Regex Performansını Arttırma: Compiled Patterns Kullanımı

1import re
2import time
3
4# Derlenmiş desen
5pattern = re.compile(r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b')
6
7# Test verisi: 10000 satır
8text_lines = ['user@example.com test data'] * 10000
9
10# Derlenmiş desen ile arama
11start = time.time()
12for line in text_lines:
13pattern.findall(line)
14compiled_time = time.time() - start
15
16# Her seferinde desen derlemek (YABANCILAŞTIRMA)
17start = time.time()
18for line in text_lines:
19re.findall(r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b', line)
20uncompiled_time = time.time() - start
21
22print(f"Derlenmiş: {compiled_time:.4f}s")
23print(f"Derlemesiz: {uncompiled_time:.4f}s")
24print(f"Hız artışı: {uncompiled_time / compiled_time:.1f}x")

1Derlenmiş: 0.0245s
2Derlemesiz: 0.2456s
3Hız artışı: 10.0x

1using System;
2using System.Text.RegularExpressions;
3using System.Diagnostics;
4
5var pattern = new Regex(@"\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b", RegexOptions.Compiled);
6
7var sw = Stopwatch.StartNew();
8for (int i = 0; i < 10000; i++)
9{
10pattern.Matches("user@example.com test data");
11}
12sw.Stop();
13
14Console.WriteLine($"Süre: {sw.ElapsedMilliseconds}ms");

1Süre: 45ms