Measured against the darkest frame on real ARW brackets, each frame is linear to 1 % up to the clip far from
clipped pixels, but within 2 px of them the +2 EV frame's green reads 3–5 % low: demosaicing interpolates it from
clipped photosites. That frame carries the most weight (k = 4) and those pixels were unclipped themselves, so they
kept it, and bright windows took a magenta tinge.
A frame's weight now goes by its brightest channel within 1 px (3 × 3; wider margins changed nothing more). On two
real interior brackets (ILCE-7M3, 0 / −2 / +2 EV), bright pixels whose green against red and blue is over 10 % below
the darkest frame's, in camera RGB: dining room 0.32 % → 0.08 %, kitchen 2.86 % → 0.28 %. Rendered at −3 EV with
main's colour tags (#1123), window pixels with G < 0.85·min(R, B) go 50 → 3 and 16 → 10 (the darkest frame alone:
2 and 205); most of the visible magenta those brackets showed before came from the missing colour tags. Test: a
bracket whose +2 EV frame has 5 % less green next to its clipped pixels merges without the deficit (fails before:
−1.96 %).
Docs: docs/merge.md, the LR-MERGE-HDR row, docs/whats-new.md (also for the frame-order fix).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015gRtdWF1E9358FfzhzYqCS
header_plan skipped an ARW6's strip but its first 64 KB and then said Ready, while the ARW6
header probe (Mode::Header) validates every tile: its picture header, stream tables and the
first lines of its residual stream, which lie in the skipped bytes. The probe of the partly
read buffer then failed ("ARW6 first-half row count outside 2..=7") where the whole file
succeeds: sparse_corpus failed on arw6-sony-ilce7rm6-apsc-land-c.arw, the first ARW6 file it
reads (it stops at the first failure). Like NEF, ORF and PEF, whose header probes read
samples, an ARW6 now gets HeaderPlan::Whole from the header plan. The preview plan is
unchanged: the embedded preview reads no tile, and the 13 ARW previews in the corpus are
still read from 21 % of their bytes.
Only the test reads the header plan today (the app's partial reads use preview_plan), so no
import changes. New test: a synthetic two-tile ARW6 whose second tile lies past a first one
of bulk size (Ready before, Whole now). sparse_corpus passes on the 8 ILCE-7RM6 ARW6 files.
cargo test --workspace: 3205 passed, 1 failed, visualize::tests::
all_masks_overlay_shows_every_visible_mask_at_once, which fails on main b3d392b4 too
(PR #1456 addresses it); fmt, clippy, parity, layers, assets and wasm pass.
Sources: own design.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The HDR merge has the same cause as the panorama: its result took the
reference frame's colour tags, and a raw whose white balance is only in the
vendor's multipliers (CR2, NEF, ARW...) has no AsShotNeutral in them.
merge_hdr now takes the median multipliers of all its inputs (the rule of
consensus_multipliers: the mean of the middle two for an even count; missing,
non-finite and non-positive values do not vote) and sets the result's
AsShotNeutral to their reciprocal, as stitch does. With no multipliers (DNG
sources) the reference frame's tags are kept, as before. HdrResult carries the
multipliers so that the brackets of an HDR panorama vote as frames of their own
when the merged brackets are stitched (the engine passes them on).
In the app and the CLI the neutral from starting_colour_tags (#965) is already
there for frames balanced alike, so those renders do not change.
Tests: an HDR merge of raw-style frames gets the median AsShotNeutral (odd and
even counts, missing and unusable multipliers ignored) and the written DNG says
it; without multipliers the reference frame's colour is unchanged.
A raw whose white balance is only in the vendor's multipliers (CR2, NEF, ARW...)
has no AsShotNeutral of its own. Loading a frame dropped the multipliers and
stitch copied the first frame's colour tags, so a panorama of frames that reach
the stitcher without a neutral wrote a DNG with neither AsShotNeutral nor
AsShotWhiteXY, and a viewer had no white to start from (a strong colour cast
until White Balance > Auto).
Frames now carry the multipliers (Frame::wb_multipliers, filled from the decoded
raw; load_frame_with is split so a decoded raw can be tested without a file).
stitch takes the median R/G and B/G of the multipliers of the frames that are in
the panorama (the mean of the middle two for an even count; frames without
multipliers, or with non-finite or non-positive ones, do not vote) and sets the
result's AsShotNeutral to their reciprocal, green = 1. The source's
AsShotWhiteXY is dropped with it: a reader takes it before the neutral and the
DNG spec makes the two exclusive. With no multipliers anywhere (DNG sources) the
first frame's tags are kept, as before.
In the app and the CLI a raw without colour matrices already gets a neutral from
starting_colour_tags (#965). For frames balanced alike the new neutral renders
the same (checked in the engine crate); it differs when the camera balanced the
frames differently: the panorama follows the median instead of the first frame.
Tests: the frame carries the multipliers; the consensus for odd and even counts,
with missing and unusable values; stitching raw-style frames gives the median
AsShotNeutral, ignores frames outside the panorama and leaves the colour as it
was without multipliers; the written DNG, read back with the raw reader, has the
reciprocal AsShotNeutral and the white balance the source's multipliers give.
Held-out and library-wide calibrate runs on real CR3s (calibrate's CR3 /
CR2 input is #1073): profiles change little (R7 2.52 -> 2.57, R6 Mark II
6.12 -> 6.09 mean dE), so none is bundled; notes the CRX adaptive QP
172/175 files that stay preview-only. The catalog test pins that an EOS
R6 Mark II (Exif "Canon EOS R6m2") edits in Kelvin through its measured
matrices and an EOS R7 stays relative.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CzWyUn77ZWLe3q3qMiQbVt
A Lightroom preset's crs:Look (name, group, UUID, amount; the Look table
in .lrtemplate files) is kept as profile.look and resolved against the
user's imported profiles by name, the group breaking ties. Lightroom's six
basic profiles map to ours; its own creative/camera profiles and profiles
not imported yet are reported per preset. References are resolved again
after profile.import / profile.deleteImported, so profiles and presets can
be imported in either order.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FxP2xEy623xzWGqCWXVwtb
`fit` to the source's own size copies the whole picture (`resize` returns a clone). The raw path
called it unconditionally, so every full-size decode (an export, and its denoised twin) made a
second float RGB buffer: 288 MB and 160 ms at 24 MP, ~390 MB and 225 ms at 33 MP. The JPEG and
preview paths in the same file already fit only when the picture is larger than `max_edge`.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Fixes for the Polish catalogs of #1100, on top of a merge of main:
- English plural suffixes ("s"/"") and the "it"/"them" words in 51 format
messages are now consumed with `{:.0}`, as docs/localization.md asks;
Polish showed "Zaimportowano 5 zdjęćs" and "utracisz they are". The
counts read "Zaimportowane zdjęcia: 5", which is right for every number
(Polish has three plural forms; ru and uk use the same pattern).
- The 19 format messages main gained since the translation was written;
without them the build fails.
- 118 plain messages the other catalogs already translate.
- "▸" and "▾" in translations become "→" and "⌄", as the other catalogs do
(translated_catalogs_do_not_use_missing_menu_triangles).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A trackpad swipe, a tilting wheel or Shift+wheel over a Develop slider nudges it by the
wheel/arrow-key steps, without selecting it first (the right panel only scrolls vertically).
Only a mostly sideways movement counts, so scrolling the panel leaves the sliders alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- The two halves of each stream are independent bit streams: they now decode in parallel
(`decode_stream`) instead of one after the other inside each of a tile's 13 streams.
- The vertical inverse 5/3 runs its rows in parallel. The horizontal pass lifts each row in place
(`lift_row`, scratch space reused per thread), rows in parallel, instead of allocating two vectors
per row and copying the result back; a test checks it against the vertical pass, row by row.
Best of 3 on 4 CPUs, previous and new decoder alternating: ILCE-7RM6 FF HQ 0.60 → 0.48 s
(process CPU 1.14 → 0.87 s), FF Compressed 0.62 → 0.48 s, APS-C HQ 0.34 → 0.23 s, APS-C
Compressed 0.31 → 0.24 s (docs/arw6-compression.md → Timings). The decoded mosaic is identical
(hash of every sample) on the 8 ARW6 files of the RawDB ILCE-7RM6 set and 3 private A7R VI photos.
Co-Authored-By: Claude <noreply@anthropic.com>
- The vertical inverse 5/3 lifts whole rows at once (`inverse_lift` with samples `w` values wide)
instead of gathering each column through a strided iterator and scattering it back: the same
integer arithmetic, sequential memory. The horizontal pass is the same function with w = 1.
- `0^n 1` codes are counted with one `leading_zeros` over the next MAX_DPT + 1 bits instead of bit by
bit (same bits consumed, same errors); each line reserves its values up front.
- Colour reconstruction reads plane rows as slices and runs over rows in parallel (it was one
thread per tile: 4 threads on a full-frame file whatever the machine).
Best of 5 on 4 CPUs, previous and new decoder alternating: ILCE-7RM6 FF HQ 0.56 → 0.46 s,
FF Compressed 0.54 → 0.38 s, APS-C HQ 0.34 → 0.23 s, APS-C Compressed 0.35 → 0.24 s
(docs/arw6-compression.md → Timings). The decoded mosaic is identical (hash of every sample) on
the 8 ARW6 corpus files and 2 private full-frame photos (ISO 100, ISO 12800); the pinned corpus
sums still match.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three changes, each leaving the output bit for bit as it was:
- Each step runs only where the tile's output depends on it (choice ±0, homogeneity ±1, Lab ±2,
greens ±3, mosaic ±5 inside the 6 px margin), with direct neighbour indices instead of an
index clamped on every read; the clamped values at the buffer edge never reached the output.
- A candidate pixel is built in registers before its Lab conversion: writing its channels
through an index and reading the pixel back stalled on store forwarding (the hottest
instructions of the profile).
- Tiles are written into the output one band at a time, instead of all 64 MP of tiles being
collected and then copied (797 MB held twice). Tiles inside the image copy their mosaic rows
instead of reflecting and taking a modulo per sample.
Sony ILCE-7RM6 9984 × 6656 Compressed HQ raw, demosaic only, best of 5 on 4 cores:
1.85 s → 1.05 s wall, 6.3 s → 4.0 s CPU. Verified identical (f32 bits) on 7 real raws
(ILCE-7RM6 FF/APS-C, ARW6 HQ/C and lossless, ILCE-7RM4A) and, against the previous code, on
random mosaics of all four Bayer layouts at 1 × 1 … 300 × 7, 257 × 130 (tile edges, reflection).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
`load_bytes_now` ran every development through `resample::fit`, which returns a clone when the
image already fits. At full size (1:1 zoom, export) that cloned the whole picture: 797 MB and
~0.3 s for a 9984 × 6656 Sony ILCE-7RM6 raw. It now fits only when the image is larger, as the
other preview paths in files.rs already do. Output unchanged (same pixels, by construction).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
docs/cr3.md gains a variant-status table: every supported and rejected CRX
coding variant, the function that gates it and the exact Unsupported text.
crates/raw/tests/cr3_variants.rs enforces the rejections (19 tests). Each
rejected variant is rebuilt as a small synthetic CR3 (container, CMP1
descriptor, CRX stream) and the test asserts the exact string, and that
probe_info (the import path) and decode return the same error. Control
fixtures decode through the same container, including both edges of the
accepted uniform quantization range (4..=44) and adaptive QP range
(131..=168), so a rejection can only come from the documented gate. The
table itself is prose and is not checked against the tests.
The doc said only QP values 131-167 were accepted; the decoder also accepts
168 (quantization value 44, step 102), so that sentence now says so.
Tests and docs only; no decoder change.
A double-click on a slider's name or thumb already reset it, but nothing said so: only the value had a
tooltip. Resting the pointer on the name or on the thumb now shows "Exposure — double-click to reset
it", the wording (and translations) the curve channels already use. The rest of the track, which is
clicked and dragged, gets no tooltip, and the value keeps "Click to type a value".
The first click of a double-click on the thumb moved the value to where the pointer landed, a few
points off its centre: one history step, then the reset another. A click on the thumb now leaves the
value alone, so the reset is one undo step, as from the name. A disabled slider's name no longer
resets it.
No per-slider reset button: Lightroom has none, and the double-click it documents is now shown where
it works. Tests: tests_slider_reset.rs (name and thumb reset Exposure and Contrast in one undo step,
the hints, a click on the thumb keeps the value, typing a value still works) and
widgets/tests_slider_reset.rs (a disabled slider's name).
crates/raw/tests/robust.rs only mutated DNGs. Add two small valid CR3
fixtures (a 4x4 lossless CRX and a 16x16 three-level C-RAW CRX) and a
property that flips bytes in them or truncates them, then runs the existing
exercise() helper (probe, embedded_preview, decode, develop and the colour
transform) plus probe_info, the header-only description used on import:
errors or clean decodes, never a panic. Unlike random TIFF-like bytes, a
structurally valid file gets past the container walk, so the mutations land
in the CRX marker walker and entropy readers. A fixture test asserts that
both files decode as built, so the mutations hit live decode paths.
Tests only; the existing DNG property keeps its 400 cases and the new one
uses proptest's default (256, PROPTEST_CASES overrides it).
The Remove tool's spot pins (Remove / Content-Aware / Heal / Clone) and the
mask pins now follow a Tool Overlay mode, saved with the UI state:
- Auto (default): shown while the pointer is over the photo (or a pin is
dragged), hidden while a new spot or brush stroke is painted
- Always: all of them
- Selected: only the selected spot's or mask's
- Never: none; the tool still works and a click paints a new spot
A hidden spot can't be grabbed. Mask pins also need Show Pins
(view.maskPins), which keeps its meaning and reply.
Command view.toolOverlay {mode?: auto|always|selected|never} (no mode steps
to the next), Shift+H (H stays Remove), View > Tool Overlay with the modes
and the cycle item, and a Tool Overlay dropdown in the Remove and Masking
panels. Headless tests cover each mode with the pointer over and away from
the photo, grabbing, the key, the command, the menu, the dropdown, mask pins
with Show Pins, and the ui.json round trip.
crates/raw/tests/crx_properties.rs drives lightcraft_raw::decode with
synthetic CR3 containers built around CRX streams that the test encodes
itself: reference entropy coders for the MED, Rice, MELCODE and QP-map
layouts, and the exact algebraic inverse of one 5/3 lifting line. The
tile-halo coefficients are free parameters of that inverse, so synthesis
must reconstruct any encoded plane exactly. Tests only; no decoder change.
- Wavelet round trip: random planes decode exactly across odd plane sizes
and multi-tile halo configurations (eight geometries, arbitrary halo
values).
- Quantization: every uniform quant value 4..=44 and every adaptive QP value
131..=168 dequantizes within the 14-bit range. Overshoot is clamped, and
an in-range plane lands exactly on level * step, so the step sequence
itself is observable. Quant 3 and 45 and QP 130 and 169 are rejected by
decode and probe_info alike. Random QP maps, including odd map heights,
with random per-band base and gain stay in range.
- Lossless samples outside the bit depth are Corrupt errors instead of being
clamped; in-range planes decode exactly.
- Adversarial entropy payloads planted in well-formed lossless, C-RAW,
version 0x200 and QP-map streams either error or decode deterministically
and never panic. A QP Rice prefix longer than the 23-zero escape is
Corrupt.
- Random IAD1 geometry either decodes with the active area inside the
sensor and the crop inside the active area, or is rejected.
- Decoding gives identical results across repeated and four-way parallel
invocations.
- Odd QP-map heights decode: the last QP row has no partner and is used
alone, and QP values are averaged before the step lookup. Large QP jumps
take the escape code.
The case budget is proptest's default (256 per property); set
PROPTEST_CASES to change it.
`Overlay::AllMasks` painted each mask's plane in turn, and every mask view
rewrites the whole image from its alpha (White on Black writes `255·a`,
Image on Black multiplies by `a`…), so each plane wiped the ones before it
and only the last visible mask showed. The planes are now merged (per-pixel
max) and painted once; the GPU path goes through the same `apply`.
The test added with #919 could not pass: besides the above, its hidden-mask
check expected the plain pixel under White on Black, where everything
outside the masks is black by definition (as the single-mask test asserts).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NH3itd85jPuM6wNjgcGfEH
In Compare a filmstrip click goes to `compare.set` (#747), so the grid's
modifier clicks never reached `library.select` and multi-select from the
filmstrip was impossible there (#187 follow-up). A ⌘/Ctrl-click now toggles
the photo and a Shift-click extends from the active photo, as in the grid;
the two photos shown stay selected and the active photo is unchanged, so a
rating key still acts on a photo that is on screen. A plain click still
replaces the active side.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NH3itd85jPuM6wNjgcGfEH
The AI Denoise panel, model download and settings were still in English
in Brazilian Portuguese; this translates its 98 strings.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The grid queues no render behind an unedited raw's quick thumbnail, since the embedded JPEG is
meant to be the thumbnail. But whether that JPEG can be read is only known when the quick job
runs, and `quick_thumb_job` had no fallback: when the preview loader returned nothing the job
failed with "no quick preview", the failure was remembered for that key, and the cell stayed
empty until an edit changed it. The quick job now carries the real thumbnail job as its last
resort, as `quick_view_job` already does for the loupe.
- Local folders in the sidebar are drawn at their tree indent like Albums and
Folders: the icon lines up with the rows above and the disclosure triangle
sits in the gutter left of it; children nest by 16 pt (#1097).
- The central panel that hosts the dock is filled with chrome, so the gaps the
dock leaves between panes no longer show the dark clear colour (#1098).
- The tool strip's bottom group is two buttons and one gap tall, so Info ends at
the bottom margin as Presets starts at the top one (#1099).
Each fix has a headless regression test that fails without it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Four CC0 EOS RP samples from raw.pixls.us (lossless RAW and C-RAW, full
frame and 1.6x crop mode) now have exact full-sensor regressions against
the DNGLab 0.8.0 black-box oracle (Windows binary, checked first against
the documented M50 RAW hash). The decoder already matched; no decoder
change was needed. Crop mode is a single 4032x2656 tile, full frame two
3192-wide tiles.
docs/cr3.md also records that the white level is Canon's specular white,
which at the -1/3-stop ISOs (e.g. 11062 at ISO 5000) is well below the
sensor's saturation point.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conflicts: the 9 locale *-formats.json catalogs (both sides appended
messages at the end; kept both) and docs/parity-checklist.md (header
change list, regenerated summary via `cargo xtask parity --write`,
revision history rows from both sides).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Many pt-BR format strings dropped the English plural argument with
{:.0}, so nouns never became plural ("Exportada 5 foto"). Also fixes
the gender of "matiz" and two terms that disagreed with the catalog.
#1123 gives HDR and panorama merges of raws without colour matrices the colour they open with, but writes
ColorMatrix1 from the forward matrix, so the as-shot white the DNG declares lands wherever the fitted look and the
generic camera model take it. On real ILCE-7M3 brackets the HDRs opened at As Shot 6113–7307 K / +115 (Duv +0.038)
where the ARWs show 3932–5268 K / +7, and any white-balance edit turned them green (CIEDE2000 11–13 against the ARW
at the same Temp/Tint). The render at As Shot was right (0.4–0.8).
The colour matrices now declare the white balance the source shows, decided by files::WbScaleOf (new dng_white):
the camera's measured matrices when they give the source's scale (lightcraft_raw::spectral; the DNG then reads every
white as the source does), else a matrix under which the as-shot neutral is the white the scale shows (Sony's
presets' reading, 6500 K / 0 on the relative scale, the file's own white otherwise). Forward matrix and neutral are
unchanged, so is the render at As Shot.
Measured on the same brackets with this change on a fork build: As Shot 4813 / 5268 / 3932 K, +8 / +8 / +7 as the
ARWs (declared white Duv +0.003); same Temp/Tint within CIEDE2000 0.8–1.7 of the ARW for most changes, about 5 for
the largest (5268 → 3200 K, 3932 → 6500 K: the ARW adapts its look with Bradford, the DNG rebalances in camera space).
Tests: a merged DNG shows its source's As Shot on all four scales (measured matrices, presets, relative, own); with a
look that tints white the DNG declares the shown white, gives the measured model's multipliers at other whites and
renders at As Shot as the source does. Both fail on main (ILCE-7M3 case: (7342, 115) for (5268, 8)).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015gRtdWF1E9358FfzhzYqCS
An unedited raw's face tile is cut from its largest embedded preview, read
the planned way (header and JPEG, never the sensor data), when the preview
holds at least 3/4 of the tile's pixels across the face; otherwise, and for
edited photos and other files, it is developed as before. Tiles keep going
to the memory + on-disk thumbnail cache, now at 16 px size steps so the
thumbnail slider reuses them. The planned preview path carries an Adobe RGB
preview colour space and applies it instead of declining to the whole-file
read (Sony alpha 1 previews).
As in PhotoCraft: when portable.txt (or LightCraft.portable) sits next to the
executable, config_dir() is <exe dir>/LightCraftData, so settings, logs and
models stay with a portable copy instead of %APPDATA%. LIGHTCRAFT_CONFIG_DIR
still wins and is used without probing beside the executable. A portable
folder that can't be written logs a warning and the platform folder is used.
The check runs once per process.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The same bracket merged in another order gave a different result: on real ARW brackets, the app's selection
order and the CLI's file order differed on 0.6 % of the pixels (up to 102 levels), while the same order gave
byte-identical files. Alignment seeds were the frames' input positions, and sums, deghost masks and brightness
ties followed them too.
merge_hdr now puts the frames in brightness order first (ties by median, then a content fingerprint, never by
position), runs everything in that order, and returns the per-frame results (EV, alignment, reference index)
in the caller's order. Test: every permutation of a bracket gives the same radiance and deghost overlay bit for
bit, and the same per-frame results (fails before: "the radiance differs").
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015gRtdWF1E9358FfzhzYqCS
Conflicts: crates/engine/src/tests_merge.rs (#1123's test landed where this branch's is; both kept),
docs/parity-checklist.md (the LR-MERGE-HDR row: both notes) and docs/whats-new.md (both entries).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015gRtdWF1E9358FfzhzYqCS
A photo browsed in a Local folder is flagged deleted by photo.delete, but
Filter::matches_in dropped every browsed photo from views without a folder,
so it vanished from the folder and never reached Recently Deleted: it could
not be restored and sat in the catalog invisibly (a deleted record is also
never forgotten as an untouched Local one).
Deleted browsed photos now match the Recently Deleted filter, so they can be
restored; the library view still excludes browsed photos.
Claude