Skip to content

Releases

1.0.2

Choose a tag to compare

@github-actions github-actions released this 16 Nov 13:15
fb3aaaa

Changed

  • A stable tag publishes server.json to the MCP registry. publish.yml has an mcp-registry job after the GitHub Release that signs in through GitHub OIDC and publishes io.github.anilcancakir/tapir, so the registry listing follows each release instead of a manual publish. (.github/workflows/publish.yml)

Fixed

  • tapir says where --cdp-port gives up tapir_evaluate. --cdp-port makes dowel run flutter on the web-server device, which has no VM Service evaluate, yet the quickstart, the skill and the README recommended the flag with no caveat, so an agent that followed them called tapir_evaluate and got an SDK error. The tapir_evaluate descriptor, skill rule 16 and its MCP reference, the MCP tool reference, the quickstart, the README and llms.txt now say it, and tapir:doctor prints an Evaluate: unavailable info line for any session that records a cdpPort. The error text itself is dowel's (1.0.4). Fixes #91. (lib/src/tapir_dowel_provider.dart, lib/src/commands/tapir_doctor_command.dart, skills/tapir/, doc/)

  • tapir_dismiss_modals, tapir_reset_overlays and tapir_navigate no longer freeze the app over a modal that guards its exit. go_router refuses the pop of a GoRoute with onExit and decides in a microtask, so the synchronous popUntil re-popped the same route forever and the page stopped answering until a reload. Modals are now popped one at a time; a guarded one is popped once and given three frames to leave. One that leaves is counted and dismissal carries on; one whose guard refuses stays open, is not counted and ends dismissal on its Navigator. A call that arrives while another dismissal is waiting on a guarded modal joins it and gets the same count, so the modal is popped once. tapir_navigate then reaches its adapter as before, and its URL check still decides navigated. (lib/src/extensions/ext_modal_router.dart, lib/src/tapir_dowel_provider.dart, skills/tapir/, doc/)

  • tapir:doctor accepts the MagicDevtools.installPost() wrapper as Magic wiring. An app on magic_devtools 1.x that calls MagicDevtools.installPre() and MagicDevtools.installPost() got Magic detected but MagicTapirIntegration missing; wire it with tapir:install, because the Magic-init check looked only for MagicTapirIntegration.install, which installPost calls itself. Either MagicTapirIntegration.install() or MagicDevtools.installPost() now reads as Magic-stack detected, integration wired; installPre() alone, or neither call, still gets the advice. (lib/src/commands/tapir_doctor_command.dart, doc/commands/tapir-doctor.md)

  • tapir_set_checkbox waits for a control that flips after a round trip. A switch the app updates only once its PATCH answers still showed the old state when the handler read it back right after the tap, so the call answered verified: false and the CLI exited 1 although the toggle worked and persisted. While the read differs from the requested value it is now repeated every 50 ms for up to 1.5 s, with no second tap; a control that flips at once answers at once, and one that ignores the tap still reads verified: false after the window. The response shape is unchanged. (lib/src/extensions/ext_checkbox.dart, lib/src/tapir_dowel_provider.dart, skills/tapir/, doc/mcp/tool-reference.md)

  • tapir:install leaves an app wired through MagicDevtools.installPre() and installPost() alone. The installer looked only for TapirPlugin.install() and MagicTapirIntegration.install(), which the magic_devtools wrapper calls itself, so it added a second TapirPlugin block, a MagicTapirIntegration block, an unused package:tapir/tapir.dart import and a WidgetsFlutterBinding.ensureInitialized() line (an app whose zone wrapper owns the binding did not want one) to an app that was already wired. installPre() now stands for the tapir block, its import and the binding line, installPost() for the Magic block; an app that calls only installPre() still gets the Magic block. The wrapper counts only when the resolved magic_devtools ships lib/tapir.dart: a release that predates tapir has the same wrapper names but installs the Dusk integrations, so the installer still wires tapir and warns to upgrade. tapir:doctor applies the same rule to installPost(). (lib/src/commands/tapir_install_command.dart, lib/src/commands/tapir_doctor_command.dart, doc/commands/tapir-install.md)

1.0.1

Choose a tag to compare

@github-actions github-actions released this 15 Nov 13:11
08253e4

Changed

  • The pub.dev topic mcp-server is mcp. Only tapir and crumbs used the first; the second is shared by many MCP packages, so the listing is easier to find. (pubspec.yaml)

Fixed

  • tapir_wait_for_network_idle refuses a pollIntervalMs below 100 as an invalid_param. The 100 ms floor was an assert, which a profile build strips: there pollIntervalMs: 0 never advanced the elapsed counter, so the wait never ended. The handler checks the floor itself now and answers -32602. (lib/src/extensions/ext_wait_find.dart)
  • The network-idle wait is described as the code behaves. The tool description, the tapir:wait_for_network_idle help, the tool reference, the skill and llms.txt say that the count is zero unless an adapter or a reader the app installs reports one: pendingHttpCountReader is a public function an app can replace with its own counter, crumbs is one option and not a requirement, and MagicHttpFacadeAdapter.pendingCount comes from magic_devtools, not from crumbs. The call matches after ceil(idleMs / pollIntervalMs) consecutive zero samples (the first sample counts, so the defaults match after about 400 ms and a window of one poll interval or less matches on the first sample; a request that starts and finishes between two samples is not seen), not on a confirmed contiguous idle window. timeoutMs is checked after a sample that did not complete the window, so a window that completes at or after it still returns matched: true (with the defaults, timeoutMs: 300 matches at about 400 ms) and only a deadline reached earlier is a timeout error. They no longer say the call returns immediately without crumbs or that timeoutMs is the maximum total wait. (lib/src/tapir_dowel_provider.dart, lib/src/commands/tapir_wait_for_network_idle_command.dart, skills/tapir/, llms.txt, doc/mcp/tool-reference.md)
  • The network-idle notes no longer name a crumbs adapter that does not exist. The skill said crumbs's DioHttpAdapter reports 0, but 1.0.0 of crumbs removed it and pendingCount has no default body. Only an adapter that implements pendingCount (MagicHttpFacadeAdapter, in magic_devtools) moves the count, and an interceptor that only calls CrumbsStore.recordHttp counts nothing. (skills/tapir/, llms.txt)
  • One install command. tapir depends on the Flutter SDK, so the README, llms.txt and the skill say flutter pub add tapir, as the installation page already did.
  • The install.yaml message and the installation page count the Wind contract right. The contract carries seven keys (className, breakpoint, brightness, platform, states, bgColor, textColor), not six, and the Magic enrichers are listed as examples because there are more than the five named. (install.yaml, doc/getting-started/installation.md)
  • The docs say which flags print the Wind fields. tapir:snap prints the wind: block only with --includeEnrichers, and tapir:observe returns just breakpoint and states unless you pass --includeEnrichers=full. The block appears under each W-widget that has a className (not WAnchor, WBreakpoint, WindAnimationWrapper or WKeyboardActions), with bgColor and textColor only when they resolve. The installation page, the README and the tapir:snap page (which still said six fields) now say so. (doc/getting-started/installation.md, README.md, doc/commands/tapir-snap.md)
  • One startup figure for the fast wrapper. The install message and the installation page quoted about 110 ms and about 50 ms; both say about 0.2 s, which is what ./bin/dowel measures on a cached build (0.17 to 0.35 s), and the "~3s" for dart run tapir says 0.4 to 2.7 s (warm and first run), the figures dowel's docs use. (install.yaml, doc/getting-started/installation.md, lib/src/commands/tapir_install_command.dart)
  • The tables in the command pages are left-aligned. The 18 pages under doc/commands/ used the unaligned |---| separator; they use |:---| like the rest of the docs. (doc/commands/)
  • The tables in the README, ARCHITECTURE.md and the skill are left-aligned too. (README.md, ARCHITECTURE.md, skills/tapir/)

1.0.0

Choose a tag to compare

@github-actions github-actions released this 14 Nov 19:06
b69aae1

Added

  • TapirPlugin.isDisabled. Whether the build set --dart-define=TAPIR_DISABLE=1 (or true, yes), read once from the define. install() returned early with no signal, so an integration went on installing its watchers and perf hooks around a tapir that registered nothing; it can now check the same flag install() does. (lib/src/tapir_plugin.dart)
  • ext.tapir.fill carries the type step's checks block. It copied only effect, so a fill whose receives-events check was indeterminate or skipped read as a clean pass; checks is now present on exactly the responses where the type step stamped it. (lib/src/extensions/ext_fill.dart)
  • tapir:doctor asks the running app. The TAPIR_DISABLE, enricher and semantics rows read the CLI process (its own define, a count hardcoded to 0, a semantics state hardcoded to on), so the one ERROR row could never fail. With a session they now call ext.tapir.boot_id over the recorded VM Service URI: no answer for ext.tapir.* warns that the app was built with TAPIR_DISABLE or never calls TapirPlugin.install() and fails the semantics row, a tree that is off fails it, and the enricher row prints the app's count. With no session, or an app that does not answer, each row says skipped. An app that registers no ext.tapir.* is asked again once, a second later, since a hot restart drops every extension until tapir registers them again and ext.tapir.boot_id registers last; only the second answer fails the row. ext.tapir.boot_id answers two more keys for it, semanticsEnabled and enrichers. Breaking: the doctor can now exit 1 where its error row could not fail before, so a CI step that runs it against an app without tapir installed starts failing. (lib/src/commands/tapir_doctor_command.dart, lib/src/extensions/ext_boot.dart)
  • tapir:install and tapir:doctor name what fluttersdk_dusk left behind. An old bin/dispatcher.dart made the installer skip dowel install in silence. Both now list a bin/dispatcher.dart that imports fluttersdk_artisan, bin/fsa, lib/app/commands/_index.g.dart, app commands still written against the Artisan* classes and a fluttersdk server in .mcp.json, then the commands to run (dart run dowel install, dart run dowel commands:refresh, dart run tapir tapir:install); the doctor reports them as a warning. (lib/src/commands/tapir_install_command.dart, lib/src/commands/tapir_doctor_command.dart)
  • Two error types, refused and invalid_param. A refusal (a ref that is not a text field, a container holding several checkboxes, a perf call with no session open) answered type: unexpected, and most bad input answered the generic extension error. refused answers JSON-RPC -32001; invalid_param and missing_param answer the standard invalid-params -32602; every other type keeps -32000. TapirErrorEnvelope.code is the one place the mapping lives. (lib/src/utils/error_envelope.dart)

Changed

  • Breaking: tapir requires dowel ^1.0.0. The ^0.0.21 range cannot resolve next to dowel 1.0.0 or the Magic packages that depend on it. dowel 1.0.0 trims its barrel and renames the install results, so the hide Error imports are gone. (pubspec.yaml)
  • Breaking: set_checkbox runs the actionability gate. It tapped wherever the control was, so an offscreen or covered checkbox reported verified: false with no reason. It now gates the control it taps like the other tapping verbs: one below the fold is scrolled into view, and one that is disabled, covered or moving is refused with the gate's reason (obscured by other widget, ...) and its envelope type. It takes checkStable and checkReceivesEvents like them, tapir:set_checkbox included as --[no-]checkStable and --[no-]checkReceivesEvents, and stamps checks when the gate could not confirm. A script no longer has to scroll the control into view first, and handles a refusal where it used to read verified: false. (lib/src/extensions/ext_checkbox.dart, lib/src/tapir_dowel_provider.dart, lib/src/commands/tapir_set_checkbox_command.dart)
  • Breaking: set_checkbox reports the settled value. value echoed the request, so a control whose owner ignored the tap answered value: true, effect.after: false, verified: false. It is now the state read back after the tap, equal to effect.after. (lib/src/extensions/ext_checkbox.dart)
  • Breaking: refusals and bad input change their error type and code. The sites that refuse a target now answer refused (-32001): the ambiguous or stateless set_checkbox ref, the text verbs on a ref that is not a text field, fill and clear with no field under the ref, focus with nothing focusable, perf_end and perf_insight and perf_trace without the session they need, and semantics_hold release outside a session. The sites that reject a value now answer invalid_param (-32602): perf_begin mode, scroll delta, screenshot rect, semantics_hold action and set_checkbox value. Every missing_param answers -32602 instead of -32000. A caller that branched on unexpected or on -32000 for these reads the new type. (lib/src/extensions/)
  • Breaking: the console and exception readers take a request object. recentLogsReader was ({int limit, String? minLevel}) and recentExceptionsReader ({int limit}), so a filter added later would have broken every host that assigns an exact-shape closure. Each is now List<Map<String, Object?>> Function(ConsoleRequest) and Function(ExceptionsRequest), with the same fields and defaults; both request classes are exported. A host writes recentLogsReader = (ConsoleRequest request) => ... and reads request.limit. (lib/src/extensions/ext_console.dart, lib/src/extensions/ext_exceptions.dart, lib/tapir.dart)
  • Breaking: the barrel exports named symbols only. export 'src/tapir_plugin.dart' and export 'src/ref_registry.dart' exported everything, including the semantics handle, the boot id minting, the install counter and the ref registry's test helpers. Each export now names its symbols, and those internals moved out of the classes to src/: TapirPlugin.releaseSemantics, acquireSemantics, semanticsReleased, mintBootId, installCount and resetSemanticsForTesting, and RefRegistry.registerForTesting, disposeGroup and activeRefs, are no longer public. TapirPlugin.aiTestDisableEnvValue, mutable and only used by tests, is gone; read TapirPlugin.isDisabled. (lib/tapir.dart, lib/src/tapir_plugin.dart, lib/src/ref_registry.dart)
  • Breaking: four CLI commands take their MCP tool's name, and dblclick is double_click. tapir:wait is tapir:wait_for, tapir:modal is tapir:dismiss_modals, tapir:resize is tapir:resize_viewport and tapir:device is tapir:device_profile (their substrate routes follow: dowel:tapir:resize_viewport, dowel:tapir:device_profile). Beside right_click and triple_click, ext.tapir.dblclick, tapir:dblclick and tapir_dblclick are ext.tapir.double_click, tapir:double_click and tapir_double_click. No alias is kept: rename them in scripts, allow lists and tapir:run steps. (lib/src/commands/, lib/src/extensions/ext_pointer.dart, lib/src/tapir_dowel_provider.dart)

Removed

  • Breaking: ext.tapir.find_by_label. No command, MCP tool, doc or skill called it, and ext.tapir.find covers it; ext.tapir.find_by_text stays for the perf runner. (lib/src/extensions/ext_wait_find.dart)

Fixed

  • tapir:install resolves a magic_devtools path dependency outside the project. The walk to .dart_tool/package_config.json went through a file path that kept a trailing separator, so the ../../magic_devtools pub writes for path: ../magic_devtools resolved one level too deep and the Magic wiring was skipped with a "no lib/tapir.dart" warning. The walk stays on directory URIs, and every rootUri resolves through a file path, so a percent-encoded or absolute file: URI works too. (lib/src/commands/tapir_install_command.dart)
  • set_checkbox taps a bare switch, not its labelled container. A Switch or Checkbox is no semantics boundary, so inside Semantics(container: true, label: ...) its flag lands on the container's node and the tap went to the container's centre, which can sit beside the control. The tap now goes to the render object that set the flag. (lib/src/extensions/ext_checkbox.dart)
  • tapir:hot_reload_and_snap recognises every reload line. It matched in \d+ms, which misses the web runner's Reloaded application in ... and any reload of a second or more, printed as 1,234ms; such a reload waited 10 s and exited 1 although it worked. (lib/src/commands/tapir_hot_reload_and_snap_command.dart)
  • set_checkbox refuses a value that is not a boolean. Anything but false or 0 read as true, so "flase" ticked the box; only true, false, 1 and 0 are accepted, and the rest is refused as invalid_param. (lib/src/extensions/ext_checkbox.dart)
  • A target half off screen with no Scrollable is tapped on its visible part. The gate passes a rect that still overlaps the viewport, and the pointer then went to its centre outside the view and reached nothing. The dispatch point, and the gate's hit probe, are the centre of the rect clipped to the visible viewport. (lib/src/utils/actionability_gate.dart)
  • The FakeCdpServer lifecycle test no longer flakes. It asserted that start() and stop() each took under 100 ms, which failed under a loaded full run; it now bounds both at 10 s to catch a hang and keeps the port re-bind check. Test-only. (test/src/cdp/fake_cdp_server_test.dart)
  • tapir:doctor finishes when a file the leftovers scan reads cannot be read. An unreadable .mcp.json threw out of the scan and aborted the whole run. The row now warns could not inspect <path>: ... instead, and never reads none found after a scan that stopped. (`lib/...
Read more

0.0.24

Choose a tag to compare

@anilcancakir anilcancakir released this 14 Nov 02:36
6e6060a

Changed

  • Breaking, nothing is migrated automatically. In a consumer project: (1) swap the dependencies with flutter pub remove fluttersdk_dusk fluttersdk_artisan && flutter pub add tapir dowel and rename the imports and the Dusk* identifiers; (2) follow the dowel 0.0.21 migration (move .artisan/ to .dowel/, delete .dowel/plugins.json), delete bin/dispatcher.dart, bin/fsa and lib/app/_plugins.g.dart (the installer keeps an existing bin/dispatcher.dart, and the old one imports fluttersdk_artisan, so bin/ no longer analyses) and run dart run tapir tapir:install again; (3) rename dusk_* to tapir_* in any saved MCP allow list, prompt or script (a Claude Code allow-list entry mcp__fluttersdk__dusk_* becomes mcp__dowel__tapir_*, since the server key is now dowel); (4) delete the stale fluttersdk server entry from .mcp.json (dowel leaves entries an earlier release wrote alone) and from every client you registered by hand (claude mcp remove fluttersdk, the VS Code or Claude Desktop settings), since they run ./bin/fsa mcp:serve, which step (2) deletes; then run dart run tapir mcp:install and register dowel with the per-client commands in the README; (5) in a Magic app, remove the import 'package:magic_devtools/dusk.dart' line and the MagicDuskIntegration.install() call that the old installer added to lib/main.dart (MagicDuskIntegration is not a Dusk* name that step (1) renames), and check that lib/main.dart calls MagicTapirIntegration.install() after tapir:install has run (the installer adds that call once the installed magic_devtools ships tapir.dart; see the Magic bullet below), without adding a second one.
  • Renamed from fluttersdk_dusk to tapir. The package, the repository (anilcancakir/tapir), the barrels (package:tapir/tapir.dart and package:tapir/cli.dart), every Dusk* class (now Tapir*, and DuskArtisanProvider is TapirDowelProvider), the executable and the file names carry the new name. Also renamed: the ext.dusk.* VM Service extensions (now ext.tapir.*), the MCP tools (dusk_* is tapir_*), the CLI commands (dusk:* is tapir:*, and the substrate commands artisan:dusk:* are dowel:tapir:*), the environment variables (DUSK_* is TAPIR_*), the perf zone key (#fluttersdk_interaction is #tapir_interaction) and the docs URLs (https://anilcancakir.com/tapir).
  • Built on dowel and wind_kit_diagnostics_contracts. fluttersdk_artisan is now dowel ^0.0.21 and fluttersdk_wind_diagnostics_contracts is wind_kit_diagnostics_contracts ^1.3.0. The .mcp.json server key written by mcp:install is dowel.
  • The Magic step of tapir:install needs a magic_devtools built on tapir. When the consumer's pubspec names magic_devtools and the resolved release ships package:magic_devtools/tapir.dart, the installer wires MagicTapirIntegration from it. The magic_devtools releases that still ship MagicDuskIntegration have no such library, so with one of them the installer skips the Magic lines and warns; upgrade magic_devtools to a release built on tapir and rerun dart run tapir tapir:install (see Fixed).
  • The floor is Dart 3.13 and Flutter 3.47. Flutter 3.50 deprecates package:flutter/material.dart in favour of material_ui, which needs that floor; the tests and the example app now build on material_ui, the package itself imports neither, and the repository is formatted with the tall style that floor enables. A consumer on an older SDK stays on the last fluttersdk_dusk release.

Fixed

  • tapir:set_checkbox reads the checked state from Semantics. It used to test the widget type (is Checkbox, is Switch from flutter/material), so a control from material_ui, wind_kit or any hand-built Semantics(checked: ...) read as "unknown, assume unchecked", and setting a checked box to checked toggled it off. The state now comes from the checked or toggled flag of the ref's Semantics node, which every library sets. A ref without a checked or toggled state is refused instead of assumed unchecked, and a mixed checkbox is always tapped. (lib/src/extensions/ext_checkbox.dart)
  • set_checkbox taps the control it read and refuses an ambiguous container. When the state came from a descendant node (a labelled container that does not merge), the tap still went to the centre of the ref's own rect, which could be a sibling control or nothing: a container holding three switches read the first and flipped the second. The tap now goes to the centre of the descendant whose state was read, and a ref whose own node has no checked or toggled flag while several descendants do is refused with an error naming the ambiguity. A ref whose own node carries a flag (a merging tile included) is read as that control, as before. (lib/src/extensions/ext_checkbox.dart)
  • tapir:set_checkbox exits 1 when the effect is not verified. It printed Checkbox e5 set to true and exited 0 without reading effect.verified, so a tap that missed looked like a success. It now prints the value read back after the tap and exits 1, notes when the box started mixed; --json and --includeSnapshot still print the whole response and exit 0. The --includeSnapshot help no longer promises a snapshot the response never carries. (lib/src/commands/tapir_set_checkbox_command.dart)
  • tapir:install wires Magic only when the installed magic_devtools ships tapir.dart. The pubspec entry alone decided it, so a magic_devtools release that predates tapir got an import of package:magic_devtools/tapir.dart and a lib/main.dart that did not compile. The installer now resolves the package through the nearest .dart_tool/package_config.json at or above the project (a pub workspace member uses the one at the workspace root; an unreadable or invalid one counts as unresolved) and, without lib/tapir.dart, skips the Magic lines and warns to upgrade and rerun. tapir:doctor follows the same check, so it no longer advises wiring a class the app cannot import. (lib/src/commands/tapir_install_command.dart, lib/src/commands/tapir_doctor_command.dart)

Removed

  • The FluttersdkDuskArtisanProvider alias in package:tapir/cli.dart. The class dowel's plugins:refresh generates for this package, TapirDowelProvider, is the provider itself, so the alias is gone.

0.0.23

Choose a tag to compare

@github-actions github-actions released this 07 Nov 18:02
f5fd128

Added

  • Dusk wakes a hidden or minimised app instead of driving a frozen one. A hidden, minimised or occluded desktop window moves the lifecycle to hidden and turns frame production off, so a tap reported "no observable change", a snapshot went stale and a screenshot returned the last cached frame. Every ext.dusk.* verb now delivers AppLifecycleState.inactive and then, once a frame has run, resumed through the flutter/lifecycle channel, and schedules a forced frame when frames are off and the lifecycle is hidden or paused (never detached); a response that forced carries framesForced: true. It is re-applied on every call because the OS flips the lifecycle back to hidden on its next occlusion event. The wake counts only when a frame actually runs after it (otherwise the previous lifecycle is delivered again, unless the engine sent one of its own meanwhile), and it is never attempted in the app on web, where Chrome withholds animation frames from a hidden tab; the CLI sends Page.bringToFront there instead (see the web recovery entry below). Verified on macOS with the window hidden through osascript (dusk:tap, dusk:snap, dusk:screenshot and dusk:navigate all work) and on web with a second tab in front (warnings returned, nothing claimed as forced). The perf verbs, boot_id, close_app, console, exceptions, semantics_hold and evaluate observe the engine as it is and do not force. The app's lifecycle observers see resumed while this happens; --dart-define=DUSK_FORCE_FRAMES=false (kDuskForceFrames) turns it off. (lib/src/utils/frame_sync.dart, lib/src/utils/dusk_register.dart, lib/src/utils/dusk_response.dart, lib/src/extensions/, doc/reference/frame-production.md)
  • dusk:snap and dusk:observe say when a control is off screen or covered. A ref line may end with [offscreen] (the widget lies outside the viewport) and/or [occluded] (a hit test at its centre reaches something painted above it), after [ref=eN]; a healthy screen prints neither. Each ext.dusk.observe candidate gains inViewport and occluded; isVisible is unchanged. (lib/src/utils/viewport_visibility.dart, lib/src/extensions/ext_snapshot.dart, lib/src/extensions/ext_observe.dart)
  • dusk:snap --grep reads an invalid regex as text. A label such as Container Sample (MP4, H.264 is an unterminated group and used to crash the snapshot; it now matches as a literal substring and the payload carries grepLiteral: true. A valid regex keeps its regex meaning. (lib/src/extensions/ext_snapshot.dart)
  • An e<N> ref survives a rebuild. Snapshots bind a label-and-index query and the match count to every labelled ref (RefBinding, with RefRegistry.bind, lookupBinding and rebind, additive statics). When the element behind a ref has left the tree, the action relocates it only if the live tree has the same number of matches as at snapshot time, and the response lists the ref under reresolved. With a different count the old defunct (element no longer mounted) error is unchanged. A ref is also pinned to the route it was minted on (RefBinding.route, held weakly): a same-label control on another route is never a match, so a ref taken on a page you have since left fails as defunct instead of acting on the next page. Known limit: the query has no role, so a node of another role with the same label in the same position on the same route could match. (lib/src/ref_registry.dart, lib/src/extensions/ext_pointer.dart, lib/src/extensions/ext_snapshot.dart)
  • dusk:screenshot infers the format from --output. .png, .jpg and .jpeg pick the format when --format is not given, so a .png file no longer holds JPEG bytes. An explicit --format wins and a mismatch with the extension warns on stderr. (lib/src/commands/dusk_screenshot_command.dart)
  • dusk:run / dusk_run: several verbs in one call. dusk:run "tap --ref=e5" "fill --ref=q1 --text=\"a b\"" "snap" takes each step as a shell-tokenised verb line (a bare verb gets the dusk: prefix), parses it with that verb's own options and runs it on one shared VM connection. The output is one line per step ([1/3] tap --ref=e8 ok: Tapped e8, the first output line cut to 120 characters), then the last step's output. It stops at the first failing step with step N failed (exit C): <step> and that exit code; --continue runs every step and exits with the first non-zero code. A nested dusk:run, an unknown verb and a verb that does not boot connected (dusk:resize, dusk:device, dusk:install, dusk:doctor, dusk:perf_campaign, dusk:perf_compare) fail as their step, as does any step that throws (it exits 3); a step that ends the app (dusk:close_app, dusk:perf_run) belongs last. The MCP tool takes {steps: string[], continue?: boolean} and routes as artisan:dusk:run, the seventh substrate-routed tool. Between steps use --text, --key or q-refs, since e-refs belong to one snapshot. 42 commands and 40 tools now. (lib/src/commands/dusk_run_command.dart, lib/src/dusk_artisan_provider.dart, doc/commands/dusk-run.md)
  • --snapshot=full|diff|none on dusk:tap, fill, type, navigate and scroll, and a snapshot enum on dusk_tap, dusk_fill and dusk_navigate. It overrides includeSnapshot. diff returns only the lines that changed against the last unfiltered snapshot (a dusk:snap or an earlier action) in unified-diff form: a changed row is - <old line> then + <new line>, an added row + , a removed row - , then a (K of N lines changed) footer, and the payload adds snapshotMode: "diff". The mode is read case-insensitively, and an unknown one comes back as snapshotError in the payload rather than as a silently missing snapshot. With no baseline yet the full text comes back. A hot restart clears the baseline, and a filtered snap (grep, within, depth, interactiveOnly) or an enricher walk never records it. (lib/src/utils/snapshot_diff.dart, lib/src/extensions/ext_snapshot.dart, the action verbs and commands)
  • dusk:snap --json prints the whole envelope. The tree alone dropped grepLiteral, warnings and framesForced; text mode is unchanged. (lib/src/commands/dusk_snap_command.dart)
  • dusk:press_key --modifiers is applied. control (or ctrl), shift, alt and meta, in any case, as a comma list on the CLI and an array over MCP, are held down before the key and released after it, even when delivery throws, so exact-match shortcuts such as Ctrl+S and Shift+Tab fire. An unknown name is an error before anything is pressed. (lib/src/extensions/ext_text_input.dart, lib/src/utils/key_press.dart, lib/src/commands/dusk_press_key_command.dart)
  • A tap on a merge boundary goes to the control the ref names (#60). A MergeSemantics boundary taps at its own centre, which can belong to another handler (a full-frame stage that pauses on any tap) while the gate still says confirmed. When exactly one labelled control owns the merged label (a Semantics with a label that is part of it and an onTap or button) and a hit test at its centre reaches it, dusk:tap sends the pointer there and the payload adds retargeted {from, to, ownerRect} (with centreHandled: true when the centre has a tap handler of its own, so the stage or row it would have tapped is named); the summary reads Tapped e7 on its control at (x, y), not the merged centre (x, y). With several candidates, or an owner that is covered or off screen, the tap stays at the centre and the payload adds tapPointOutsideOwner {ownerRect, point} with one stderr line. retarget: false (CLI --no-retarget) taps the centre as before. Snapshot and observe say so before any tap: a tapOwner: {"x","y","w","h"} sub-line under the ref and a tapOwner candidate field, both present only when the centre misses the control, with centreHandled: true added when the centre has its own handler and retargets: false when a tap cannot go to the control. The gate and its reasons are unchanged; hover, dblclick, right_click, triple_click and drag still use the centre. (lib/src/utils/tap_owner.dart, lib/src/extensions/ext_pointer.dart, lib/src/extensions/ext_snapshot.dart, lib/src/extensions/ext_observe.dart, lib/src/commands/dusk_tap_command.dart)
  • effect.target says which field a text verb wrote. dusk:fill, dusk:type and dusk:clear add target {matchKind: overlap|nearest|element, label, rect} to the text effect, and the CLI prints a warning when the match was nearest. (lib/src/utils/effect_report.dart, lib/src/extensions/ext_text_input.dart, lib/src/commands/dusk_fill_command.dart, lib/src/commands/dusk_type_command.dart)
  • The CLI recovers a hidden web tab. snap, observe, tap, fill, type, navigate, scroll and press_key call the extension through callExtensionVisible. When the answer carries a web warnings block, or the call times out (the VM Service gives up on a hidden tab that never answers) and the tab reports hidden, and the session has a cdpPort, the CLI sends CDP Page.bringToFront once (it raises the Chrome window). Read-only verbs (snap, observe) are then called again once; a verb that acts is not repeated, since its gesture may already have fired, and a stderr notice says to run dusk:snap. A CDP failure leaves the original result and exit code; without a cdpPort nothing changes. (lib/src/cdp/page_visibility.dart, lib/src/commands/web_visibility.dart)

Changed

  • Every MCP tool schema is closed, and lists the knobs its handler reads. artisan validates each tools/call against the schema, so a key the schema did not list was either let through to a handler that ignored it (a typo answered with a clean response) or, when the schema was incomplete, impossible to pass. All 40 schemas now declare additionalProperties: false, and the keys the handlers read but the schemas o...
Read more

0.0.22

Choose a tag to compare

@github-actions github-actions released this 18 Oct 11:43
a6c2042

Fixed

  • dusk:observe lists a merged control once. The walk emitted every node that carries an action, and MergeSemantics keeps each absorbed descendant in the tree as its own node flagged isMergedIntoParent, so an MSSwitch row came back as a 324x44 boundary plus a 44x24 duplicate and an MSButton "Kaydet" as 79x44 plus 79x36, while the platform sees one node. A merged node now yields no candidate and no qN handle; the walk still descends through it, the candidate shape is unchanged, and the boundary is still listed. The candidate count can drop for a screen with merged controls, and a script that held a candidate by position has to re-observe; a dusk:perf_run {role, name, index} target with an index above 0 now counts the same nodes dusk:snap lists. The per-label counter still counts merged nodes, because the find resolver does, so the matchIndex of every surviving handle is unchanged. (lib/src/extensions/ext_observe.dart)

0.0.21

Choose a tag to compare

@github-actions github-actions released this 18 Oct 09:28
e2ac478

Fixed

  • dusk:snap lists a merged control once. The walk printed every semantics node that carries an action, and MergeSemantics keeps each absorbed descendant in the tree as its own node flagged isMergedIntoParent, so a magic_starter MSSwitch row showed up as a button "X" [ref=e1] followed by a second button "X" [ref=e2] while the platform sees one node. A merged node is no longer printed and gets no ref; the walk still descends through it, and the line shape, the ref syntax and the textbox collapse are unchanged. Refs after a removed duplicate shift down by one each, so a script that hard-coded ref numbers has to re-snap. A WSelect that still prints twice is the wind side (two real platform nodes), fixed in fluttersdk_wind 1.9.3. (lib/src/extensions/ext_snapshot.dart)

0.0.20

Choose a tag to compare

@github-actions github-actions released this 12 Oct 23:06
de72030

Changed

  • The artisan floor names this batch's release. fluttersdk_artisan moves ^0.0.19 to ^0.0.20. The old range already admitted 0.0.20, so a fresh pub get resolves nothing differently; what changes is that the floor names the release this package is verified against. artisan 0.0.20 is not breaking, and a consumer inherits one behaviour from it: a dusk:* command run while the app is still booting, before ext.dusk.* is registered, now prints one retry line and exits 75 instead of an RPC stack and exit 3. The requirements tables in doc/getting-started/installation.md and doc/getting-started/index.md and the stack lines in CLAUDE.md and .github/copilot-instructions.md quote the new floor. (pubspec.yaml, doc/getting-started/, CLAUDE.md, .github/copilot-instructions.md)

0.0.19

Choose a tag to compare

@github-actions github-actions released this 09 Oct 02:04
f5d23ba

Changed

  • The artisan floor names this batch's release. fluttersdk_artisan moves ^0.0.17 to ^0.0.19. The old range already admitted 0.0.19, so a fresh pub get resolves nothing differently; what changes is that the floor names the release this package is verified against. Neither artisan 0.0.18 nor 0.0.19 is breaking: 0.0.18 adds installer operations dusk does not use, and 0.0.19 widens artisan's xml constraint to admit 7.x, which a consumer now inherits. The requirements tables in doc/getting-started/installation.md and doc/getting-started/index.md and the stack lines in CLAUDE.md and .github/copilot-instructions.md quote the new floor. (pubspec.yaml, doc/getting-started/, CLAUDE.md, .github/copilot-instructions.md)
  • The example app moves cupertino_icons to ^2.0.0. It touches only the unpublished example/ app, not the package. (#53, example/pubspec.yaml, example/pubspec.lock)

0.0.18

Choose a tag to compare

@github-actions github-actions released this 07 Oct 10:29
59911aa

Fixed

  • dusk:press_key / dusk_press_key reaches the focused widget. The key was handed to HardwareKeyboard.instance.handleKeyEvent, which updates the pressed-key state and calls the keyboard's global handlers and nothing else, so Focus.onKeyEvent, Shortcuts and CallbackShortcuts never heard it, whatever the tool description promised: three ArrowDown presses on an app whose focused FocusScope zaps channels on the arrows changed nothing. Both halves of the press now enter through the binding's onKeyData, the handler real key data reaches, marked synthesized so each is dispatched at once, which delivers to HardwareKeyboard and to the focus tree alike. Every key also carries its own physical key now, where every press used to report the Enter key's. dusk:reset_overlays pressed its Escape the same broken way and now goes through the same helper, so a Shortcuts-bound dismiss hears it. Two limits are documented on the helper: a real key whose raw message is still pending holds the press back until the next real key, and on an embedder that sends only raw key messages (none of Flutter's own) a press before the first real key fixes the transit mode to key data and real keys stop reaching the focus tree until a restart, so such an embedder must not be driven with it; reset_overlays stays best-effort there, catching the debug assert as well as a missing handler. (lib/src/utils/key_press.dart, lib/src/extensions/ext_text_input.dart, lib/src/extensions/ext_modal_router.dart)

Added

  • dusk:press_key --key=<letter or digit>. A single ASCII letter or digit presses the key that carries it (G and g both press the G key with the character g), so a shortcut bound to a letter can be driven; text still goes through dusk:type, since a text field takes its text from the text input channel. Anything else outside the named keys is still refused. (lib/src/extensions/ext_text_input.dart, lib/src/commands/dusk_press_key_command.dart, lib/src/dusk_artisan_provider.dart, doc/mcp/tool-reference.md)