RepeatedCall.message: def of bend-lint-rules@0.2.0.0
Import: import bend-lint-rules@0.2.0.0/performance/repeated_call.bend as RepeatedCall
Lines 66-70 of bend-lint-rules@0.2.0.0/performance/repeated_call.bend: https://jend.leeeet.dev/read.txt?path=bend-lint-rules@0.2.0.0/performance/repeated_call.bend&start_line=66&end_line=70

import Base

def message(callee: String, loop: String) -> String:
  callee
    ++ " is called again here, with the same arguments, on this path of "
    ++ loop
    ++ ", so Bend computes it twice: clang does not merge calls of a definition that loops. All lanes. Measured on the CPU: pow2(30) written `Nat.add(pow2(q), pow2(q))` took 3.4 s, against 0.1 s after the fix; in eldergrove-faire and in llama.bend's tokenizer, the fix changed nothing measurable, as those calls walk short lists. Confidence: medium, as the gain is what the call costs. Bind its value once, with `+` if it is used twice, and use that name. If the call is cheap, such as a walk of a short list, expect this finding."

Uses (2):
  https://jend.leeeet.dev/definition.txt?path=Base/base.bend&name=String
  https://jend.leeeet.dev/definition.txt?path=Base/base.bend&name=String.append