Between a kept luminance share of 0.15 and 0.25 the table colour is now blended in with a smoothstep,
so noisy shadows show no step between pixels that get the table colour and pixels that keep theirs.
At 0.25 and above pixels render exactly as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A slideshow template is the Playback (not "prepare previews"), Layout /
Options / Backdrop, Overlays and Titles settings as one versioned JSON
document (format lightcraft.slideshowTemplate, version 1).
- Engine: slide_template (SlideLook, Template, built-ins, user template
files in <settings>/slideshow-templates read through each section's own
merged() validation; size and count caps; newer versions refused; bad
files reported and skipped; file names sanitised so ids never leave the
folder). Applying replaces the four settings in one step and keeps the
replaced ones as "previous" for the session (prefs have no undo).
- Five built-ins of our own: Plain Black, Gallery Wall, Edge to Edge,
Cinema Wide, Shooting Data (no Lightroom names or designs, no images).
- Commands slideshow.templates.list/apply/save/update/rename/delete/
import/export; slideshow.preview {template} previews without applying.
- UI: Template Browser at the top of Settings > Slideshow (built-in, user,
previous settings; click applies, hover previews on the selected photo,
right-click update / rename / export / delete; save and import), file
dialogs on the desktop; strings in the nine locales.
- Docs: parity row (partial), whats-new, control protocol and MCP.
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>
Rebased onto the activity stack (#601). An export adding its files to
Photos keeps its row through that step, with "Adding to Apple Photos…" as
the detail. The worker turns the row's ✕ off (TaskHandle::set_cancellable)
before its last look at the cancel flag and reports what it saw, so a
cancel either stops the export before Photos is asked or is refused, never
both "cancelled" and imported. activity: a per-task gate orders a cancel's
decision against a step switching cancelling off, so a cancel can't be
accepted between the switch and the step's look at the flag and then be
ignored. An import of its own (export.addToPhotos, or
the Photos step left running after a foreground app.export) holds a row of
kind applePhotos while Photos works: no ✕, no count, the album as detail.
The desktop app refuses export.addToPhotos {wait: true}: its window's
thread never waits for Photos. The label is in every catalog; tests assert
the rows, the refused cancel and the refused wait.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Ten minutes plus five seconds a file per import (was one minute): the
first live run timed out while macOS's permission prompt was still up,
and Photos' duplicates question waits for a person too.
- Files go to osascript 200 at a time, all within the import's one time
limit: one run with every path went over macOS's 1 MiB argument limit
at about 10,000 files, and Photos got nothing. A failed later batch says
how many files Photos had added.
- The progress line "Adding to Apple Photos…" is translated on the UI
thread; on the export worker it was always English.
- docs/apple-photos.md: who can add what (the control port is reachable by
any local program, other users' included), batching, the new limit, and
tested end to end (macOS 26.6.2, Photos 11.0, packaged app). The parity
row says so; What's New has an entry.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Export dialog gets "Add to Apple Photos" with an optional album (macOS
only). app.export takes addToPhotos / photosAlbum (dialog, Export with
Previous, presets, MCP, lightcraft-cli run and render --opt); the new
export.addToPhotos {paths, album?, wait?} adds files already on disk and
export.photosImports {job?} reports how imports went.
A constant AppleScript run by /usr/bin/osascript asks Photos' import
command (Apple Events, not synthetic input) to import the files into a
top-level album of exactly that name (made when missing; albums in folders
are never used; several of that name: refused), skip check duplicates
false, and returns the media item ids. Album names and paths are passed
only as separate arguments (on run argv), never spliced into the script.
The runner is injectable (Session::apple_photos): with_fs installs the real
one on macOS only, and tests use fakes, so no test starts osascript or
Photos.
Imports are jobs (Session::apple_photos_imports), one at a time. The
desktop app (Session::apple_photos_background, set in open_session) never
waits for Photos on its UI thread: the background export adds on its
worker, app.export without background and export.addToPhotos answer with
the running job, and the app announces the outcome. Without a window (MCP,
lightcraft-cli) imports wait by default, and both headless front ends (the
MCP/CLI backend and the snapshot UI) wait for a running one before the
process ends, on every exit path, through one engine helper
(Session::wait_for_photos_imports) whose deadline follows the job: a
writing allowance while an export still writes, then the import's own
time limit from when Photos was asked, under an overall cap. A runner that panics, waiting or on a
worker, ends its import as failed and frees the slot (panic guard plus a
completion guard). Session::after_export reserves
the import slot before an export writes anything and frees it on every
exit path (handed the files, cancelled, failed, dropped), so a competing
import is refused with the export's job number instead of the export
failing its Photos step after writing. Permission (-1743), missing Photos,
cancel, timeouts, ambiguous albums, busy imports, other osascript failures
and partial imports produce actionable messages; the export's files are
kept either way. Hidden on other platforms; the web build compiles it out.
Packaging: the hardened-runtime entitlements gain
com.apple.security.automation.apple-events (app and CLI) and Info.plist
gains NSAppleEventsUsageDescription, without which macOS refuses the
events without asking.
Not yet tested end to end against a real Photos library: manual steps,
with the packaged app, in docs/apple-photos.md. Parity row LR-EXP-PHOTOS is
now partial.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Open Library given a Lightroom Classic catalog folder, one of its .lrdata /
.lrcat-data folders, Backups/, the .lrcat or its -wal/-shm/.lock files used to
create an empty LightCraft library inside Lightroom's folder (or, for the
.lrcat file, close the current library and then fail). Now:
- engine: lightroom_folder::lightroom_catalog_at (names only, top-level scan
bounded to 10 000 entries, never fails); Session::open_library refuses those
places with EngineError::LightroomCatalog {path, catalog} before creating
anything. A LightCraft library (catalog.snap / catalog.log) always wins.
- app: app.openLibrary checks before closing the current library, keeps it,
returns {opened: false, lightroomCatalog, path} and shows a dialog with
Import Lightroom Catalog… (file.importLightroom with the catalog found) and
Cancel. The library problem window gets the error instead.
- Lightroom import status keeps the catalog path after the task ends.
- CLI --library explains how to import instead.
Intro and ending title screens (Lightroom Classic's Titles panel): each on
or off, a background colour and, centred, the Overlays identity plate or a
line of text, in the plate's colour or an override colour, at a scale, for
its own duration (default: the slide duration).
- Playback timeline: titles are items (Item::Title) of Playback; the intro
plays once before the first photo, the ending after the last photo when
repeat is off; random order never moves them; arrows step across them;
fades, pause and readiness work as for slides; n of N counts photos only.
- Engine: slide_title (TitleSettings in prefs.json, TitleCompose drawing
through the slide composer and overlay text renderer), slideshow.titles
get/set, slideshow.preview {title}.
- UI: titles composed on a render worker (Renderer::request_title), the
control bar shows Intro / End, Settings > Slideshow > Titles section and
a Photo / Intro / Ending preview choice; strings in the nine locales.
- Docs: parity row (partial), playback note, whats-new, control and MCP.
Compare and Reference drew each pane from a whole-frame render of at most
2560 px, so 1:1 was a magnified preview and focus could not be checked.
Each pane now also renders the window on screen from the photo's own
pixels (Slot::CompareRegion), as the loupe does (#323). The media cache
keeps up to two full-size originals while either view is open, so one
pane's window does not evict the other's source.
Docs: gaps.md entry 4 shrinks to overlays and soft proofing; checklist
LR-VIEW-ZOOM / LR-VIEW-COMPARE and ui-parity zoom row updated.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01UyHnpE9573a7nJFQqnGNex
Lightroom's Light panel "HDR" switch (`crs:HDREditMode`, `1`/`True` on,
`0`/`False` off) now sets `hdr.enabled` when importing XMP sidecars, presets,
DNG presets and Lightroom catalogs, instead of being reported as unmapped (an
off value was already silent). The rest of the HDR settings keep their
defaults. Follow-up to #646, ported from #466.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An HDR edit's preview shows its SDR rendition, so its histogram said nothing
about the highlights above SDR white. Now it shows the HDR rendition:
`pipeline::hdr_histogram` renders the edit small (384 px, float HDR, same
framing) and bins it with SDR on the left half and the HDR range on the right
(`Histogram::of_hdr`, `HdrBins`), on a fixed scale of Hdr::MAX_EV_LIMIT (5)
stops above SDR white so it doesn't rescale when the headroom limit changes.
The engine's `develop()` (CPU and GPU renders) and the web worker swap it in
for HDR edits; high-bit-depth, windowed and proofed requests keep the image's
histogram, and a display profile's histogram doesn't replace it. Web workers
carry the HDR bins across their wire (`Histogram::to_words` / `from_words`).
The panel draws it in Lightroom's layout: a solid line at SDR white, dashed
lines between the stops, the headroom limit as an orange line with the range
past it dimmed, a bar under the plot (white; one band per stop in Visualize
HDR range's colours while that is on), SDR / HDR labels and an info tooltip.
Follow-up to #646, ported from #466.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Readers that don't read the PNG cICP chunk show the PQ samples as if they
were sRGB (flat, washed out): expected for PQ PNG, but it surprises people.
Noted in What's new (with the gain map JPEG as the look-right-everywhere
choice), as the export dialog's HDR PNG hint, and in the option and encoder
docs. Review suggestion on #664.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With HDR output on, a photo edited in HDR exports as a 16-bit PNG with the PQ
curve in Rec. 2020 (SDR white at HDR_REFERENCE_WHITE_NITS), described by a
`cICP` chunk (PNG third edition: BT.2020 primaries, PQ, RGB, full range) and
a `cLLI` content light level chunk (the levels the HDR AVIF records). EXIF,
XMP and print resolution are kept; no ICC profile next to cICP. Photos without
an HDR edit, and HDR edits with HDR output off, export as before.
`codecs::encode_png_pq` sits next to `encode_avif_pq` and takes the same
input. Follow-up to #646, ported from #466.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Exercise both separator directions for folder and individual-file imports.
Verify duplicate detection, stable photo IDs, added/deleted file detection,
and reuse of a confirmed scan when the folder spelling changes.
Co-Authored-By: GPT-6 Astra <noreply@openai.com>
Rebuild fixture path components with native separators so expected paths
match directory scans on Windows, including files that have been removed.
This is a prerequisite for Windows validation and overlaps the path-helper
fix in upstream PR #660; it can be dropped once that change lands.
Co-Authored-By: GPT-6 Astra <noreply@openai.com>
The grid already showed the camera's embedded preview first, then queued a render of the whole raw
to replace it. For a 512 px thumbnail that render decodes 50 of a 56 MB Fuji RAF's bytes for a
picture the 4 416 × 2 944 embedded JPEG already covers — 45 s a photo on the 1.2 MB/s share this was
measured on, 8.3 GB of traffic for a 149-photo folder, and a queue that delayed the render the user
was actually waiting for (measured: the first loupe render of a cold photo finished 174 s after the
click, most of it spent behind thumbnails). Idle for a minute afterwards, the same folder read 12
previews (26.9 MB) and decoded zero raws for thumbnails, against 638 MB of whole-file reads before.
- `Session::embedded_thumb_is_final` answers whether a photo is an unedited raw whose camera JPEG the
viewer is showing — the same `has_import_look` gate the preview path uses.
- `grid::skips_the_real_render` is the decision (unit-tested): a cached render as stand-in, an edited
photo and a raw without an embedded JPEG all still render, because their thumbnail must show the
pipeline's output rather than the camera's.
- `Renderer::remember_embedded_thumb` records the quick render's key and the bucket, and
`thumb_quick_is_final` / `thumb_quick_pending` let the per-frame grid request stop at the embedded
preview instead of rebuilding the render job (and its cache key) every frame.
- The thumbnail key still follows the settings hash, so editing a photo re-renders its thumbnail
through the ordinary background path.
`docs/raw-preview-reads.md` and the `LR-IMP-PREVIEW-READ` parity row record the behaviour and the
numbers.
Co-Authored-By: DeepSeek V3.2 <noreply@deepseek.com>
HueSat::apply restores luminance with before / after. Near black, where noise makes some channels
negative, the hue/saturation table can bring after close to 0, and the gain turned single pixels
white (isolated white dots in dark areas, most visible at fit). Luminance is now restored only if
the table kept at least 25 % of it; such colours keep their colour, like black and non-finite ones.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A Sony body shot in Adobe RGB writes ColorSpace Uncalibrated with InteropIndex R03 in the raw's EXIF and leaves the
embedded JPEG untagged, so the look was fitted against the JPEG decoded as sRGB. The EXIF fallback now applies to
all makes (the Nikon maker note still wins); DNGs are skipped, since their previews come from the converter.
A watermark (text, horizontal or vertical, or a graphic) can be turned 0,
90, 180 or 270 degrees clockwise, like Lightroom Classic's rotate buttons.
It is sized before it is turned and placed and inset by the turned box; the
shadow still falls down and to the right. New watermark key `rotation`
(other values are errors); settings without it draw unturned. The Export
dialog gets a Rotate row.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Watermark.inset becomes inset_x (from the left or right edge) and inset_y
(from the top or bottom edge), like Lightroom Classic's watermark editor.
The watermark keys are insetX and insetY; the old inset key still reads as
both, so saved presets and existing MCP calls keep working.
The Export dialog's Inset slider becomes Horizontal inset and Vertical
inset, each greyed out when the position is centred along its axis.
cargo fmt reflows WM_POSITIONS and its test from the previous commits.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With Watermark on, the Export dialog shows the first photo it will export
(its cached thumbnail) with the watermark drawn as the file will have it.
The watermark is redrawn on a copy only when it or the output size changes.
draw_watermark_preview scales the 6 px minimum text size and the 1 px
minimum shadow offset by the preview size over the export size, so a small
watermark doesn't look bigger in the preview than in the file. Exports draw
with a scale of 1 and are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Color Grading gets 3-Way / Shadows / Midtones / Highlights tabs (3-Way by
default). 3-Way shows three equal wheels; a range tab shows one large wheel
with that range's Hue, Saturation and Luminance sliders. Blending and Balance
are in every tab.
Precision (#559):
- Shift-drag on a wheel locks the hue and changes saturation only; the whole
spoke is drawn as a guide. Shift can be pressed or released mid-drag.
- A white dot marks each wheel's neutral centre.
- The large wheel has dashed 25/50/75 % saturation rings.
- 3-Way shows the hovered wheel's hue and saturation above it.
Each wheel has an eye (bottom-left, on hover) that hides its effect for
viewing only: RenderJob::with_hidden_grading renders those wheels neutral
under their own key, never cached as the photo's preview. Not an edit, not
saved, reset on another photo; export is unaffected.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Canon EOS R50 C-RAW files reach QP 168 (quant 44, step 102, measured about 100). The quantization step and the adaptive QP map range accept it; anything above stays rejected. Bumps RENDER_CACHE_VERSION since such files change from the embedded preview to a raw decode.
- Engine: slide_overlay (OverlaySettings in prefs.json, the new slideshow.overlays
command): identity plate (text, or a PNG/JPEG file with its transparency; size,
colour, opacity; render behind image), rating stars (own vector stars), up to 16
text overlays with tokens ({caption}, {exposure}, {file}, {date}, {focal}… plus
every rename token; unknown tags show as typed and are reported) and a drop
shadow, each at one of nine anchors of the slide or photo plus an offset, kept on
the slide; the export watermark on the photo. Sizes relative to the slide's
shorter side; validated and clamped.
- Drawn by the engine as a second pass over lightcraft_slideshow::compose_with
(new behind hook for "render behind image"); the watermark's text renderer moved
to engine text.rs (TextLayout, Inter + craft-fonts CJK) and is shared; glyphs
off the image are skipped so huge text stays cheap.
- rename::expand_tokens_with lets overlays add tokens without changing rename.
- Playback composes slides when overlays are on; slideshow.preview returns each
overlay's box. Settings > Slideshow > Overlays section with live preview.
- Docs: parity row (partial), whats-new, control/MCP docs, nine locale catalogs.
- Tests: tokens, nine anchors, star count, plate image opacity and missing file,
text shadow, resolution independence, hostile input, big-slide cost, CJK with
craft-fonts, prefs persistence, preview and headless playback pixel checks.
Measured with RawNIND on a 20.8 MP raw: an Apple M3 does a tile in 116 ms, an
RTX 3060 Ti over Tailscale 345 ms there and back; sharing blindly made the
export take 34 s instead of 6.4 s while the photo waited for the server's last
tiles. The greeting now times a check tile there and back, and tiles are
shared only when that beats this computer's own time per tile (the card's, or
the CPU's with its threads counted as `card_is_faster` does). Otherwise
`denoise.status` → `device.remote` says `slower` with both times. With the
server configured but slower the export takes 7.2 s (greeting + check tile).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Preserve Target-aware sidecar operations, add French/Ukrainian warnings, and ensure omitted diagnostic fields decode in the native UI.
Generated with Codex.
- `lightcraft_denoise::remote`: protocol v1 over TCP (JSON lines + byte-shuffled,
deflated half-float tiles), a `Remote` client (connection pool, four slots,
greeting checks the model's SHA-256 and tile size, gives up after three failed
tiles) and `serve` (shared token compared in constant time, size caps including
payloads that inflate past their tile, connection limit, timeouts).
- Engine: with LIGHTCRAFT_DENOISE_URL / LIGHTCRAFT_DENOISE_TOKEN set, the loaded
model greets the server on first use and adds its slots to this computer's
tiles at once; a tile goes to a free server slot, else runs here, and a tile
the server fails runs here too. `denoise.status` → `device.remote` says where.
- `lightcraft-cli denoise-serve` serves the installed, selected model on this
computer's card (checked against the CPU, as in the app) or CPU.
- docs/denoise.md: setup, safety (unencrypted: VPN/LAN only) and the protocol.
Tested: protocol round trips, refusals (token, model, tile size, hostile
payloads), fallback and give-up; the engine model with a live, a mismatched
and a dead server; the CLI binary serving a synthetic model end to end.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- New crate lightcraft-slideshow (L1, no UI): SlideSettings (margins linked or
independent, guides, aspect preview, zoom to fill, stroke border, cast shadow,
backdrop colour, colour wash, background image; lengths relative to the slide's
shorter side; validated and clamped) and a CPU compose() used by playback, the
Settings preview and slideshow.preview.
- Engine: slideshow.layout (partial merge, saved in prefs.json) and
slideshow.preview {id?, width?, height?, path?} (PNG to path or base64);
RenderJob::with_slide composes on the render worker; the background photo is
rendered once and shared.
- Playback: with a non-default layout each slide is the composed screen-sized
slide, prepared ahead like before; the default layout plays as before.
- Settings > Slideshow: Layout, Options and Backdrop sections and a debounced
live preview of the active photo's slide (with margin guides).
- Tests: composer geometry/pixels/hostile input, command get/set/persist/preview,
headless playback of a composed slide and of the default layout.
Reset Edits rebuilt a photo without a default-preset look from DevelopSettings::default(),
so a raw imported with capture sharpening 40 and colour noise reduction 25
(DevelopSettings::for_raw) came back with both at 0, and with its embedded lens
corrections off. It now starts from the photo's own built-in defaults
(Photo::camera_defaults, the same as import) and keeps the white balance of its
current as-shot reference, as before. Rendered files are unchanged (their defaults
are DevelopSettings::default()). Reported in #763's notes.
The highlights/shadows base (fast guided filter of log luminance, eps 0.35)
followed thin high-contrast detail, so the per-pixel shift flattened twigs,
lines and texture in bright areas (and dark ones under Shadows); Exposure
kept them.
This changes how every existing edit with Highlights or Shadows renders, so
it ships as a new rendering process (docs/process-versions.md):
- Process::V2 (version 2) is added and made LATEST. New photos and Reset get
it; existing edits stay on V1 until Update to Current Process. Settings
saved without the field still load as V1.
- From V2 the base is the same guided filter of a separable median of log
luminance (radius 0.15 % of the long edge, <= 24 px): fine detail takes its
surroundings' shift and keeps its contrast, region steps stay put (no new
halos). V1's base is untouched: local::tone_base_detail is 0 under V1 and
tone_base then is exactly guided_fast(l, sigma, BASE_EPS), as before.
- CPU (local::prepare) and GPU prepare (median_h / median_v kernels, bit for
bit with the CPU median) both branch on it via PlaneSigmas::base_detail.
- The base plane's cache key is now (sigma, process, median radius)
(local::BaseKey), so a warm cache never serves one process the other's
base. plane_at is shared with the GPU crate.
- finish::base_tone: V2 uses V1's tone map.
Tests: V2 keeps twig/local contrast (>= 0.9 of Exposure's) with no halo; V1
plane sigmas have no median and its base equals guided_fast bit for bit; V1
and V2 render alike without Highlights/Shadows; warm-cache renders across
V1/V2 switches equal fresh ones (fails if the key drops the process); GPU
matches CPU on both processes at 720 and 2400 px, cached and not.
Pentax 645D 2401 at 4000 px: main and V1 renders bit-identical (plain,
Highlights -100, Shadows +100). Highlights -100 bright rel. detail
0.095 (V1) -> 0.147 (V2), Exposure -1: 0.124; Shadows +100 dark rel. detail
0.126 -> 0.148.
RENDER_CACHE_VERSION 54 (adjust at merge).
Export as JPEG XL from the dialog, CLI, MCP and control channel (format jxl): lossy with
quality or lossless, 8/16-bit, effort fast/normal, progressive, EXIF/XMP boxes; HDR edits
as 16-bit Rec. 2020 PQ. Encoder: jixel (pure Rust, BSD-3-Clause OR Apache-2.0, native
only: it spawns its own threads), wrapped in catch_unwind. Lossy files declare an enum
encoding (sRGB / P3 / Rec. 2020; Adobe RGB and ProPhoto written as Rec. 2020); lossless
files embed the ICC profile.
Decoder fix: jxl-oxide desaturated colours outside sRGB when rendering XYB images to
wider primaries; lossy wide-gamut JXL now renders to linear sRGB primaries and is
converted to Rec. 2020 here. jixel's slowest effort is left out: jxl-oxide misdecodes
its (valid) files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A grid thumbnail and a loupe preview of an unedited raw show the camera's embedded JPEG, which
sits near the start of the file — for a 24 MP Fuji RAF, 5.6 MB of a 56 MB file (measured:
DSCF9492.RAF). The loader read the file whole and threw the sensor data away, so every preview
paid for 50 MB nobody looked at. On a local disk that is merely wasteful; on a network share,
where photographers keep libraries, it is the wait: the same preview took 10 s over a GVFS/SMB
Wi-Fi share against 0.96 s after this change.
- `lightcraft_raw::bounded_preview(bytes)` locates the preview from the header alone and reports
its byte range; `embedded_preview_from_range` extracts it, byte for byte what `embedded_preview`
returns for the whole file (the pointer is read from the RAF header's fixed u32s at 84/88).
- `engine::files::preview_window` reads a 4 KiB prefix, then — because the container reader clamps
a slice to the buffer it was given, so a short prefix reports a plausible but truncated 3948
bytes — re-reads exactly the extent the header declares, and takes the window out of the prefix
rather than reading those bytes twice. `LIGHTCRAFT_PROFILE=1` prints what each preview cost.
- Formats whose preview only the whole file reveals (CR3 boxes, TIFF pointer tags past the prefix,
CRW/MRW/X3F scans), a lying pointer, a short read and a file that changes underneath us all fall
back to the whole-file path: a damaged header never costs the user their preview.
- Tests: the header range matches `embedded_preview` byte for byte, with a clamped prefix, a padded
extent, a late pointer, a lying pointer and a non-JPEG window covered; on a synthetic RAF with a
200 KB tail the quick path and the whole-file path produce identical pixels and the read stays
under half the file; `crates/engine/tests/preview_read.rs` (opt-in, `#[ignore]`) measures a real
file on a share. No raw fixture is committed.
- `docs/raw-preview-reads.md` records the mechanism and the numbers.
Co-Authored-By: DeepSeek V3.2 <noreply@deepseek.com>
The R5 Mark II's APS-C crop-mode files (5376 x 3574) state an IAD1 active area of x 132..5243,
but columns 132..~256 sit at the black level and the image runs to the right edge (measured on
raw.pixls.us 7883/7884), while their recommended crop, x 272..5359 (5088 x 3392, the camera JPEG's
size), lies on the image. The recommended crop didn't fit inside the stated active area, so the
whole active area was kept: a black band down the left and ~130 image columns lost.
When the recommended crop doesn't fit inside the stated active area, it is now the valid area.
On 7883 the render's high-pass correlation with the camera JPEG goes from -0.13 to 0.99. The
active area and crop of the other 108 decodable CC0 CR3s on raw.pixls.us are unchanged.
RENDER_CACHE_VERSION 49.
Raws without matrices in the file (ARW, NEF, RW2, RAF, CR2, CR3, PEF, SRW,
ORF) were all edited on the relative white-balance scale (6500 K / 0 = as
shot), even when lightcraft_raw::spectral has measured matrices for the
camera. Those 52 bodies (ILCE-7M3, ILCE-7M4, D850, EOS 5D Mark III...) now
edit in Kelvin like a DNG and like Lightroom.
Design:
- One decision on both sides: spectral::find_camera("Make Model") backs
catalog::relative_wb_camera (Photo::relative_wb); files::WbScaleOf decides
the same from the decoded file. A corpus test asserts they agree.
Photo::fitted_look keeps the old predicate for smart previews.
- The matrices only convert: As Shot is the file's multipliers read through
the model's ColorMatrix1/2 (spectral::as_shot_white_through). The look
(JPEG fit / profile / spectral fallback) is unchanged, and the slider
adapts from that white with Bradford. The camera-space rebalance stays
reserved for files with their own matrices.
- Presets resolve to their Kelvin values (WbMode::preset, sources cited).
- The Color panel shows the white balance in effect (effective_wb).
- XMP: crs::Target::for_file takes the camera; matrix cameras read
crs:Temperature as Kelvin (marked "scale": "kelvin"), others keep the
#510 as-shot shift.
Migration (no look changes):
- WhiteBalance.scale: WbScale { Legacy (absent in JSON), Kelvin }. A
white balance without a scale keeps its old meaning. On a spectral-Kelvin
raw (SourceInfo.legacy_relative_wb) the pipeline composes the old
adaptation (6500 -> value) with the new one (as shot -> ...) in Bradford
cone space (local::legacy_white). Von Kries scalings compose exactly, so
the matrix matches to f32 and the pixels are identical
(pipeline legacy_relative_white_balance_renders_as_before_on_a_kelvin_raw,
media old_relative_white_balance_on_a_kelvin_camera_keeps_its_look...).
- Every white-balance writer (develop.set/adjust/quickAdjust/wb/wbPick/
reset) first runs local::normalize_wb: a legacy value becomes the Custom
Kelvin value with the same look and is marked Kelvin. Partials without a
scale (old presets, develop.merge) stay legacy.
- Catalog format 7: Op::SetAsShotWb replaces the 6500/0 as-shot recorded
by older imports when the file is next decoded or reloaded (system op,
not an undo step). Format-version test updated.
- The web wire carries legacy_relative_wb.
Docs: parity (LR-EDIT-COLOR-TEMP), camera-preview-colour, xmp-interop,
lightroom-catalog-import, control-protocol, mcp, whats-new.
develop.set / develop.adjust / develop.quickAdjust / develop.resetControl on crop.angle
(MCP set_develop, the control channel, lightcraft-cli render --set) and crop.set {angle}
now turn the crop as crop.straighten does, through the new CropGeometry::with_angle:
kept if it still fits, else the largest crop of its aspect inside the rotated photo.
Before, they left the full rect, and the export took in smeared edge pixels in the corners.
Thai folder and file names drew boxes: the UI fell back only on craft-fonts' CJK faces, so a
Thai face (craft-fonts #15) or the Arabic one already there was never used. fonts::ui_fallback
keeps the CJK faces first, in their per-language order, then appends every face of another
non-Latin script, whatever the interface language (a Latin-only face is no fallback: Inter covers
Latin). theme.rs builds the Proportional, SemiBold and Monospace families from it.
Tests: the fallback order on synthetic faces; with a craft-fonts build that has a Thai face,
every UI family falls back on it, and the shipped Thai and Arabic faces have their script's glyphs.
Fujifilm RAFs carry per-shot GeometricDistortionParams (0xf00b) and VignettingParams (0xf010)
in the raw IFD. Derived by measurement against the cameras' own JPEGs and LightCraft's
uncorrected renders: [scale, n knots, n values], scale x n = half diagonal of the recorded image,
last knot = farthest corner, distortion in percent indexed by the source radius with the
frame-filling zoom, vignetting as 100/V gain. They become a WarpRectilinear and a
FixVignetteRadial for 59 validated bodies (same path as the Sony ARW and Panasonic RW2
corrections); unmatched bodies and tables with another scale stay uncorrected.
On 67 body/lens combinations the corrected render's radial shape error against the camera JPEG
falls from a median of 0.72 % to 0.05 % of the half diagonal. The lateral CA table (0xf00f) is
real but is not applied: our demosaic already registers the colour planes and the stored shift
over-corrects. See docs/fuji-lens-corrections.md.
RENDER_CACHE_VERSION 48.
