Add benchmark for deepcopy (#201)

* Add benchmark for deepcopy

* fix typo

* Update pyperformance/data-files/benchmarks/bm_deepcopy/run_benchmark.py

Co-authored-by: Eric Snow <[email protected]>

* Update pyperformance/data-files/benchmarks/bm_deepcopy/run_benchmark.py

Co-authored-by: Eric Snow <[email protected]>

* address review comments

* Update pyperformance/data-files/benchmarks/bm_deepcopy/run_benchmark.py

Co-authored-by: Eric Snow <[email protected]>

Co-authored-by: Eric Snow <[email protected]>

Author: eendebakpt

doc/benchmarks.rst

@@ -124,6 +124,11 @@ See `pyaes <https://github.com/ricmoo/pyaes>`_: A pure-Python implementation of
 the AES block cipher algorithm and the common modes of operation (CBC, CFB,
 CTR, ECB and OFB).
 
+deepcopy
+--------
+
+Benchmark the Python `copy.deepcopy` method. The `deepcopy` method is
+performed on a nested dictionary and a dataclass.
 
 deltablue
 ---------

pyperformance/data-files/benchmarks/MANIFEST

@@ -11,6 +11,7 @@ generators	<local>
 chameleon	<local>
 chaos	<local>
 crypto_pyaes	<local>
+deepcopy	<local>
 deltablue	<local>
 django_template	<local>
 dulwich_log	<local>

pyperformance/data-files/benchmarks/bm_deepcopy/pyproject.toml

@@ -0,0 +1,9 @@
+[project]
+name = "pyperformance_bm_deepcopy"
+requires-python = ">=3.8"
+dependencies = ["pyperf"]
+urls = {repository = "https://github.com/python/pyperformance"}
+dynamic = ["version"]
+
+[tool.pyperformance]
+name = "deepcopy"

pyperformance/data-files/benchmarks/bm_deepcopy/run_benchmark.py

@@ -0,0 +1,88 @@
+"""
+Benchmark to measure performance of the python builtin method copy.deepcopy
+
+Performance is tested on a nested dictionary and a dataclass
+
+Author: Pieter Eendebak
+
+"""
+import copy
+import pyperf
+from dataclasses import dataclass
+
+
+@dataclass
+class A:
+    string: str
+    lst: list
+    boolean: bool
+
+
+def benchmark_reduce(n):
+    """ Benchmark where the __reduce__ functionality is used """
+    class C(object):
+        def __init__(self):
+            self.a = 1
+            self.b = 2
+
+        def __reduce__(self):
+            return (C, (), self.__dict__)
+
+        def __setstate__(self, state):
+            self.__dict__.update(state)
+    c = C()
+
+    t0 = pyperf.perf_counter()
+    for ii in range(n):
+        _ = copy.deepcopy(c)
+    dt = pyperf.perf_counter() - t0
+    return dt
+
+
+def benchmark_memo(n):
+    """ Benchmark where the memo functionality is used """
+    A = [1] * 100
+    data = {'a': (A, A, A), 'b': [A] * 100}
+
+    t0 = pyperf.perf_counter()
+    for ii in range(n):
+        _ = copy.deepcopy(data)
+    dt = pyperf.perf_counter() - t0
+    return dt
+
+
+def benchmark(n):
+    """ Benchmark on some standard data types """
+    a = {
+        'list': [1, 2, 3, 43],
+        't': (1 ,2, 3),
+        'str': 'hello',
+        'subdict': {'a': True}
+    }
+    dc = A('hello', [1, 2, 3], True)
+
+    dt = 0
+    for ii in range(n):
+        for jj in range(30):
+            t0 = pyperf.perf_counter()
+            _ = copy.deepcopy(a)
+            dt += pyperf.perf_counter() - t0
+        for s in ['red', 'blue', 'green']:
+            dc.string = s
+            for kk in range(5):
+                dc.lst[0] = kk
+                for b in [True, False]:
+                    dc.boolean = b
+                    t0 = pyperf.perf_counter()
+                    _ = copy.deepcopy(dc)
+                    dt += pyperf.perf_counter() - t0
+    return dt
+
+
+if __name__ == "__main__":
+    runner = pyperf.Runner()
+    runner.metadata['description'] = "deepcopy benchmark"
+
+    runner.bench_time_func('deepcopy', benchmark)
+    runner.bench_time_func('deepcopy_reduce', benchmark_reduce)
+    runner.bench_time_func('deepcopy_memo', benchmark_memo)