Measured on 68 photos straightened by hand in Lightroom Classic and 300 left as shot, Auto
straighten turned 79 of the 300 by more than 1° (65 by more than 2°): the edges of planks, a
branch or a bird's wing were taken for the horizon. It now answers only for at most 10° and on at
least 3 agreeing lines (Upright Level is unchanged): 19 of the 300, with the same 14 of the 68
levelled within 0.5° and 1 instead of 3 more than 1° off. When it finds nothing, the app says so
instead of leaving the photo without a word. LR-CROP-AUTO goes to 🟡: it still leaves most tilted
photos alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
`lightcraft_raw::sparse::header_plan` says which byte ranges of a raw file the header
probe (`probe_info`, `lightcraft_meta::extract`) reads, given a buffer of the file's length
in which only some ranges are loaded: everything but the bulk sample data (large TIFF
strips/tiles minus the first chunk's coding headers, a CR3's image tracks minus the raw
track's CRX headers and QP maps, a RAF's CFA strip, an RW2's raw data). `preview_plan`
also reads what `embedded_preview` needs, for raws shown from their preview. NEF, ORF and
PEF (their header probe looks at samples) and anything else are read whole.
The corpus test checks, for every raw.pixls.us sample the plan describes, that probing the
partly loaded buffer gives exactly what the whole file gives, with the skipped bytes zero and
again 0xFF: CR3 2.7 %, DNG 2.4 %, CR2 3 %, RAF 6 %, RW2 7 %, ARW 8 % of their bytes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
export_progress_and_cancel failed every time on a fast machine (M5 Max, 4
export lanes): the 8 demo photos exported within the 100 ms progress
throttle, so the first progress above 0 was the final 8/8 and the cancel
arrived after the export had finished (a reply to the cancelled request,
or all 8 files on disk).
The test now wraps Headless in a test-only backend that, at the first
photo boundary, keeps polling the server's real progress hook (which
answers pings and reads cancels) until the cancel is read or the test
releases it, bounded at 20 s so a broken cancel fails rather than hangs.
The first export is released once the mid-batch ping is answered (that
was the same race), the second only ends by the cancel. That makes the
outcome exact, so the check tightens from "1..7 files" to "exactly the
one completed photo", which also proves the photos the other lanes had
already rendered are never written. No production code changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The CRX 0x200 QP map opens with its first symbol as a 32-bit literal; every
later symbol is Rice coded with a limited-length escape: 23 zeros, the
terminating 1 and the symbol in 8 bits (the ITU-T T.87 layout for a 32-bit
limit and 8-bit symbols, without its minus-one offset). The old rule (an
escape after 16 zeros with a 16-bit literal) matched only the first symbol,
which is all the R8 and R7 maps need. R5 Mark II maps (QP 150-162) jump by up
to 12 where the Rice parameter is 0, giving 18- to 22-zero prefixes that
misparsed ("Canon CRX adaptive QP value -21087"): 174 of 306 sampled files
from a private R5 Mark II library opened as previews. Every QP map of its
33,665 C-RAW files now decodes to the exact end of its entropy stream; the six
oracle fixtures are unchanged. Established by black-box analysis of the
bitstream; no third-party decoder source consulted.
Adds the four CC0 R5 Mark II samples (raw.pixls.us 7881-7884) to the corpus
with pinned sensor hashes (the crop-mode framing is left to #754), bumps
RENDER_CACHE_VERSION to 43, and updates cr3.md, raw-parity (235 verified
models), coverage, gaps, the checklist rows, ROADMAP, target-app-parity and
What's New.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
