生成单个字符排序规则及正则字符范围扩展可行性技术问询

Absolutely! Since you're working with regex character classes and the C locale's strict ASCII byte-value sorting rules (focused on visible characters 33–126), extending these ranges is totally doable—you just need to lean into how ASCII values map to characters. Let’s break down the key ways to expand your character classes:

1. Extend Single Continuous Ranges #

In C locale, regex character ranges like [b-d] work because they correspond to consecutive ASCII byte values (98 for b up to 100 for d). You can expand this to cover larger continuous blocks as long as the start character’s ASCII value is less than or equal to the end character’s value:

• Match all lowercase letters: [a-z] (covers ASCII 97–122)
• Match all digits: [0-9] (covers ASCII 48–57)
• Match all visible ASCII characters (33–126): [!-~] (since ! is 33 and ~ is 126, this spans every printable non-whitespace ASCII character)
• Match digits, symbols between digits and uppercase letters, and uppercase letters: [0-Z] (note this includes :;<=>?@ too, since their ASCII values fall between 57 (9) and 65 (A))

2. Combine Multiple Discontinuous Ranges #

If you need to match non-consecutive groups of characters, just string multiple ranges together in the same character class—no separators needed:

• Match lowercase letters, uppercase letters, and digits: [a-zA-Z0-9]
• Match b-d and x-z (skipping the letters in between): [b-dx-z]
• Match digits plus the symbols #, ?, and @: [0-9#?@]

3. Add Individual Special Characters #

You can also throw single, non-range characters directly into the class, even if they don’t fit into a continuous range:

• Match b-d plus the exclamation mark: [b-d!]
• Match lowercase letters plus the period and comma: [a-z.,]

Critical Caveat to Remember #

Always make sure your range’s start character has a lower or equal ASCII value than the end character. For example, [a-Z] is invalid because a (ASCII 97) has a higher value than Z (ASCII 90)—regex engines will reject this as an invalid range. If you need both lowercase and uppercase letters, use [a-zA-Z] instead.

As long as you reference the ASCII byte values for the characters you want to include, you can build exactly the character set you need for your regex patterns in the C locale.

内容的提问来源于stack exchange，提问作者user232326