# `Alaja.Syntax.Special`
[🔗](https://github.com/Lorenzo-SF/alaja/blob/2.1.0/lib/alaja/syntax/special.ex#L1)

Defines a special tokenization rule for edge cases.

Tested in priority order before standard keyword/string/number matching.

## Fields

  * `pattern` — regex to match at the current position
  * `type` — token type atom to emit (e.g. `:decorator`, `:lifetime`)
  * `priority` — lower values are tested first (default: 50)
  * `multiline` — if true, the pattern can span multiple lines
  * `interpolate` — if true, the matched region contains sub-tokens
    (e.g. `#{...}`) that should be tokenized recursively
  * `consume` — if true, the matched characters are consumed but not
    emitted as a token (useful for sigil prefixes like `~r` in Elixir)

## Examples

    # Python decorator
    %Special{pattern: ~r/@\w+/, type: :decorator, priority: 10}

    # Rust lifetime (before char literal)
    %Special{pattern: ~r/'[a-z_]\w*/, type: :lifetime, priority: 10}

---

*Consult [api-reference.md](api-reference.md) for complete listing*