Dot escaping \. vs [.]

They’re effectively the same in most regex engines, but they come from different contexts and have slightly different implications.

1. \. (escaped dot)

• Means: literal dot
• Normally . = “any character”, so escaping it removes that special meaning.
• Works everywhere.

2. [.] (character class with a dot)

• Also means: literal dot
• Inside a character class ([...]), most metacharacters lose their special meaning — including .

Key differences

Readability / intent

• \. → clearly “I’m escaping a metacharacter”
• [.] → looks like a set (even though it only has one char)

Escaping layers (like HCL, JSON, etc.)

• In HCL or other string-escaped contexts:

  • \. often becomes "\\." (double escape needed)
  • [.] stays "[.]" (no escaping needed)

👉 This is the main practical reason people use [.] — fewer backslashes.

Behavior edge cases

• In most regex engines: identical match behavior

• Rare edge cases:

  • Inside some complex character class logic (like nested or POSIX classes), [.] behaves like a class, not a literal escape
  • But for simple use: no difference

Rule of thumb

• Use \. when writing normal regex

• Use [.] when:

  • escaping is annoying (HCL, Terraform, JSON, etc.)
  • you want cleaner strings with fewer backslashes

Example (HCL)

# annoying
pattern = "\\."

# cleaner
pattern = "[.]"