FAILURE MAP
← Case archive

FA-79779 / Barcode symbology encoding / Member archive

Elements on the threshold are silently called wide · case 04

Borderline scans decode to wrong data instead of being retried.

Member previewVariant 4 · 3 implementations · 8 checks per implementation

Case contract

Classify measured element widths of a two-width symbology scanline (bars at even indexes, spaces at odd) into N/W. Bars and spaces are classified separately because ink spread widens bars. Within a class, if max < 1.5 * min all elements are narrow; otherwise the threshold is (min + max) / 2, elements above it are wide, below it narrow, and an element exactly on it is {"error": "ambiguous", "index": i}.

Why this case matters

Retail, logistics, pharmacy and document workflows depend on encoders that produce exactly the module pattern, code-set switches, separators and quiet zones scanners expect; one misplaced module or separator makes a label unreadable or, worse, scan as different data.

One recorded failure

Sample boundary fixture

This sample comes from the broken implementation of a controlled reproducer.

Boundary fixtureActualExpectedOutcome
regression: threshold tie 0"NWNWWNN"{"error": "ambiguous", "index": 1}Failed

MEMBER ARCHIVE

The complete case is available to members.

This record includes three runnable implementations, regression fixtures, execution results, and source hashes.

Member access is invitation-based. Sign in with your invited account to inspect the sources.

Sign in to the archive ↗