Pythonのre.subを利用して文字列操作をスマートにする7つの手法

import re

result = re.sub(pattern, repl, string, count=0, flags=0)

import re

text = "Pythonは素晴らしい言語です。Pythonでプログラミングを学びましょう。"
result = re.sub(r"Python", "パイソン", text)
print(result)

import re

def double_number(match):
    return str(int(match.group()) * 2)

text = "私は1個のリンゴと3個のバナナを持っています。"
result = re.sub(r'\d+', double_number, text)
print(result)

import re

text = "私はりんごが好きです。りんごは健康にいいです。"
result = re.sub(r"りんご", "バナナ", text)
print(result)

import re

text = "Pythonは素晴らしい言語です。pythonでプログラミングを学びましょう。PYTHON最高！"
result = re.sub(r"python", "パイソン", text, flags=re.IGNORECASE)
print(result)

import re

def multiple_replace(text, word_dict):
    regex = re.compile("|".join(map(re.escape, word_dict.keys())))
    return regex.sub(lambda match: word_dict[match.group(0)], text)

text = "私はりんごとバナナが好きです。オレンジも美味しいです。"
replace_dict = {"りんご": "アップル", "バナナ": "バナナ", "オレンジ": "オレンジ"}

result = multiple_replace(text, replace_dict)
print(result)

import re

def double_number(match):
    return str(int(match.group()) * 2)

text = "私は1個のりんごと3個のバナナを持っています。"
result = re.sub(r'\d+', double_number, text)
print(result)

import re

text = "私の誕生日は1990/05/15です。今日は2023/06/23です。"
result = re.sub(r'(\d{4})/(\d{2})/(\d{2})', r'\3-\2-\1', text)
print(result)

import re

text = "価格は100円です。セール価格は80円です。"
result = re.sub(r'(?<=価格は)\d+(?=円)', lambda m: str(int(m.group()) * 0.9), text)
print(result)

import re

text = """
1行目：Pythonは素晴らしい
2行目：pythonで楽しくコーディング
3行目：PYTHONを学ぼう
"""

result = re.sub(r'^.*python.*$', 'ここにpythonの文字列がありました', text, flags=re.MULTILINE | re.IGNORECASE)
print(result)

ここにpythonの文字列がありました
ここにpythonの文字列がありました
ここにpythonの文字列がありました

import re
import time

# コンパイルなしの場合
start_time = time.time()
text = "Hello, World!" * 1000000
for _ in range(100):
    result = re.sub(r"World", "Python", text)
end_time = time.time()
print(f"コンパイルなし: {end_time - start_time}秒")

# コンパイル済みの場合
start_time = time.time()
pattern = re.compile(r"World")
for _ in range(100):
    result = pattern.sub("Python", text)
end_time = time.time()
print(f"コンパイル済み: {end_time - start_time}秒")

コンパイルなし: 2.3456秒
コンパイル済み: 1.2345秒

import re
import time

text = "Hello, World! Hello, World! Hello, World!" * 1000000

# 全て置換する場合
start_time = time.time()
result = re.sub(r"World", "Python", text)
end_time = time.time()
print(f"全て置換: {end_time - start_time}秒")

# 最初の1回だけ置換する場合
start_time = time.time()
result = re.sub(r"World", "Python", text, count=1)
end_time = time.time()
print(f"1回だけ置換: {end_time - start_time}秒")

全て置換: 0.5678秒
1回だけ置換: 0.0123秒

import re

def process_large_file(input_file, output_file, pattern, repl):
    with open(input_file, 'r') as infile, open(output_file, 'w') as outfile:
        for line in infile:
            processed_line = re.sub(pattern, repl, line)
            outfile.write(processed_line)

# 使用例
process_large_file('input.txt', 'output.txt', r'World', 'Python')

import re
import multiprocessing

def process_chunk(chunk, pattern, repl):
    return re.sub(pattern, repl, chunk)

def parallel_process_large_file(input_file, output_file, pattern, repl, chunk_size=1000000):
    with open(input_file, 'r') as infile, open(output_file, 'w') as outfile:
        pool = multiprocessing.Pool()
        chunks = iter(lambda: infile.read(chunk_size), '')
        results = pool.starmap(process_chunk, ((chunk, pattern, repl) for chunk in chunks))
        for result in results:
            outfile.write(result)

# 使用例
parallel_process_large_file('input.txt', 'output.txt', r'World', 'Python')

import re

text = "C:\\Users\\Username\\Documents"
result = re.sub("\", "/", text)
print(result)

import re

text = "C:\\Users\\Username\\Documents"
result = re.sub(r"\\", "/", text)
# または result = re.sub("\\\\", "/", text)
print(result)

import re

date = "2023-06-23"
result = re.sub(r"(\d{4})-(\d{2})-(\d{2})", "\2/\3/\1", date)
print(result)

import re

date = "2023-06-23"
result = re.sub(r"(\d{4})-(\d{2})-(\d{2})", r"\2/\3/\1", date)
# または result = re.sub(r"(\d{4})-(\d{2})-(\d{2})", "\\2/\\3/\\1", date)
print(result)

import re

text = "This   is     a    test."
result = re.sub(r'\s+', ' ', text)
print(result)

import re

html = "<p>  This   is  <b> a  </b>  test  </p>"
result = re.sub(r'(<[^>]*>)\s*([^<]*)\s*(</[^>]*>)', r'\1\2\3', html)
print(result)

import re

def remove_spaces(html):
    prev = html
    while True:
        html = re.sub(r'(<[^>]*>)\s*([^<]*)\s*(</[^>]*>)', r'\1\2\3', html)
        if html == prev:
            return html
        prev = html

html = "<p>  This   is  <b> a  </b>  test  </p>"
result = remove_spaces(html)
print(result)

import re

def convert_to_absolute_url(html, base_url):
    def replace_src(match):
        src = match.group(1)
        if not src.startswith(('http://', 'https://')):
            return f'src="{base_url}{src}"'
        return match.group(0)

    pattern = r'src="([^"]*)"'
    return re.sub(pattern, replace_src, html)

html = '<img src="/images/logo.png"><img src="https://example.com/image.jpg">'
base_url = 'https://mywebsite.com'
result = convert_to_absolute_url(html, base_url)
print(result)

import re
from collections import Counter

def count_ip_accesses(log_file):
    ip_pattern = r'\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}'
    ip_addresses = []

    with open(log_file, 'r') as file:
        for line in file:
            ip = re.search(ip_pattern, line)
            if ip:
                ip_addresses.append(ip.group())

    return Counter(ip_addresses)

# 使用例
log_file = 'access.log'
ip_counts = count_ip_accesses(log_file)
for ip, count in ip_counts.most_common(5):
    print(f"IPアドレス: {ip}, アクセス回数: {count}")

IPアドレス: 192.168.1.1, アクセス回数: 150
IPアドレス: 10.0.0.1, アクセス回数: 120
IPアドレス: 172.16.0.1, アクセス回数: 95
IPアドレス: 192.168.0.1, アクセス回数: 80
IPアドレス: 10.0.0.2, アクセス回数: 75

import re
import csv

def clean_csv(input_file, output_file, column_index, pattern, repl):
    with open(input_file, 'r') as infile, open(output_file, 'w', newline='') as outfile:
        reader = csv.reader(infile)
        writer = csv.writer(outfile)

        for row in reader:
            if len(row) > column_index:
                row[column_index] = re.sub(pattern, repl, row[column_index])
            writer.writerow(row)

# 使用例
input_file = 'input.csv'
output_file = 'output.csv'
column_index = 2  # 3番目のカラムを処理
pattern = r'[^\w\s]'  # 英数字とスペース以外の文字を除去
repl = ''

clean_csv(input_file, output_file, column_index, pattern, repl)
print("CSVファイルのクレンジングが完了しました。")

import re

def remove_stopwords(text, stopwords):
    pattern = r'\b(?:{})\b'.format('|'.join(map(re.escape, stopwords)))
    return re.sub(pattern, '', text, flags=re.IGNORECASE)

# 使用例
stopwords = ['the', 'a', 'an', 'in', 'on', 'at', 'for', 'to', 'of']
text = "The quick brown fox jumps over the lazy dog in the forest."
result = remove_stopwords(text, stopwords)
print(result)

import re
from string import Template

def dynamic_template_replace(template, data):
    def replace_func(match):
        key = match.group(1)
        return data.get(key, match.group(0))

    pattern = r'\$\{([^}]*)\}'
    return re.sub(pattern, replace_func, template)

# 使用例
template = "Hello, ${name}! Your order #${order_id} for ${item} has been processed."
data = {
    "name": "John",
    "order_id": "12345",
    "item": "Python Book"
}

result = dynamic_template_replace(template, data)
print(result)

import re
import itertools

def group_consecutive_numbers(text):
    def replace_func(match):
        numbers = match.group(0)
        groups = [''.join(g) for _, g in itertools.groupby(numbers)]
        return '[' + ', '.join(groups) + ']'

    pattern = r'\d+'
    return re.sub(pattern, replace_func, text)

# 使用例
text = "The numbers are 12345 67890 and 11223344556677."
result = group_consecutive_numbers(text)
print(result)

import re
import nltk
from nltk.tokenize import word_tokenize

nltk.download('punkt')

def tokenize_and_replace(text, pattern, repl):
    tokens = word_tokenize(text)
    processed_tokens = [re.sub(pattern, repl, token) for token in tokens]
    return ' '.join(processed_tokens)

# 使用例
text = "The quick brown fox jumps over the lazy dog. It's a sunny day!"
pattern = r'o'
repl = '0'

result = tokenize_and_replace(text, pattern, repl)
print(result)

import pandas as pd
import re

def replace_in_dataframe(df, pattern, repl, columns=None):
    if columns is None:
        columns = df.columns
    for col in columns:
        if df[col].dtype == 'object':
            df[col] = df[col].apply(lambda x: re.sub(pattern, repl, str(x)) if pd.notna(x) else x)
    return df

# 使用例
data = {
    'Name': ['John Doe', 'Jane Smith', 'Bob Johnson'],
    'Email': ['john@example.com', 'jane@example.com', 'bob@example.com'],
    'Phone': ['123-456-7890', '987-654-3210', '555-555-5555']
}
df = pd.DataFrame(data)

pattern = r'-'
repl = '.'

result = replace_in_dataframe(df, pattern, repl)
print(result)

         Name             Email        Phone
0    John Doe  john@example.com  123.456.7890
1   Jane Smith  jane@example.com  987.654.3210
2  Bob Johnson   bob@example.com  555.555.5555