dtourolleandClaude Opus 5 4dc954f01a
Benchmarks / CPU and I/O (per commit) (push) Successful in 3m53s
Benchmarks / Frame budget (on demand) (push) Skipped
Build and test / Desktop (Linux) (push) Failing after 58s
Build and test / Layer separation (push) Successful in 45s
🐳 Android image / Build and push (push) Successful in 4s
Build and test / android-image (push) Successful in 4s
Traceability / Requirement traces (push) Failing after 1m0s
Build and test / Android (aarch64) (push) Failing after 31s
Take the photo roll's grab band off the buttons that end a mode
"Done Cropping", "Done Repairing", "Done Masking" and "Fit" float over
the foot of the canvas. So does the photo roll's swipe handler, and a
gesture handler is not a layout box — it is an input surface. A press
inside one is delayed, then offered to that handler's own children and
to nothing else: `input_event_filter_before_children` returns
`DelayForwarding`, which aborts the hit-test traversal outright, and the
replay afterwards visits only the handler's subtree. Everything behind
it is never asked, hover included.

The band was `strip-height + reach` — 136px along the bottom — whether
the roll was out or away. So the button that ends a mode was drawn, was
lit, and did nothing for as long as a library was open, which is the
whole time anybody is developing from one. The tool rail kept working
because it is a sibling of the canvas rather than behind the roll, which
is exactly why this looked like two dead buttons rather than a dead
region.

The band now goes where the roll goes. The handler carries the strip
instead of standing still while the strip animates inside it: closed,
only `reach` is on screen and the rest hangs below the window where
nothing can press it; open, it still covers the thumbnails, which is
what lets a swipe down anywhere across them put the roll away. The
180ms travel moved from the strip onto the handler, so the drawn
positions in both states are what they were.

The controls are then positioned against that band rather than against
the bottom of the canvas, and ride up with the strip when it comes out.
Reordering them in front of the roll would have been the other fix, and
it is the wrong one — the band would become the thing that cannot be
reached, and a gesture nobody can start is worse than a button with a
second way out.

`roll-strip` and `roll-reach` are tokens now, because two files have to
agree on where that band is for either of them to keep out of it.

The bottom of the photograph comes back with it: the crop's lower
handles and a repair placed near the bottom edge were inside the same
136px and had the same fault.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-30 19:42:17 +02:00
2026-08-30 16:52:05 +02:00
2026-08-26 10:08:51 +02:00
2026-08-30 16:52:05 +02:00
2026-08-30 16:52:05 +02:00

DarkRoom

A cross-platform, non-destructive RAW photo editor for Linux and Android.

Status: 0.9.0, and no longer a spike. A library opens, culls, develops and exports on both platforms, across eight tagged releases. What is not built is written down rather than merely absent — see docs/outstanding.md for the requirements that have no implementation and why, and docs/technical-debt.md for the compromises that were chosen.

Documentation

Document Contents
CONTRIBUTING.md How to land a first change without reading the rest
requirements.md What the software must do — 179 numbered requirements
architecture.md How it is built — crates, GPU pipeline, data model, sync
technical-debt.md Compromises taken deliberately, each with the condition that retires it
outstanding.md What is not built, and whether that is a decision or a gap
code-health.md What a contribution costs, per seam, measured
traceability.md Generated: which requirement is claimed by which file
faces.md Face detection and identity — the models, the licence problem, and what S14 measured

Building

Desktop:

cargo run -p darkroom-desktop

Android (containerised toolchain, see docker/android):

./docker/android/build.sh cargo ndk -t arm64-v8a build --release

Git LFS is required for the model weights, and the toolchain pins itself. CONTRIBUTING.md has the details and the four commands CI will run against what you send.

Current state

Working. A catalog over a local folder, a Nextcloud account, or a folder a sync client keeps in virtual-files mode — where a placeholder is treated as the photograph rather than as a one-byte file. A virtualised library grid with a capture-time timeline, ratings, labels, keywords, collections and a trash that survives a crash mid-operation. Card ingest. Face detection and identity, with the index syncing between devices. A develop pipeline of fifteen declared operations fused into a single compute dispatch, plus the neighbourhood operations that cannot be — clarity, texture, capture sharpening, noise reduction, lens correction, spectral film simulation. Crop, straighten, spot removal, gradient and subject-segmentation masks, named presets, and a generated panel that no operation in ui/ is allowed to name. Export to JPEG, PNG and 8- or 16-bit TIFF with resize and output sharpening.

The zero-copy display path works on desktop. The compute pass writes a texture that Slint composites directly, which is what ARCH §6.1 requires; the readback it forbids costs 96% of frame time at 4K, and

cargo run -p dr-gpu --example bench --features readback

still reproduces that measurement. The one exception is the Android develop view, which reads the frame back through the CPU because zero-copy there needs wgpu's Vulkan swapchain, and that tears a portrait window on a tablet whose panel is mounted landscape. It is debt, not a revision of the rule: the reasoning, the on-device measurements that forced it, and the three separate things any one of which would remove it are in technical-debt.md TD-1.

Not built. Plugins, compare and survey culling, focus peaking, burst grouping, AI denoise, tiled and progressive rendering, and most of the Android platform integration beyond running. The performance targets in §4.1 are unverified rather than unmet — the per-commit benchmark suite §8 requires does not exist, so nothing fails a build on a regression. docs/outstanding.md is the list, with the reasoning.

Licence

GPL-3.0-or-later.

S
Description
No description provided
Readme GPL-3.0
1 GiB
2026-10-07 11:27:59 +00:00
Languages
Rust 86.1%
Slint 10.3%
Python 1.1%
Shell 1%
WGSL 0.9%
Other 0.6%