str.translateで複数文字を一括置換する方法とreplaceの違い

# 基本: 同時に複数置換＋削除
s = "ABC-123-XYZ"

# dictで置換マップを記述。削除は None を指定します。
table = str.maketrans({
    "A": "@",
    "B": "8",
    "-": None,  # ハイフンは削除
})

result = s.translate(table)
print(result)

# 部分文字列の置換を試みる(誤りの例)
try:
    table = str.maketrans({"th": "X"})  # キーが2文字なのでNG
except Exception as e:
    print(type(e).__name__, ":", e)

# "abc" を "ABC" へ対応させる
table = str.maketrans("abc", "ABC")
print("abracadabra".translate(table))

# 丸数字や記号を通常の数字・記号へ
table = str.maketrans({
    ord("①"): "1",
    ord("②"): "2",
    ord("③"): "3",
    ord("—"): "-",   # エムダッシュをハイフンへ
})
print("得点①—③を選択".translate(table))

# 数字を全角→半角に、かつカンマと空白は削除
table = str.maketrans("０１２３４５６７８９", "0123456789", ", ")
print("１,２ ３".translate(table))

table = str.maketrans({"ß": "ss", "æ": "ae"})
print("straße • encyclopædia".translate(table))

s = "ABABA"

# replaceの連鎖(順序依存)
r1 = s.replace("A", "B").replace("B", "A")

# translateは同時置換なので意図通りに入れ替わる
table = str.maketrans({"A": "B", "B": "A"})
r2 = s.translate(table)

print("replace:", r1)
print("translate:", r2)

s = "spam spam spam"
print(s.replace("spam", "eggs", 1))  # 最初の1回だけ置換

import timeit

text = "A-B_C:D;E " * 10000
table = str.maketrans("", "", "-_:; ")

def via_translate():
    return text.translate(table)

def via_replace_chain():
    return text.replace("-", "").replace("_", "").replace(":", "").replace(";", "").replace(" ", "")

# ベンチマーク
t1 = timeit.timeit(via_translate, number=50)
t2 = timeit.timeit(via_replace_chain, number=50)
print("translate    :", t1)
print("replace chain:", t2)

import re
s = "a   b\t\tc\n  d"
print(re.sub(r"\s+", " ", s).strip())

text = '“Python”—それは「汎用」言語。価格は１，２３４円。メールはfoo—bar@example.com'
# 代表的なUnicode記号をASCIIへ正規化
table = str.maketrans({
    "“": '"', "”": '"',
    "‘": "'", "’": "'",
    "—": "-", "–": "-", "−": "-",  # ダッシュ/マイナスもハイフンへ
    "〜": "~", "～": "~",
    "：": ":", "；": ";",
    "（": "(", "）": ")",
    "［": "[", "］": "]",
    "｛": "{", "｝": "}",
    "，": ",", "。": ".",
    "１": "1", "２": "2", "３": "3", "４": "4",
})
normalized = text.translate(table)
print(normalized)

s = "注文番号: No.ＡＢ-123 ４５６-xyz"
# 文字列中に現れる、数字以外の文字を集めて削除指定
delete_chars = "".join({ch for ch in s if not ch.isdigit()})
table = str.maketrans("", "", delete_chars)
digits_only = s.translate(table)
print(digits_only)

import string

s = " \tHello,-_ World!! \nNext\tLine... "
# 空白類(string.whitespace)と句読点(string.punctuation)を削除
delete = string.whitespace + string.punctuation
table = str.maketrans("", "", delete)
cleaned = s.translate(table)
print(cleaned)