random.randint()            | Drop-in replacement                                           |
|  4 | random.randint()                         | cythonpowered.random.n_randint()          | n_randint(a, b, k) replaces [randint(a, b) for i in range(k)] |
|  5 | random.uniform()                         | cythonpowered.random.uniform()            | Drop-in replacement                                           |
|  6 | random.uniform()                         | cythonpowered.random.n_uniform()          | n_uniform(a, b, k) replaces [uniform(a, b) for i in range(k)] |
|  7 | random.choice()                          | cythonpowered.random.choice()             | Drop-in replacement                                           |
|  8 | random.choices()                         | cythonpowered.random.choices()            | Drop-in replacement, only supports the 'k' keyword argument   |
+----+------------------------------------------+-------------------------------------------+---------------------------------------------------------------+
|  9 | datetime.date.today()                    | cythonpowered.dateutil.date.today()       | Drop-in replacement                                           |
| 10 | calendar.isleap()                        | cythonpowered.dateutil.date.isleap()      | Drop-in replacement                                           |
| 11 | calendar.monthrange()                    | cythonpowered.dateutil.date.monthrange()  | Drop-in replacement                                           |
| 12 | datetime.datetime.strptime().date()      | cythonpowered.dateutil.date.fromstring()  | Assumes '%Y-%m-%d' format, returns cythonpowered date object  |
| 13 | datetime.date().strftime()               | cythonpowered.dateutil.date().tostring()  | Assumes '%Y-%m-%d' format, uses cythonpowered date object     |
| 14 | datetime.date().weekday()                | cythonpowered.dateutil.date().weekday()   | Drop-in replacement, uses cythonpowered date object           |
| 15 | datetime.date().timetuple().tm_yday      | cythonpowered.dateutil.date().yearday()   | Drop-in replacement, uses cythonpowered date object           |
| 16 | datetime.date.fromordinal()              | cythonpowered.dateutil.date.fromordinal() | Drop-in replacement, returns cythonpowered date object        |
| 17 | datetime.date().toordinal()              | cythonpowered.dateutil.date().toordinal() | Drop-in replacement, uses cythonpowered date object           |
| 18 | datetime.date() +/- datetime.timedelta() | cythonpowered.dateutil.date().offset()    | Supports only days offset, returns cythonpowered date object  |
+----+------------------------------------------+-------------------------------------------+---------------------------------------------------------------+

Benchmark

Run cythonpowered --benchmark o view the performance gains on your system for all cythonpowered functions, compared to their Python counterparts.

Example benchmark output:

               _   _                                                      _ 
     ___ _   _| |_| |__   ___  _ __  _ __   _____      _____ _ __ ___  __| |
    / __| | | | __| '_ \ / _ \| '_ \| '_ \ / _ \ \ /\ / / _ \ '__/ _ \/ _` |
   | (__| |_| | |_| | | | (_) | | | | |_) | (_) \ V  V /  __/ | |  __/ (_| |
    \___|\__, |\__|_| |_|\___/|_| |_| .__/ \___/ \_/\_/ \___|_|  \___|\__,_|
         |___/                      |_|                                     
                                                                  ver. 0.2.2

CPU model:             11th Gen Intel(R) Core(TM) i7-11370H @ 3.30GHz
CPU base frequency:    3.3000 GHz
CPU cores:             4
CPU threads:           4
Architecture:          x86_64
Memory (RAM):          15.31 GB
Operating System:      Linux 6.8.0-100-generic
Python version:        3.12.3
C compiler:            GCC 13.3.0

================================================================================
Running benchmark for the [cythonpowered.random] module (5 benchmarks)...
================================================================================
Comparing Python random.random() with cythonpowered alternative(s)... 100.00%
Comparing Python random.randint() with cythonpowered alternative(s)... 100.00%
Comparing Python random.uniform() with cythonpowered alternative(s)... 100.00%
Comparing Python random.choice() with cythonpowered alternative(s)... 100.00%
Comparing Python random.choices() with cythonpowered alternative(s)... 100.00%
+----------------------------------+-----------------+--------------------+--------------+
|          Function name           |   No. of runs   |    Speed factor    | Avg. speedup |
+----------------------------------+-----------------+--------------------+--------------+
|     [Python] random.random()     | [10K, 100K, 1M] |        1.00        |     1.00     |
|  cythonpowered.random.random()   | [10K, 100K, 1M] | [1.05, 1.00, 1.02] |     1.03     |
| cythonpowered.random.n_random()  | [10K, 100K, 1M] | [3.11, 3.16, 3.13] |     3.13     |
+----------------------------------+-----------------+--------------------+--------------+
|    [Python] random.randint()     | [10K, 100K, 1M] |        1.00        |     1.00     |
|  cythonpowered.random.randint()  | [10K, 100K, 1M] | [5.26, 4.62, 4.50] |     4.79     |
| cythonpowered.random.n_randint() | [10K, 100K, 1M] | [25.1, 16.7, 15.6] |     19.1     |
+----------------------------------+-----------------+--------------------+--------------+
|    [Python] random.uniform()     | [10K, 100K, 1M] |        1.00        |     1.00     |
|  cythonpowered.random.uniform()  | [10K, 100K, 1M] | [2.33, 1.94, 1.97] |     2.08     |
| cythonpowered.random.n_uniform() | [10K, 100K, 1M] | [12.4, 7.61, 7.66] |     9.23     |
+----------------------------------+-----------------+--------------------+--------------+
|     [Python] random.choice()     | [10K, 100K, 1M] |        1.00        |     1.00     |
|  cythonpowered.random.choice()   | [10K, 100K, 1M] | [4.99, 4.55, 4.51] |     4.68     |
+----------------------------------+-----------------+--------------------+--------------+
|    [Python] random.choices()     | [1K, 10K, 100K] |        1.00        |     1.00     |
|  cythonpowered.random.choices()  | [1K, 10K, 100K] | [3.27, 2.23, 2.18] |     2.56     |
+----------------------------------+-----------------+--------------------+--------------+

================================================================================
Running benchmark for the [cythonpowered.dateutil] module (10 benchmarks)...
================================================================================
Comparing Python datetime.date.today() with cythonpowered alternative(s)... 100.00%
Comparing Python calendar.isleap() with cythonpowered alternative(s)... 100.00%
Comparing Python calendar.monthrange() with cythonpowered alternative(s)... 100.00%
Comparing Python datetime.datetime.strptime().date() with cythonpowered alternative(s)... 100.00%
Comparing Python datetime.date().strftime() with cythonpowered alternative(s)... 100.00%
Comparing Python datetime.date().weekday() with cythonpowered alternative(s)... 100.00%
Comparing Python datetime.date().timetuple().tm_yday with cythonpowered alternative(s)... 100.00%
Comparing Python datetime.date.fromordinal() with cythonpowered alternative(s)... 100.00%
Comparing Python datetime.date().toordinal() with cythonpowered alternative(s)... 100.00%
Comparing Python datetime.date() +/- datetime.timedelta() with cythonpowered alternative(s)... 100.00%
+---------------------------------------------------+-----------------+--------------------+--------------+
|                   Function name                   |   No. of runs   |    Speed factor    | Avg. speedup |
+---------------------------------------------------+-----------------+--------------------+--------------+
|           [Python] datetime.date.today()          | [10K, 100K, 1M] |        1.00        |     1.00     |
|        cythonpowered.dateutil.date.today()        | [10K, 100K, 1M] | [1.86, 1.78, 1.89] |     1.84     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|             [Python] calendar.isleap()            | [10K, 100K, 1M] |        1.00        |     1.00     |
|        cythonpowered.dateutil.date.isleap()       | [10K, 100K, 1M] | [1.79, 1.78, 1.76] |     1.78     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|           [Python] calendar.monthrange()          | [10K, 100K, 1M] |        1.00        |     1.00     |
|      cythonpowered.dateutil.date.monthrange()     | [10K, 100K, 1M] | [3.23, 2.33, 4.25] |     3.27     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|    [Python] datetime.datetime.strptime().date()   | [10K, 100K, 1M] |        1.00        |     1.00     |
|      cythonpowered.dateutil.date.fromstring()     | [10K, 100K, 1M] | [10.1, 9.60, 9.56] |     9.75     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|        [Python] datetime.date().strftime()        | [10K, 100K, 1M] |        1.00        |     1.00     |
|      cythonpowered.dateutil.date().tostring()     | [10K, 100K, 1M] | [4.66, 4.49, 4.35] |     4.50     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|         [Python] datetime.date().weekday()        | [10K, 100K, 1M] |        1.00        |     1.00     |
|      cythonpowered.dateutil.date().weekday()      | [10K, 100K, 1M] | [1.03, 0.99, 0.87] |     0.96     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|    [Python] datetime.date().timetuple().tm_yday   | [10K, 100K, 1M] |        1.00        |     1.00     |
|      cythonpowered.dateutil.date().yearday()      | [10K, 100K, 1M] | [15.2, 6.67, 11.3] |     11.1     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|        [Python] datetime.date.fromordinal()       | [10K, 100K, 1M] |        1.00        |     1.00     |
|     cythonpowered.dateutil.date.fromordinal()     | [10K, 100K, 1M] | [0.65, 0.65, 0.68] |     0.66     |
+---------------------------------------------------+-----------------+--------------------+--------------+
|        [Python] datetime.date().toordinal()       | [10K, 100K, 1M] |        1.00        |     1.00     |
|     cythonpowered.dateutil.date().toordinal()     | [10K, 100K, 1M] | [0.58, 0.58, 0.55] |     0.57     |
+---------------------------------------------------+-----------------+--------------------+--------------+
| [Python] datetime.date() +/- datetime.timedelta() | [10K, 100K, 1M] |        1.00        |     1.00     |
|       cythonpowered.dateutil.date().offset()      | [10K, 100K, 1M] | [2.00, 2.01, 1.96] |     1.99     |
+---------------------------------------------------+-----------------+--------------------+--------------+

---