Skip to content

barcodes: README audited against the code - #328

Merged
shreeve merged 1 commit into
mainfrom
barcodes-docs
Sep 18, 2026
Merged

shreeve merged 1 commit into
mainfrom
barcodes-docs

Conversation

@shreeve

@shreeve shreeve commented Sep 18, 2026

Copy link
Copy Markdown
Owner

Twelve documentation defects found by an audit of packages/barcodes/README.md against the code on main, eight of them older than #327: the QR speed range, the corpus timing sentence, the nativeEvery description, lock color defaults, which symbologies share the ECI table, the encoder verification claim, the probes' stated scopes, and the race section's description of the misc/rip-prior checkout. The bench now prints the README's row names verbatim so the tables have one source, the race labels its baseline column rip main, and the PDF417 result type gains the three orientation fields the reader sets. 46 tests pass; a fresh bench run's rows match the README's rows exactly.

Every claim checked against main: the QR speed range reads from the race
table (1.7 to 5.5x), the corpus sentence carries the measured mean and the
slowest image's warm time, nativeEvery is described as the gate applies it
(1 never forces the native search), the lock colors are the literal
defaults, the ECI table is shared by QR and PDF417 only, the encoder is
pinned against generated vectors rather than a shipped oracle, the probes
paragraph states each probe's exact scope, and the race describes the
misc/rip-prior checkout as main. The bench prints the README's row names
verbatim, the race's baseline column is labeled rip main, and the PDF417
result type carries vertical, reversed and inverted as the reader sets
them.
@shreeve
shreeve merged commit b4bb772 into main Sep 18, 2026
2 checks passed
@shreeve
shreeve deleted the barcodes-docs branch September 18, 2026 09:03
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant