252 Commits

Include renames

Author	SHA1	Message	Date
dtourolle	e6ac31d39d	Give the face sweep a size budget, so a panorama is never fetched ... The sweep fetches the whole original before it can learn anything about it, and the one file in the reference library the decoder refuses on sight is a 521 MB stitched panorama — so every pass on the tablet spent half a gigabyte of Wi-Fi to find that out again. The catalog already knows the byte count, and that is enough to decide before the fetch: originals over 256 MB are marked examined with nothing found and a zero edge, counted as failed, and named in the log. Below the line is every camera RAW the library holds; above it, four files, all panoramas. A budget and not a verdict on panoramas. The right treatment for one is a tiled pass — read it in strips, detect in each, stitch the boxes back — and the zero edge is what that pass would select on. Until it exists, this is what keeps a background sweep on a phone from paying for the decision the decoder cannot make.	2026-09-19 12:14:05 +02:00
dtourolle	7db999c1f6	Require judgement anywhere, and evidence that never becomes a verdict ... Rating and flagging were reachable from the grid alone, so a photograph opened in develop could not be judged without leaving it; FR-UI-5 said "rating" without qualifying the view and was built as though it had. And FR-UI-1's expanded row has said "filmstrip" since it was written while the roll stayed on demand in both classes. Both are amended to say what they meant: judgement follows the photograph, without auto-advance outside the culling mode, and the roll is open by default where there is room for it. The larger change is a rule. Per-face signals — eye state from a classifier, head pose from the five landmarks the detector already yields — are worth having for culling, and §3.9.1 excluded detecting a blink outright. The exclusion was always of judgement, not of knowing: a blink is a fact about a frame of the same kind as a clipped highlight. FR-CULL-8a specifies the two signals; FR-CULL-13 says what any signal may do (be shown, filtered, sorted, propose a burst representative) and what none may (write a rating or flag without a user action between). R7 states the same thing as a user need. Licensing was read before either was written. OCEC's eye-state weights are MIT with a clean data chain; every open gaze model is trained on Gaze360 or its peers, whose licences restrict derived models by name, so gaze is deferred in §7 and head pose stands in for it. D13 records both so they are not re-searched. Replacing a closed-eyed face from a neighbouring frame was raised and is written down as D17 rather than built: it is the multi-source schema question §7 already defers for panorama and HDR, with its non-goals — never automatic, provenance declared — fixed now. Traceability regenerated: three new IDs, none yet tagged.	2026-09-19 11:47:27 +02:00
dtourolle	327decfab1	Fuse every detector's faces into one population per embedder ... Choosing "Thorough" made the library look empty. The detector setting writes under its own faces.model_id, and every reader of "the faces" keyed on that exact id: the clustering pass, the coverage figure, the sweep's work list, the shard export and import, and the sync merge's face matching. On the reference library that restarted coverage at 1,834 of 19,140, drew a People rail of 36 faces for a person with 520, queued a ~400 GB re-fetch on each device, and stranded the desktop's 3,583 confirmations under the old id: the tablet held the same faces under the new one and the merge refused to match them. Same photograph, same box, same embedder, two ids — that is one face, not two libraries. The embedder half of the id is now the key. embedder_of and embedder_sql give it to every query; writes keep the full id, so which detector drew a box stays on record. record_detections is unchanged and is where the generations meet: an image holds one pipeline's faces at a time, and a re-detection carries confirmations across by box overlap. The merge's match_faces applies the same rule within an embedder. The calibration is keyed on the embedder too, since the similarity space did not change. Shards travel every generation, each under its own id, and a peer adopts whichever it is sent — including a stronger detector's pass over an image it indexed itself with a weaker one, which is the re-detection its own sweep would otherwise queue, already done. Never downwards: a tablet on Fast keeps the desktop's Thorough faces. The sweep gains the same tail — images a weaker detector indexed, after the ones nothing has — driven by FaceDetector::supersedes, so choosing a stronger detector still improves the library over time without first making it disappear.	2026-09-19 10:49:28 +02:00
dtourolle	f8addbee53	Mark a file the decoder cannot open, so the sweep stops fetching it ... A decode failure in the face sweep was counted, logged at debug where nobody saw it, and left unmarked — so the next pass fetched the same file and failed the same way. For the 521 MB panorama behind rawler's panic that was half a gigabyte per sweep, on a tablet. It is now marked examined with nothing found and a zero edge, which is what a later "try again with a better decoder" pass would select on, and the warning names the file. The failure count is unchanged: it did fail.	2026-09-19 10:45:08 +02:00
dtourolle	c0b1e78f7c	Return a panic inside the decoder as an error, not as the end of the thread ... rawler panics on some input rather than returning Err — a DNG whose IFD claims a >50000 px image, which the reference library has: a 521 MB stitched panorama, IMG_4181-Pano.dng. On a worker thread a panic is the end of the thread, so the face sweep that met it stopped thirteen seconds in, three sweeps running on the tablet and three on the desktop, with "17301 image(s) to index" as the last word. FR-RAW-4 says a malformed file must not abort a batch, and that is this crate's promise whatever the library beneath it does: every entry point that calls into rawler now runs under catch_unwind, and a file that panics the decoder is one failed file with the panic's message in the error. Verified on the panorama itself: metadata reads, decode returns the error, the thread survives. The crash hook still records the panic, which is right — it is a defect in a dependency and the record is how it gets reported.	2026-09-19 10:45:07 +02:00
dtourolle	78cb00634e	Fetch the photographs around the open one ahead of the step to them ... Walking the photo roll was one download per frame: every step showed "Downloading…" over an empty canvas while tens of megabytes came down, and moving between a pair of near-identical frames paid that a dozen times. Now, once the opened photograph has landed, the ones around it are fetched into the originals cache while it is being looked at, so the next step is a disk read. A single worker serves the latest wish only, closest first and working outwards — next, previous, next-but-one, previous-but-one… — one file at a time. Each open replaces the wish, so a fast walk never leaves a trail of stale downloads competing with the one being waited on. A process-wide in-flight registry makes a click on a photograph that is still being fetched ahead wait for that transfer and read it from disk, rather than start a second download of the same file. How far each side is a setting under STORAGE — Off, 2, 5, 10 or 20, defaulting to 5 — and it is moot while "keep originals after opening" is off, since a fetch the cache would discard on arrival is transfer for nothing. Nothing is fetched ahead while offline. The transfers show in the activity list while they run and are removed when they end.	2026-09-19 10:36:50 +02:00
dtourolle	33e2e277a2	Set the Wayland app id late enough for it to take ... The launcher and the task bar have shown a generic tile for a working window since the call was written. set_xdg_app_id sat at the top of run(), on the reasoning that the app id is read when the surface is created — true, and beside the point: the call goes through Slint's global context, and there is no global context until something installs a platform. That is BackendSelector inside shared_gpu, or AppWindow::new falling back to the default, and both happen further down. Called before either, it returned NoPlatform and did nothing at all. It moves to just after the window is constructed, which is not the same as shown — run() is far below — so there is a platform to talk to and the surface does not exist yet. The failure was logged at debug, which is why a year of grey squares went unremarked: the whole symptom is invisible from inside the application. It is a warning now, naming the consequence.	2026-09-19 10:15:46 +02:00
dtourolle	9b627e7713	Let a catalog writer wait for its turn instead of losing its work ... SQLite's busy timeout defaults to zero, and nothing ever set one: the loser of a write race got SQLITE_BUSY at the moment it asked. WAL does not cover this — it makes one writer and many readers free, and this application constantly has two writers, the face sweep committing a batch while the derived sync imports shards or reclustering reads. The cost was not a retry but lost work. A sweep that had already paid for the detection and the embedding — seconds per image, the expensive part — discarded the result on "storing faces for 214: database is locked" and moved on to the next image. Both the desktop and the tablet logged runs of those on consecutive images, which is a face sweep quietly failing to store the faces it had just computed. Ten seconds, on every connection, set in configure() so that nothing can open the catalog without it — the figure the job runner's own tests have used for this reason since they were written. It is far longer than any transaction here, so it bounds pathology rather than making anyone wait.	2026-09-14 20:05:53 +02:00
dtourolle	5e67f026ec	Merge: a catalog the server cannot damage for long, and backups on both sides ... fix/corrupt-remote-catalog-deadlock. The catalog on the server had been malformed since 7 September and every device declined to overwrite it, so collections and people stopped syncing everywhere at once; the damage came from two devices assembling chunks in one upload directory. Each chunked upload now has its own directory, the server keeps three generations of the catalog behind the current one, every push is verified before and after, a damaged copy that arrived whole is merged from the generation before it rather than pinned in place, and the catalog is backed up daily as NFR-R2 always asked.	2026-09-13 20:04:55 +02:00
dtourolle	eeee3d920a	Back the catalog up daily, not only before migrations ... NFR-R2 asks for the catalog to be backed up on a schedule and before schema migrations. Only the second half existed: every backup on disk was a pre-migration copy, and a library that never migrated was never backed up at all. A backup is now also taken at the end of a library sweep when the newest one is more than a day old — the moment the catalog is quiet and a day's collection and people edits have just been folded in — on its own thread and its own connection, so the copy of a 130 MB file is not spent on the UI. Whether one is due is read from the backup directory, not the catalog, so the ordinary case costs nothing. An empty catalog is skipped: there is nothing in it a rescan would not rebuild. Pruning to KEEP_BACKUPS applies as before.	2026-09-13 19:32:05 +02:00
dtourolle	f100db89ca	Verify the catalog snapshot before it is sent, and after it lands ... Two checks around the upload, both cheap next to what they prevent. Before: the snapshot is quick_checked before it leaves. It is the copy every other device merges from, and a damaged one costs each of them a download, a failed merge and a refusal to push. After: the staged upload's size on the server is compared to the bytes sent before it is rotated into place. A chunked upload is assembled server-side, and an assembly that goes wrong is a file of plausible size no device can open — caught here, on the device that caused it, for one listing; otherwise on every other device, after the fact. A mismatch, or a size the server will not confirm, discards the upload and leaves the current copy and its generations untouched.	2026-09-13 19:31:58 +02:00
dtourolle	2ce0fcc74a	Keep three generations of the catalog on the server ... The server held one copy of the catalog, overwritten in place on every push. When that copy was damaged there were two answers, both bad: refuse to touch it for ever, which is what every device did for a week, or overwrite it with ours, which loses whatever another device had added since — the escape hatch of the previous commit. A push now uploads to catalog.upload.sqlite, rotates catalog.sqlite to .1 (and .1 to .2, .2 to .3, dropping the oldest), and moves the upload into place. Rotation is server-side renames, oldest first so that every destination is empty when it is written to — move_to refuses to overwrite, by design — and a failure at any step leaves a gap in the generations and never a missing current copy. The only transfer is the upload itself. A damaged current copy that arrived whole now merges from the newest readable generation before ours goes over it, which loses nothing, and is kept as .1 by the ordinary rotation rather than by a separate 40 MB upload. NFR-R2 asks for the catalog to be backed up; this is the half of it that lives with the copy other devices read.	2026-09-13 19:31:58 +02:00
dtourolle	6f62ac09f8	Give every chunked upload its own directory on the server ... The upload directory was named from the destination path alone, on the reasoning that two files could then not collide. Two devices uploading the same file could, and did: both wrote 00001…00009 into one directory, and whichever MOVEd first assembled a mix of the two — a catalog of exactly the right size whose pages came from two different databases. SQLite called it malformed, every client declined to overwrite it, and collections stopped syncing on all of them for a week. A transfer that died on a phone's link left its chunks there for the next device to assemble in, by the same mechanism. The name now carries a nonce as well, so no two uploads share a directory, and a failed transfer deletes its own directory on the way out rather than leaving 5 MB chunks for the server to sweep eventually.	2026-09-13 19:31:58 +02:00
dtourolle	c1e0f09be7	Say where the face models were looked for when they are not found ... "The chosen detector is not installed" was the whole of what a user saw, on a machine where the files were three directories away from where the lookup went. The search order was in a doc comment and nowhere a user could read it. Now a missing pair logs the detector file it wanted and every directory it tried, which is what the first Windows install needed and what the next misplaced download will.	2026-09-13 19:20:30 +02:00
dtourolle	693195fa96	Replace a damaged catalog on the server instead of pinning it there ... The catalog sync refuses to upload when it cannot read the server's copy, because the upload is a read-modify-write and writing blind would discard another device's collections. That is the right rule for a timeout, a dropped connection or a newer schema — the remote is fine, only our view of it failed. A file SQLite calls malformed is not that. No device will ever read it again, so refusing to write over it preserves nothing — and every client declines in turn, pinning the damaged file in place for good. Collections and people then stop crossing between devices on all of them at once, each logging "catalog not pushed" on every pass. This library did exactly that from 2026-09-07, on the desktop and on a freshly installed phone alike, while 32 collections sat undelivered. Now a copy that arrived whole and still will not open is set aside under a dated name and replaced by ours. Whole is checked against the size the server advertises: a truncated download will not open either, and on a phone that is the far likelier story, so anything short — or any size the listing cannot confirm — is treated as the transport failure it is and the server's copy is left alone. A placeholder's size is not trusted for the comparison, since it means nothing. The report says when this happened, and the log line calls it "pushed over a damaged copy" rather than folding it into an ordinary push: it is the one push that discarded something.	2026-09-13 19:01:00 +02:00
dtourolle	b396096787	Open the sign-in URL on Windows ... Login Flow v2 cannot complete without a browser, and the launcher had a branch for xdg-open, one for Android's Intent, and an honest Unsupported error for everything else — which on Windows stranded the flow on "approve the sign-in in your browser". rundll32 url.dll,FileProtocolHandler is ShellExecute on the URL and needs no crate; chosen over cmd /C start, whose quoting of & in a query string is a known trap. Not verified: Wine has no browser to open.	2026-09-12 07:34:09 +02:00
dtourolle	ef1154af94	Resolve every base directory in one place, and on Windows ... Five sites each read XDG_*_HOME and fell back to $HOME/.local/… on their own, which is fine on Linux and wrong everywhere else: Windows sets neither variable, so every one of them degraded to a path relative to the working directory — for a Start Menu launch, C:\Windows\System32. The models lookup walked XDG_DATA_DIRS the same way. dr_plat::dirs now holds the rule per platform: XDG on Unix, the known folders on Windows — %APPDATA% for config, which roams, and %LOCALAPPDATA% for data and state, which do not — and the executable's own directory as the system data dir, which is where the installer puts the models. The Android overrides stay where they were; only the fallback behind them moved. Both rule sets are unit-tested on either host, and the Windows one was confirmed by running the application under Wine: its log landed in AppData\Local\darkroom\state and nothing was written anywhere else.	2026-09-12 07:34:05 +02:00
dtourolle	896188a489	Read the sidecars other editors write, and write them back on request ... FR-CAT-13 asked for standard XMP and `core/dr-xmp` answered the file: it has read and written `dc:subject`, `xmp:Rating`, `xmp:Label` and the IPTC core since 5fa4c07, under an ownership rule that leaves everything else in the document untouched. What nothing did was call it. No scan found an `.xmp` beside a raw, no catalog row was filled from one, no judgement wrote one back, and the "external modification detected, reload offered" clause had no mechanism. A library imported from Lightroom came in and could not go back out. The scan collects `.xmp` beside `.drsc` from the listings it was already paying for, and the pull reads each one whose ETag has moved. Both namings resolve: darktable's `IMG_0001.CR3.xmp` names its file exactly, Lightroom's `IMG_0001.xmp` names the stem, and under the stem the JPEG beside a RAW is the same photograph and takes the same document, as DarkRoom's own sidecar already does. Each is reconciled with the catalog winning — keywords union, a rating or label taken only where the catalog has none — because a standard XMP carries nothing that could say whether its value is newer. A genuine disagreement is not resolved; it is written to a table, and the settings page offers the sidecars' values against it. That button is the reload the requirement asks to be offered, and the ETag that moved is the detection it asks for: an `.xmp` edited elsewhere is exactly a file the pull's ordinary incrementality re-reads. Writing goes the other way behind a setting that starts off, since NFR-R4 makes writes beside somebody's originals theirs to switch on. With it on, a judgement or a keyword rewrites the sidecar of whichever spelling exists, or creates Lightroom's. The record is read from the catalog whole at that moment rather than carried from the gesture, so a rating and a keyword a second apart are two writes of one file that agree. And the file's own title, caption, copyright and hierarchy come through the rewrite: the catalog has no columns for them, `rewrite` replaces the owned set wholesale, and a record that said nothing about them would have deleted them from a Lightroom sidecar on every star. The rating's two axes cross the format's one field both ways: a rejection is Adobe's `-1` and stars are stars, and stars arriving on a rejected frame lift the rejection, since the file said it was worth a number. An unrated file says nothing and clears nothing, on the rule the `.drsc` merge keeps. `versions.label` finally has a reader and a writer, with the code table moved out of the query so the two cannot drift.	2026-09-12 01:08:11 +02:00
dtourolle	d3b6127db6	Let a photographer name the state they liked, and go back to it or look at it ... FR-DEV-5 asked for named snapshots of an edit state and FR-DEV-7 for a comparison against a chosen one, and neither existed. The history stack is per sitting and forgotten with it, on purpose — the gap that mattered was an automatically saved mis-drag with no way back, and that was closed first. What was left was the other half: a state the photographer wants to keep *because* it is worth keeping, which is a different thing from a step and is not served by making the steps last longer. A snapshot is an edit state, and an edit state is exactly what a sidecar version stores, so it is stored as one: a `[version]` block carrying `snapshot-of = <uuid>`. The parameters, the masks and their parts, the repairs and the film all arrive through the blocks that already carry them, a merge keys on the uuid as it does for any version, and a build that predates the key reads the block as a named version and keeps it — the right failure. Only the pointer is new. The one reader that has to know is `default_version`, which must never answer with a snapshot: a file whose edit is missing is not a file whose edit is one of its saved moments. The snapshots of an edit are listed by that pointer, oldest first, the same on every device. Writing them back removes what this sitting deleted and puts in what it holds, and leaves standing whatever it never saw — a snapshot the other device took since the photograph was opened here is not this device's to remove by not knowing about it. That is the rule the version merge already keeps, applied one level down, and it is why the save carries the deleted ids rather than replacing the list wholesale as the masks are. Each is re-pointed at the uuid the save settled on, because the default may have been fused onto its canonical identity since the snapshot was taken. Restoring is one history step, so undo takes it back whole, as a paste is. Taking and deleting are not steps: they change nothing about the photograph, and an undo that removed a snapshot would be undoing a decision to remember. Holding the eye beside one renders the snapshot and hands the edit straight back — the same suspension "Before" uses, against a point the photographer chose rather than the file. Two sessions on the same photograph get ids that cannot collide, stamped with the second and a random word, because the merge folds equal ids into one.	2026-09-12 01:08:10 +02:00
dtourolle	369eb8fbf0	Put the log and the crash records in one file, and show it before writing it ... NFR-OPS-1 asks for a diagnostics bundle — the log, the schema version, the GPU and driver, the app version — "with an explicit preview-and-consent step before anything leaves the device". The log and the crash records have existed since August; what did not exist was any way to hand them over that was not `adb pull` and a knowledge of where the state directory is, which on the tablet the requirement was written for is nobody. Nothing here sends anything, and that is the design rather than a gap: crash.rs already says why a transport built ahead of the consent is the shape of thing that gets switched on by default. The bundle writes one text file to a place the user can find, so that they can attach it. That is the moment it leaves, and it is theirs. So the consent guards the write, not a send. Preparing gathers everything into memory and shows what would be written — each section, its size, what was taken out, and where the file would go — and only the second press puts bytes on disk. A user who reads the preview and presses the other button has changed nothing anywhere. The gathered bundle is held between the presses so what is saved is exactly what was shown, not a second gathering that differs by whatever was logged while they were reading. One text file rather than an archive, because a `.txt` opens wherever the user is sitting and pastes into an issue, and because the preview can then be the file rather than a summary of it. Every line goes through the blunter of the two redactions on the way in, whatever the sink already did to it: the log's own rule keeps paths, since a path read over `adb` is context, but a file meant to be attached to a public report by someone who may not read it first is held to the crash record's rule instead. The About page's graphics line gains the driver, which the requirement names and the adapter has always reported. And docs/outstanding.md is corrected on both OPS requirements: it said crash reporting was a log::error! hook and NFR-OPS-1 had nothing behind it, and neither had been true since 2026-08-30.	2026-09-12 01:08:10 +02:00
dtourolle	4574c35236	Let a part be left out of a mask without being taken out of it ... A layer built from parts was missing the one control a correction most often wants: seeing what it did. The question a subtracted gradient raises is whether it took only the sky, and the question a stroke raises is whether it filled the shoulder — and the only way to ask either was to remove the part and look, which answered the question and lost the part. The layer's own ring answers a different question, about the adjustment, and hiding eight layers to check one correction is not an A/B anybody performs. So a part carries `hidden`. It is an edit and a history step, as the layer's switch is, and it is folded into the render fingerprint because hiding a part changes the mask as surely as removing it does. Where the mask is built the shown parts are walked rather than the parts, which is what makes a hidden base hand the fold to the first part that is shown — and a revealed layer whose every part is hidden clears its slice rather than leaving whatever the last rasterisation put there to be read back. `covers` asks the same shown parts, so a layer whose only adding part is hidden costs no slice at all. In the sidecar the key is `hidden`, in the part's block or, for the base, in the mask block — under a word that cannot be confused with the layer's `enabled`, which has always meant the layer. Absent means shown, so no file written before the switch existed reads any differently. The row wears the same ring the layer does, one row down, because it is the same question about a smaller thing.	2026-09-12 01:08:10 +02:00
dtourolle	9cc52fd72b	Bind the two develop gestures that were described and not bound ... FR-DEV-16's book said resetting a control and hiding a mask layer were reachable by pointer and by finger, and stopped there. The reason was honest: the generated rows have no focus, so "reset the focused control" named a thing the panel could not point at. But a photographer at the keyboard means something narrower than focus. They mean the slider they just dragged too far, and that is a thing the panel can remember. So the Adjustments global keeps the last control moved — two indices, written where the panel forwards the change and cleared when the next photograph opens, so a reset cannot reach back into the previous edit through an index that happens to be shared. R puts it back, through the same callback the track's double-click takes, and is silent until something has moved. The mask layer needs no such notion, because the panel already has a selection: the rows the edge controls point at. H hides or shows those, through the path the ring at the head of the row takes, so it is an edit and a history step exactly as the ring is. A mixed selection goes to shown, since the layer nobody can see is the one being asked about. Both tags now carry the key, and the book says so.	2026-09-12 01:08:09 +02:00
dtourolle	fa4dca327f	Give the desktop executable a version flag and a Windows identity ... Three things the Windows build showed the entry point was missing, and that a Linux build never asks for. `--version`, answered before the logger and the crash hook install: a binary built on a machine that cannot run the application — the Linux CI producing the Windows executable, checked under Wine — needs an exit that proves it starts without opening a window or touching the user's directories. It is the smoke test in docs/windows.md §6. A GUI-subsystem executable in release, or Windows keeps a console window open behind the application for the life of the process. Debug builds keep the console, which is where their log goes. A resource block, or Explorer, the Start Menu and the taskbar show the generic executable icon and the Details tab is empty. build.rs wraps the PNG every other platform uses into an .ico at build time — an ICO entry may be a PNG, so the wrapper is a 22-byte header — and hands it to winresource with the version cargo already knows. The crate is an unconditional build-dependency because a cfg(windows) on one is evaluated against the host, which here is Linux; the script itself returns before touching it on any other target.	2026-09-12 00:54:11 +02:00
dtourolle	0a2c49dd10	Guard two constants the Windows target leaves unused ... secrets.rs names the keyring service and desktop_client.rs the socket timeout, and every use of both sits under a cfg that a Windows build does not satisfy — the placeholder secret store has nothing to file under, and the Nextcloud client's named pipe is not opened yet. The first cross-compile reported both as dead code, which is a failed clippy job the moment the Windows leg runs with -D warnings. Guarded by the same cfgs as their users, with the reason beside each.	2026-09-12 00:54:10 +02:00
dtourolle	b718c70b11	Specify a Windows installer built by the Linux CI ... The tree is closer to Windows than a Linux-only project usually is: every image library, the TLS stack and the inference engine are pure Rust, and dr-plat already keeps the Linux-only code behind cfgs with a loud fallback where none exists for another platform. What remains is a short list above dr-plat — five XDG path lookups, an xdg-open, the secret store's third implementation, the models' lookup beside the executable — and none of it touches core, which is the NFR-PORT-3 test this would be the first real run of. docs/windows.md decides the GNU target over MSVC-via-xwin, Vulkan only as on every other platform, a per-user NSIS installer that leaves the library alone on uninstall, and a CI leg in the shape of the Android one. It is explicit about what a runner with no Windows can verify — that it links, is PE32+, starts under Wine and installs under Wine — and what it cannot, which is everything involving a real GPU driver. Three FR-PLAT-WIN requirements and a channel row record the decisions; the ordering puts a first cross-compile on the developer machine before any container exists, because the list of cfg gaps is a reading of the source and the compiler's list will be longer.	2026-09-11 23:30:16 +02:00
dtourolle	4f31123b0c	Let the user choose which SCRFD finds their faces ... faces.md §12.3 measured what the cheapest detector costs: the small faces in every group shot, and a dog embedded a dozen times. Which trade is right depends on the machine doing the sweep — a desktop left overnight and a tablet on a battery want different answers — so the detector is now a per-device setting, Fast / Balanced / Thorough on the settings page beside the indexing button, persisted with the rest of the settings file. A detector is half of a model id. Every face, marker, shard and calibration is keyed on faces.model_id precisely so that a model change is a new id and a re-index rather than a silent change under existing data, and a detector change is a model change: it decides which faces exist and where the landmarks that align them land. So each choice names its own pipeline. 500M keeps the bare "w600k_mbf" every existing library was written under, so an upgrade disturbs nothing; the others are qualified. Choosing one restarts coverage from zero under the new id, the sweep re-detects, confirmed names carry across by box overlap, and the sync shards are keyed by the same id so a peer on another setting neither adopts nor pollutes them. The library controller carries the id into the sync the same way it carries the cache budget, because the sync starts from places that have no settings in reach. All three shape-fixed exports ship — APK, Arch, Flatpak — since a tablet has no other way to obtain the one it was not installed with; the APK grows by twenty megabytes for the choice.	2026-09-11 22:12:53 +02:00
dtourolle	9d35addd86	Measure what the cheapest SCRFD actually costs in faces ... §1 chose scrfd_500m on FLOPs and never measured the recall it gave up. A dr-ui example now runs several detectors over the same sample of stored proxies, matches boxes by IoU against the first, buckets the result by face size, times each, and writes contact sheets of the disagreements in both directions — because a count of extra faces says nothing until someone has looked at whether they are faces. Over 400 proxies from the reference library: 2.5G finds 14% more faces for 12% more time, 10G a further 12% for 3.1× the time. The extras are small real faces. The 86 faces only 500M found are a dog a dozen times, a stop sign, a wheel and the backs of heads. Recorded in faces.md §12.3.	2026-09-11 22:12:39 +02:00
dtourolle	3d6d69ec90	Wrap the face-sweep repair match the way rustfmt wants it ... CI's Desktop job failed at the Format step on 16f3fb4: rustfmt puts the `match faces_without_proxy(...)` on its own line under the `let` and re-indents its arms, and the commit was written without running it. No code changes; only the layout of that one match in library.rs.	2026-09-11 22:00:38 +02:00
dtourolle	16f3fb41a3	Measure the faces already found rather than finding them again ... Every face stored before its quality was kept holds a unit vector, and V14 forgot the run marker of each image holding one so that the next sweep would look again. Looking again meant detecting again: a whole re-detection per image, with every suggestion on it thrown away and the confirmations carried across by box overlap, to recover one number. The sweep now has a measuring pass between the proxy repair and the un-indexed images. It lists every image holding an unmeasured face, fetches the original once, warps each stored face from the landmarks it already has, embeds it, and writes the raw vector and its length over the old row. Ids, boxes and identities are untouched; the marker is re-written fresh so the sync exports the measured vectors. A face whose landmarks no longer make a warp is dropped, as detection would have refused to store it. `faces_unindexed` leaves those images to the measuring pass, so the V14 deletion no longer costs a second detection.	2026-09-11 21:50:12 +02:00
dtourolle	8b3abdb787	Keep each face's quality, and never compare against a poor one ... The embedder's raw output has a length, and the length is a reading of how recognisable the crop was: a blur, an occlusion or a hard profile comes out short. Normalising threw it away. A short vector sits near the middle of the sphere and matches a little of everyone, which is how one bad crop bridges two people in a grouping pass. So the length is kept — the store now holds the raw vector, re-normalised on load, with the length beside it as `faces.quality` — and a face under MIN_GALLERY_QUALITY (14) is a probe: measured against the gallery and placed where it fits, but never what another face is measured against. Two probes are never paired, and a probe is nobody's evidence for a confidence. The People screen shows the number as "Quality 17.3", dimmed below the floor. Faces indexed before this stored unit vectors and have no reading; they are admitted to the gallery, and schema V14 forgets the run marker of every image holding one so the next indexing pass measures them. A peer's unmeasured shard faces are not adopted, or a sync would write that marker back.	2026-09-11 21:50:12 +02:00
dtourolle	a87139b838	Give every mask an eye and a colour, and put the brush where the mask is ... The first build of seeing a mask showed the selected layer's, in one global style, from a strip at the top of the panel. It answered the wrong question and answered it somewhere nobody looked. What a photographer asks of two masks is how they meet — where the sky's edge sits against the building's — and that needs both on screen at once, in colours that can be told apart. So each row of the stack has an eye, drawn in the colour its mask is shown in, and each mask has six swatches to choose that colour from. Several can be open at once; a new one comes up open, in the first colour nothing else is using. The style — tint, alpha, outline — is the one setting that stays global, above the stack, because three styles at once are three pictures that cannot be read against each other. Alpha now draws every shown mask, each in its colour, on black. In the pipeline a `Reveal` is a list of `(layer, colour)` rather than one layer, and every reveal block carries its own colour. The brush moves too. Select, Paint and Erase and the three sliders under them sat at the top of the panel, appeared only once a row was selected, and said nothing about which mask they acted on — so "how do I paint" and "how do I correct the model's outline" both had the same answer and nobody found it. They sit under the selected mask's parts now, beside the swatches, and on a subject or a category the hint says what a stroke there does: it becomes a part of this mask, joined to the model's, and can be taken out again. Eyes and colours are viewing state, on the session and not on the layer, so a photograph reopened has every eye closed — the stored-mask round-trip test asserts it.	2026-09-11 19:03:51 +02:00
dtourolle	5d175cc668	Let a press on the photograph reach the tool that was armed for it ... The brush did nothing, and neither did three other things nobody had tried lately: clicking a subject on the photograph to select it, placing a repair, and sampling a neutral. All four are TouchAreas over the canvas, and all four sat behind the pan/zoom area, which is full-canvas and enabled for everything but a crop. It took every press in the viewport and they were never offered one. Slint hit-tests siblings front-to-back (`send_mouse_event_to_item` visits children `TraversalOrder::FrontToBack`), a TouchArea answers `GrabMouse` on any press it is enabled for, and the first grab aborts the traversal. Front means *last declared*. Each of the four carried a comment saying it sat "above the pan/zoom area so a click reaches it first" — true of the order they were written in, and backwards. Nothing about the geometry decides this, so nothing about the geometry could have fixed it. The pan area is declared first now, as the backstop it always meant to be, and the rule it leaves behind is that the general case goes above the specific ones. `GradientHandles` is the other end of that rule and is why dragging a handle has worked all along while everything between it and the pan area did not. The order is asserted in a test, because this is a fault that compiles, passes every other test, and silently removes four tools at once.	2026-09-10 20:56:10 +02:00
dtourolle	193b35a249	Start a category mask where the photograph can bear it ... Clicking "architecture" made a layer whose mask was gone. Every category layer began at STRICTNESS_DEFAULT, and that constant was fitted on the synthetic sky the refine tests build — its own note warns that a real photograph's noise "moves every crossing down together", which turns out to be a considerable understatement. Measured over seven ordinary frames, half scale removes 76% to 99.5% of `architecture`, 36% to 93% of `ground` and 18% to 91% of `vegetation`. Only sky, the category the number was calibrated against, survives it. An empty mask is indistinguishable from a broken one: the layer is listed, the adjustment moves, and no pixel changes. So what this looks like from outside is that the segmentation does not make masks at all. No smaller constant fixes it either, because a nat of evidence means different things over a smooth sky and over a stone facade — the useful position is above 5 on one frame and below 1 on the next. So the frame is asked instead: `Refinement::gentle` walks down from half scale and takes the first rung whose gate removes no more than a sixth of the category's weight, and the model's own outline when none of them does. One `apply` on a friendly photograph and four on an unfriendly one, paid when a layer is made rather than for eight categories nobody masked. The slider's reset went to 4 as well, so taking the control back to its "default" emptied the mask. It goes to zero now, which is the one position documented to mean something: exactly what the model weighted.	2026-09-10 20:27:40 +02:00
dtourolle	404fea47a8	Wrap the lines the merge resolution left long ... `cargo fmt --check` failed the desktop job, on three files and for one reason: routing the mask handlers through the `Masking` global was done by substituting the call prefix, which is a text edit rather than a Rust one. It left `window.global::<Masking>().on_part_join_picked(...)` on a line that had been short enough as `window.on_mask_part_join_picked(...)` and no longer was. Formatting only. The whitespace-stripped source is identical in the two `ui/` files; the third differs by the trailing commas rustfmt adds when it breaks a call across lines. The matrix moves with it, because the tags shift by a few lines and the check compares line numbers.	2026-09-08 08:57:32 +02:00
dtourolle	2d878c2117	Offer the film stock in its own group, and let its list scroll itself ... Two faults in one control, both reported from the tablet. The stock picker appeared in every group. It is not a parameter, so it is not a row, so the filter that hides every other control when a group is chosen never saw it — "Kodachrome" sat at the top of Light, of Colour and of Detail alike. Three places it does not belong, and the one it does no more prominent than the rest. The descriptor has said `Effect` and only `Effect` since the film moved there; nothing was asking it. So the panel now asks. It cannot ask directly — a generated panel may not know which operation a control belongs to — so the session answers, from what the operation declares it is about, and a stock re-declared as something else would move on its own. The flag is recomputed when the group changes as well as when the film does, which is the half that would have made it stale exactly when it mattered. And the open list was unbounded, so it made the develop column taller and the column scrolled as one: reaching Velvia dragged every slider below it off the screen, an answer given once pushing aside the controls used constantly. It now scrolls within a bounded height of its own. That viewport is counted rather than measured, for the reason the tool rail records a few files away: a viewport that asks a layout how tall it wants to be, while the layout takes its height from the viewport, is a cycle Slint settles by handing back the height it was given — and the content is then clipped in silence rather than scrolling. Every row here is one fixed height, so multiplying is exact. The group rule has a test. The scrolling does not, and cannot: it is a layout, and a layout fault is invisible to the compiler and to every assertion that can be written about it.	2026-09-07 20:42:15 +02:00
dtourolle	4c217c9be6	Show what a control does to a photograph, one parameter at a time ... A node that has just been declared can be read, reasoned about and tested, and none of that answers the question a photographer asks first: what does moving this do to the picture. Colour grading, dehaze and the range masks were all argued into the tree on their behaviour and none of them had been *looked* at. So two diagnostics. `sweep` walks one operation from its minimum to its maximum and writes a frame per step; `rangesweep` does the same for a mask band, which is not an ordinary parameter — it lives on a layer, is rasterised by its own pass, and only becomes visible through whatever adjustment the layer carries, so it gets two stops down to make the selection legible. `sweep` names no operation. The id arrives as a string and the parameters and their ranges come from the graph's own capabilities, so a node declared yesterday sweeps on the same terms as one that shipped a year ago — the property `ops/README.md` promises, used rather than asserted. Three things it learned the hard way and now records. It renders through `render_detailed` unconditionally, because the fused path refuses a shader composed with a detail stage rather than rendering it wrongly, and that call falls through when there is no such stage. It takes `SWEEP_HOLD`, because a parameter grouped under one widget is not meaningful alone: a hue with no strength behind it renders the same frame every time, which reads as a broken node rather than a correctly declared neutral. And it bounds the output, since a 25 MP frame is a 75 MB PPM and a sweep is hundreds of them. Both read a rendered file as readily as a raw one, so a JPEG can stand in where no raw is to hand — on the terms `from_rgba8` documents, with the controls still working and their neutral being what the camera left rather than what the sensor recorded.	2026-09-07 20:01:08 +02:00
dtourolle	e44cb8cbe0	Regenerate the traceability matrix for the rebased line numbers	2026-09-07 20:01:01 +02:00
dtourolle	0a5eab0487	Wire the develop panels through globals, so a second copy is one line ... Every panel in the develop column declared its inputs and its callbacks and had `app.slint` bind each one to a property or a callback on the window root. That is fine while a panel is drawn once. N9 draws them a second time, in the portrait dock, and the wiring is what would have to be copied: `MaskPanel` alone ran to forty lines of forwarding, and a callback added to one copy and not the other compiles, renders, and simply does nothing on the layout nobody was looking at. So the wiring moved to Slint globals. A panel reads the global and calls the global; Rust hooks the global instead of the window; and the instantiation in the column is now the panel's name and a pair of braces — every one of the ten children of the column, with no property that differs by placement left to supply. There is a global per panel family rather than one for all of them, and the reason is an import cycle. Each panel's model struct — `ParamRow`, `MaskRow`, `HistogramView` — is declared in the panel's own file, so a single global holding `[MaskRow]` and `[ParamRow]` would have to live in a file importing `masks.slint` and `adjust.slint` while both imported the global back, which Slint rejects. Breaking that needs six model declarations relocated, which is a change to the data model and not to the plumbing this is about. A global beside the panel it serves also lets each name drop the prefix it was carrying only because the window root is one flat namespace: `root.spot-radius` is `Repair.radius`, and `root.peaking-on` is `Peaking.showing`. `session.slint` is new and holds the two facts every family needs and none of them owns: whether there is an open photograph to edit, and which mode the view is in, with the three readings of the mode derived once instead of at each of the dozen places that tested one. `ViewMode` moves there from `adjust.slint`, where it was only ever a lodger. Nothing on screen changes. What is not here: the tool rail and the status strip still take their properties at the instantiation, because they are drawn once and N9 does not copy them; the preset sheet's own state stays on the window, because the library grid opens the same sheet and a global cannot bind the window's state — which is why `Transfer.open-presets` is handled in `presets.rs`, beside the summary it already had to compute.	2026-09-07 20:01:01 +02:00
dtourolle	30468c4c69	Put the window-metrics doc on the function it describes ... The comment explaining why both coordinate systems go on one line was written for `log_window_metrics` and sat above `window_metrics_level`, where it read as the start of that function's much longer note. Two doc blocks ran into each other and the one that prints had none.	2026-09-07 20:00:50 +02:00
dtourolle	428d8c4a51	Say what size the window actually is, in both coordinate systems ... Every figure in D-N7's table was computed at a guessed scale factor. The tablet's panel is 3000 by 1920 physical and nothing in this repository has ever recorded the density Android reports for it, so the dock's width is either 900 or 1037 and its available height is 200px either way. N6 asks for the measurement; this is the line that carries it. Beside the existing `apply_layout_class` call, because that is where the window is already being asked for its size and its scale, and again on every resize, so turning the tablet over records the other orientation in the same logcat. Both coordinate systems on one line: a logical size cannot be checked when the scale is the thing in doubt, and a physical size that does not divide by the scale printed next to it says the reading is of something other than the panel. The level is not fixed, because none of the three obvious choices works. `android_main` caps the facade at info, so debug never leaves the device and a debug-only line answers nothing. A drag emits a resize per frame and each accepted record is also appended to the on-disk log, so info on every resize is not a diagnostic. And the first reading is not the settled one: on X11 the window reports 0x0, then 360x320 at scale 1.0, then 1100x720 at scale 2.0, so reporting only the first would put a number in logcat that is not the window's. So the pair that decides is the scale factor and the orientation -- exactly what N6 is asking for, and exactly what a drag leaves alone. A window with no area is not a reading and records nothing. Every later change to either half, the scale resolving or the tablet turning over, is a new answer and goes out at info; everything else is debug.	2026-09-07 20:00:49 +02:00
dtourolle	df741a8a49	Let one mask be built from more than one selection, and paint into it ... A mask the model draws arrives approximately right — stopping inside a shoulder, leaking into the hair — and FR-DEV-3's edge controls move the *whole* boundary, so no value of feather or dilation fixes two errors that go opposite ways. What fixes them is a second selection joined to the first, and a layer that held exactly one source had nowhere to put one. The brush the core has had all along was reachable from no control in the application. A layer is now an ordered list of parts. Each names a source and how it joins the mask before it — added to it, or taken out of it — and carries its own edge treatment, because a model's soft coverage and a stroke painted where it stopped short do not want the same feather. Invert and opacity stay on the layer, where the composed shader already reads them. The sidecar grows `[part]` blocks and nothing else. A layer of one part writes exactly the bytes it always did; a mask block with no part blocks after it reads back as one part; and a stroke, a join or a source this build cannot read costs that part rather than the layer. So every sidecar in every library still parses to the edit it always was. On the device the parts fold into the layer's one slice, so eight layers still cost eight channels: union is a `max` blend and subtraction is the erase blend the brush already used. A part is drawn into a scratch texture before it is joined, and that is not incidental — an erase stroke means a hole in *that part*, not a hole in the mask, and drawn straight onto the accumulator it would punch through the subject underneath. A layer of one part skips all of it and takes the path it always took. In the interface: a part list under the selected layer with a chip saying which way each joins, Add and Subtract beside it, a Select/Paint/Erase strip with the brush's size, hardness and flow, and a drag on the photograph that paints. Pressing Paint on a mask that cannot hold a stroke joins a part that can, rather than explaining that a subject is not a brush. A whole stroke is one step in the history. The edge controls now shape the part that is selected rather than the layer, which is the one behaviour change to an existing control: with a correction selected, the feather slider softens the correction and leaves the model's mask alone.	2026-09-07 20:00:40 +02:00
dtourolle	1e171c6d31	Let a collection be picked up, rearranged, and emptied after the fact ... Collections could be made and filled and never reorganised. Nesting had a drag; un-nesting had nothing, in either direction — "All photographs" refused every drop, which is right for a photograph and wrong for a collection, which has a top level to be returned to. So a collection put inside another was in there permanently. Right-click deleted an *empty* collection outright and refused otherwise, which is wrong in both directions at once: destructive with no confirmation, and no way at all to delete a collection that held anything without emptying it by hand, child by child. And a photograph could only leave the collection the grid was scoped to, since that is the only one a button in the header can name — the cell's badge says a photograph is in three collections and never which three. Three ways in, one vocabulary: **Hold a row.** The tree is inside a Flickable, which claims any drag beginning inside it, so with a finger a drag on a row is a scroll until something says otherwise. The hold is that something. It lifts the row — drawn before anything moves, so the gesture says it has been understood — and then what the user does decides which of two things they meant: move, and it is a rearrangement; let go, and it is the row menu. The same fork the grid already uses to tell hold-to-select from drag-to-file. `decide_release` is that fork, and it is tested, because getting it wrong one way puts a sheet over every tidied tree and the other way makes the menu unreachable by touch. **The row menu.** Rename, new collection inside, move to top level, keep offline, delete. Deleting asks once when there is anything to lose and says what survives: the photographs stay in the library, and nested collections move up rather than going with it — which is what the catalog does, and what a user would never assume. An empty collection goes on the first press, because a dialogue about losing nothing is how people learn to dismiss dialogues. **"Collections…" on a selection.** Every collection the selection is filed in, each with a count — "3 of 40", so nobody takes forty photographs out of a collection thirty-seven were never in — and a way out of any of them without navigating there first. The long press used to open the offline question by itself. That question is one item in this menu now: there is one hold per row, and while it was spent on a single action nothing else the tree can do had a touch route at all. Nothing is lost — the tray on the row keeps its tap, and the question gains a full-width control in place of a 30px icon in a row shorter than the touch minimum. The row-press handler moves to `collections_ui` with the rest of what a collection row does; it lived in `library_ui` only because it opened that prompt.	2026-09-07 19:59:44 +02:00
dtourolle	577bbd82b0	Return the action the drag offered, not the one this row prefers ... Slint negotiates a drag action between source and target, and the runtime clamps whatever `can-drop` returns against the set the source allowed: an action outside it becomes `none`. Both drop targets here named a constant instead of echoing what was on offer, and each named the wrong one for half its traffic. A collection row takes two kinds of payload. Photographs come from a DragArea allowing `copy`; a collection being nested comes from one allowing `move`. The row asked for `copy` unconditionally, so images filed correctly and every collection dropped on a collection was refused — nesting by drag has never worked. The trash had the same fault mirrored: it insisted on `move` while the grid's cells allow only `copy`, so it refused every photograph dragged to it. Neither failure had anything to see. A clamped action is delivered as a refusal, which looks exactly like a target that declined on purpose, so the drag simply did nothing and left no error to search for. `decide_drop`'s Reparent branch was tested and passing throughout. It tests the decision, not the negotiation, and nothing was reaching it.	2026-09-07 19:59:44 +02:00
dtourolle	bac5801618	Ask which collections a selection is filed in, and how much of it ... `collections_for_image` answers this for one photograph and has no counts, which is enough to badge a cell and not enough to offer a removal: with forty selected and three of them in "Iceland", a sheet that says only "Iceland" invites the user to take all forty out of a collection thirty-seven were never in. `membership_of` returns the count alongside the name so the row can say "3 of 40". Chunked over the image list rather than one `IN (...)`, because the list is a selection and a select-all makes it as large as the library — past SQLite's bound-parameter cap on exactly the gesture most likely to produce it. Counts are summed across chunks, so the answer is the one the unchunked query would have given. Smart collections are excluded by construction: they have no member rows, so there is nothing a removal could do.	2026-09-07 19:59:44 +02:00
dtourolle	2cd49d1cb7	Measure the rail by counting it, so it can scroll instead of clipping ... With the adjustment groups in the rail, a short window silently lost them. At a 1500x680 window the rail drew Photo, Compose, Local, Repair, the seam and "All", and then stopped: Optics, Light, Colour, Effects and Detail were not scrolled off, they were gone, with nothing on screen to say so. The develop view's primary navigation, unreachable by any means. The Flickable was put here to prevent exactly that, and it could not, because its viewport asked the layout how tall it wanted to be while the layout was already taking its height from the viewport. Slint settles that cycle by handing back the height it was given, so `max(self.height, preferred-height)` could never exceed `self.height` and there was never anything to scroll. Counting breaks the cycle. Every entry here is a fixed height by construction — a tool is `rail-entry-height`, a group is a touch target — so the content is six plus four fifty-fours plus a gap plus a touch target for each group and "All", which is exact rather than an estimate and depends on nothing that depends on it. Four tools and five groups come to 498, against the 340 a 680-pixel window leaves at 2x, and the difference is now scrollable rather than absent. Found by shrinking the window with the groups forced into the rail. The interaction itself is unverified: synthetic input does not reach a Slint window on this desktop, and the tablet was disconnected, so what is confirmed is the arithmetic and the clipping it explains, not the scrolling it should restore.	2026-09-07 00:17:08 +02:00
dtourolle	3f6dbce2aa	Name a lone control after its operation, so three cannot all read "Amount" ... The Detail group ended with three consecutive sliders labelled "Amount" and nothing to tell them apart. They are dehaze, clarity and texture: each declares exactly one parameter, and `ops/README.md` tells an author to reach for the `amount` kind first, so all three named it the same thing. The panel withholds a heading from a group of one, and the reasoning it gives is sound — a lone control names itself, and the group reset it loses costs nothing because the slider already resets on double-click. But that argument rests on the parameter being named after what it does. It holds for exposure, contrast, vibrance, saturation and brilliance, whose single parameter shares the operation's name, and it fails for the three whose parameter is called after its kind rather than its subject. So a lone parameter now takes its operation's label — the name the withheld heading would have carried. For the five that already agreed, nothing changes. Found on the tablet, and only there: every row was correct, every label resolved, and the panel was still unusable. The test that guards it asserts over the real chain rather than a fixture, because the fault was a property of what is actually declared — a fixture would have had to be written to reproduce it, and would then only have proved itself.	2026-09-06 19:01:54 +02:00
dtourolle	124b2d99c6	Take the first control that moves the picture, not the first one drawn ... `showing_the_original_leaves_the_edit_exactly_as_it_was` set row zero to its maximum and then asserted the photograph was modified. It was not, and the test failed on its own premise rather than on the thing it exists to check. `EditGraph::capabilities` puts the lens corrections at the head of the list, matching where they sit in the shader. Those carry profile coefficients rather than parameters, so with no profile loaded a slider on one is a control with nothing behind it: the graph stays neutral and the premise assertion fires. The test was written against a panel whose first row happened to be an adjustment, and it stopped being one. So it now walks the rows until it finds a control that actually changes the edit, which is what it meant by "row zero" all along. Still addressed by index, so it still names no operation, and it no longer depends on where in the chain the first *adjustment* happens to sit.	2026-09-06 19:01:53 +02:00
dtourolle	3994caba12	Write down the develop gestures, since only their author knew them ... FR-UI-4 says a gesture with no visible counterpart is a feature only its author knows about, and the vocabulary the application actually publishes had two sections in it — the library grid and people. Develop had none. Every one of its gestures was documented in the comment beside the `TouchArea` that implements it, which is where the previous sixteen were before this scanner existed, and unreachable to anybody not reading the source. Thirteen now carry tags: magnify by pinch or wheel, pan a magnified frame, fit and 1:1, hold to see the original, sample a neutral, undo and redo, step through the folder, reset one control, show or hide a mask layer, choose a group of adjustments, and copy and paste the settings. Each has a pointer and a touch route, so none of them is keyboard-only. Three bindings were genuinely missing and are added here rather than merely described. Ctrl+C and Ctrl+V for the settings clipboard, which the Settings panel's own comment has claimed existed for as long as the panel has and nothing bound; and [ and ] to step through the adjustment groups. The groups are whatever the operation set declares itself to be about, so there are as many as the pipeline has and no key can name one of them — stepping is the binding that survives a node being added, and "everything" is part of the cycle rather than a way out of it. Two are described and not bound. Resetting a control from the keyboard and toggling a mask layer from the keyboard both need a notion of which control or layer has focus, and the generated panel has none — the rows are a model the repeater rebuilds, and inventing a focus ring for them is a larger change than a keyboard shortcut. Both are reachable by pointer and by finger, and the tags say so rather than promising a key that is not there.	2026-09-06 19:01:52 +02:00
dtourolle	901f51e6c4	Point at something grey and let the pipeline work out the rest ... FR-DEV-3 has asked for "white balance (temperature/tint, and picker)" since it was written, and only the first half existed. `WidgetKind::WhitePoint` was in the vocabulary and `develop::supported` answered false for it, so the node degraded to two sliders — correct behaviour that had quietly become the only behaviour. Sampling a neutral is the first move of the global tonal pass and every colour judgement afterwards is measured against where the grey was put, so guessing at two sliders until a wall stops looking green is the wrong way round. The awkward part is that a picker genuinely needs to know how far a hundred units of temperature move red against blue, and that number is declared in the node's own file. So the inversion lives in `dr_pipeline::neutral` rather than in the interface: the canvas hands over a colour, the core finds the operation that asked to be driven by a pixel and bisects its declared response until the sample comes back grey. Nothing in `ui/` names white balance, and nothing holds a second copy of a response that would be wrong the first time somebody adjusted the range. A bisection rather than a closed-form inverse because only monotonicity is part of the bargain — the expression is free to become a table tomorrow. The result is rounded to the precision the control is drawn at, which is not cosmetic: unrounded, sampling something already neutral lands a ten-thousandth off zero, and the photograph comes back modified with an undo step for a correction of nothing. On the panel side this needed one distinction the generated path was missing. `is_on_canvas` was being read as "and so the panel draws nothing for it", which is right for a crop — four edge fractions are not controls anyone drags in a list — and wrong for an eyedropper, which *writes* temperature and tint and leaves them exactly the controls a photographer reaches for next. So a sampling widget keeps its sliders and puts the affordance that arms the canvas in the group's heading, built like the reset beside it. One click, one sample, one history step: `Edit::Action` never coalesces, and there is no hover preview to fill the stack with temperatures nobody chose. Declaring the presentation also groups temperature and tint under one undo step, where they were two. That follows from what `Presentation` means and reads correctly — white balance is one decision — but it is a change, and worth saying so.	2026-09-06 19:01:52 +02:00
dtourolle	2584b9ecbc	Hold one key to see the photograph before you touched it ... FR-DEV-7 asks for the current edit against the unedited original and nothing implemented it. What the develop view had was history navigation, which *changes* the edit rather than previewing against it — so the only way to look was to undo, look, and redo, and that puts two real steps on the stack at exactly the moment a photographer suspects they have overcooked a frame and is least sure of what they are doing. Holding the "Before" button, or backslash, renders the graph with every adjustment stripped and hands it straight back afterwards: the same suspend-render-restore shape the crop overlay already uses to show an uncropped frame and an export uses to suspend the zoom. Nothing is recorded, no rows are re-synced, and the photograph is still modified when the key comes up — the panel goes on describing the edit the photographer has, because only the canvas is answering a question. The framing deliberately stays on. A held comparison is a question about tone and colour, and re-cropping the canvas under someone's thumb would move the detail they are comparing; worse, the zoom is a rectangle of the *framed* image, so dropping the crop at 4× would quietly show a different part of the photograph rather than the same part unedited. What the crop took away is already compared in Compose, which shows the whole frame. Not a split screen: that halves the working image on the tablet this column was sized for, and the comparison photographers describe making is a flick back and forth rather than two pictures side by side. Press-and-hold is one gesture on a finger and on a mouse, which is what FR-DEV-3b's mapping wants, and it has no mode to be stranded in — the button reports both edges, so a press the system cancels puts the original down too.	2026-09-06 19:01:51 +02:00