Lesson 34 · Penalty rule 3 - finder-like patterns

Penalty rule 3 - finder-like patterns

The third rule punishes any run that mimics the finder pattern's distinctive 1:1:3:1:1 signature, which could fool a scanner into seeing a finder where there is none. Today you write it.

The goal

Score a grid by penalizing finder-like patterns in any row or column.

Start here - the target

TO DO

Scenario: Finder-like sequences are penalized

Givenpenalty rule 3: each occurrence of the 11-module pattern dark-light-dark-dark-dark-light-dark next to four light modules scores 40

Whena line containing 1,0,1,1,1,0,1,0,0,0,0 is scanned

Thenthat occurrence contributes 40

Andthe reversed pattern 0,0,0,0,1,0,1,1,1,0,1 also scores 40, and both orientations are checked in every row and every column

Background

The finder pattern is recognized by its 1:1:3:1:1 dark-light ratio, and a scanner hunts for exactly that signature. If the data region accidentally produces the same sequence, the scanner might mistake it for a finder and misalign. Rule 3 guards against this: wherever the 7-module core 1,0,1,1,1,0,1 appears with four light modules on one side - the 11-module patterns 1,0,1,1,1,0,1,0,0,0,0 or its reverse 0,0,0,0,1,0,1,1,1,0,1 - it scores a hefty 40.

You detect it by sliding an 11-wide window along every row and every column and matching either orientation. The large penalty reflects how badly a false finder can break scanning, so this rule often dominates the choice of mask. One rule remains - the overall dark/light balance - and then you total all four.

Make it work

// Slide an 11-wide window; match either orientation.
var p1 = []int8{1,0,1,1,1,0,1,0,0,0,0}
var p2 = []int8{0,0,0,0,1,0,1,1,1,0,1}
// for each row and column, for each window of 11:
//   if it equals p1 or p2, add 40

CheckpointDONE

You can score a grid on finder-like patterns. Commit and stop here.