Create bm_regex_generic.py

pyperformance/benchmarks/bm_regex_generic.py

@@ -0,0 +1,114 @@
+
+"""Benchmarks for Python's regex engine.
+
+Benchmarks for checking begining and end of strings. Should have reasonable times after https://bugs.python.org/issue42885 is fixed.
+
+Real life use case is e.g. comparing 500.000 filenames and paths each with thousands of regex of known malicous files, https://github.com/Neo23x0/signature-base/blob/master/iocs/filename-iocs.txt
+(^ and $ mostly not yet included because it wouldn't give a speed advantage in https://github.com/Neo23x0/Loki)
+
+Adopted from bm_regex_effbot.py by Arnim Rupp
+
+Feel free to add other generic regex.
+
+"""
+
+# Python imports
+import re
+
+# Local imports
+import pyperf
+
+# These are the regular expressions to be tested. 
+
+def gen_regex_table():
+    return [
+        re.compile('^c:'), # very slow if false
+        re.compile('\Ac:'), # very slow if false
+        re.compile('evil\.exe$'),
+        re.compile('evil\.exe\Z'),
+        re.compile('\Ad:'),
+        re.compile('evil\.dll$'), # very slow if true
+        re.compile('evil\.dll\Z'), # very slow if true
+        ]
+
+
+def gen_string_table(n):
+    """Generates the list of strings that will be used in the benchmarks.
+
+    """
+    strings = []
+
+    strings.append('d:' + ('A' * ( 10 ** n )) + 'evil.dll')
+    return strings
+
+
+def init_benchmarks(n_values=None):
+    """Initialize the strings we'll run the regexes against.
+
+    The generated list of strings is cached in the string_tables
+    variable, which is indexed by n.
+
+    Returns:
+    A list of string + lengths.
+    """
+
+    if n_values is None:
+        n_values = (3,5,7)
+
+    string_tables = {n: gen_string_table(n) for n in n_values}
+    #print(string_tables)
+    regexs = gen_regex_table()
+
+    data = []
+    for n in n_values:
+        for id in range(len(regexs)):
+            regex = regexs[id]
+            string = string_tables[n]
+            data.append((regex, string))
+    return data
+
+
+def bench_regex_generic(loops):
+    if bench_regex_generic.data is None:
+        bench_regex_generic.data = init_benchmarks()
+    data = bench_regex_generic.data
+
+    range_it = range(loops)
+    search = re.search
+    t0 = pyperf.perf_counter()
+
+    for _ in range_it:
+        # Runs all of the benchmarks for a given value of n.
+        for regex, string in data:
+            # search 10 times
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+            search(regex, string[0])
+
+    return pyperf.perf_counter() - t0
+
+
+# cached data, generated at the first call
+bench_regex_generic.data = None
+
+
+def add_cmdline_args(cmd, args):
+    pass
+    #if args.force_bytes:
+        #cmd.append("--force_bytes")
+
+
+if __name__ == '__main__':
+    runner = pyperf.Runner(add_cmdline_args=add_cmdline_args)
+    runner.metadata['description'] = ("Test the performance of regexps "
+                                      "with long strings.")
+    options = runner.parse_args()
+
+    runner.bench_time_func('regex_generic', bench_regex_generic, inner_loops=10)