tom_d4rt_dcli 1.5.0

tom_d4rt_dcli: ^1.5.0 copied to clipboard

Changed — realigned with the dcli 10 line #

dcli, dcli_core and dcli_terminal move from ^8.4.2 to ^10.0.0, and dart_console to ^5.0.0 (required by dcli_terminal 10). win32 resolves 6.4.0 in consequence.

The move was blocked, not deferred: tom_build_base pinned dcli: ^8.4.2, and because it sits transitively under this package via tom_d4rt_generator, version solving rejected any request for dcli 10 regardless of what this pubspec said. tom_build_base 2.11.0 widens that constraint to admit both majors, which is what unblocks this.

Changed — bridges regenerated against the dcli 10 API #

lib/src/bridges/*.b.dart are regenerated. dcli 9 removed a set of members that had been deprecated through the 8.x line, so the committed bridges described methods the package no longer has and every test that loaded them failed to compile. The removals visible here are Env.addToPATHIfAbsent, Terminal.previousLine, Terminal.lines, DartProject.current, DartScript.current, and DartSdk.globalActivate / globalActivateFromPath.

Removed — the FileSync.tempFile() bridge test #

FileSync.tempFile() was deprecated in favour of the top-level createTempFilename() and removed in dcli 9, so the test asserted an API that no longer exists. createTempFilename() is already covered by two tests in the Temp Files group; a third copy in the FileSync group would have been duplication rather than coverage.

Suite 704/13/0, against a 705/13/0 baseline — the difference is exactly the deleted test.

Fixed — the --version banner reports the real version (sce9) #

lib/src/version.versioner.dart said 1.3.1 while the pubspec said 1.4.0, so dcli --version named a release that is not what was running. The stamp is regenerated and version_stamp_test passes; published so the correction reaches the package people install rather than only this tree.

Changed — the bridges are regenerated after seven months, and the package can now check them (sce3) #

tom_d4rt_dcli configured d4rtgen in its buildkit.yaml but listed tom_d4rt_generator in no dependency block, so dart run tom_d4rt_generator:d4rtgen failed in it and checkBridgeFreshness could not be called from its suite. It was one of two consumers in the workspace that could neither regenerate its bridges nor check what it had committed.

The dependency was there once, commented out, with a reason:

# analyzer-10 migration: tom_d4rt_generator (still analyzer ^8.4.1) cannot
# share a resolution with tom_d4rt's analyzer ^10.

That has not been true for some time. The generator depends on analyzer: ^10.0.0 and on tom_d4rt: ^1.77.0, which is exactly what this package already declared; adding it back as a DEV dependency resolves on the first attempt and changes nothing the package ships.

With it in place the measurement was possible, and eight of the nine generated files were stale — every bridge module, the dartscript, the test runner and the relaxers, last written in February. They are now what the generator produces, and test/bridges_fresh_test.dart will say so the next time they drift.

Changed — formatted the tree once (scd82) #

Every package in this repo already declares an SDK floor above the 3.7 tall-style boundary, so the formatter can no longer produce two layouts here. What was not true is that the trees were formatted: until this commit, running dart format on any single file rewrote it wholesale and buried whatever real edit came with it. 46 of 60 files under lib and test were affected.

This package and tom_dcli_exec are twins in the sense SCC26 cares about — the same REPL on the analyzer line and the analyzer-free line, kept in step by diffing — so unformatted layout was actively costing the check that keeps them equal, not merely untidy.

The commit is layout plus twelve brace pairs, and which is which was proven rather than asserted. git diff -w cannot establish inertness, because the tall style splits lines and a whitespace-insensitive diff still counts a moved boundary as a change. What was checked is the token stream, per file, twice: whitespace stripped, then whitespace and commas stripped. 44 of the 46 files are identical to HEAD under that normalisation.

The other two are lib/src/cli/repl_base.dart and lib/src/cli/vscode_integration.dart, which gained braces — ten pairs and two pairs. Re-wrapping a long braceless if splits it across lines, which is what makes curly_braces_in_flow_control_structures fire, and it fired twelve times here. The braces were added by dart fix --code=curly_braces_in_flow_control_structures rather than by hand: a first attempt to brace them with a regex over the source corrupted a multi-line string, which is the argument for using the parser. Every inserted chunk in those two files was then checked to be exactly { or } and nothing else.

dart analyze is clean, and the formatter is now idempotent here.

Added — the full VS Code scripting API bridge surface #

lib/src/bridges/tom_vscode_scripting_api_bridges.b.dart was regenerated against the current tom_vscode_scripting_api and grew from a partial surface to the complete one (~9 000 lines of additional bridged members). Scripts that previously hit Undefined member on a VS Code scripting call now resolve it.

The remaining bridge files (cli_api, dcli, path, tom_chattools) were regenerated in the same pass; their content is unchanged.

Fixed — barrel re-exports were registered under an unmatched key (GEN-125) #

bridgeReExports() recorded its source as lib/tom_d4rt_cli_api.dart instead of package:tom_d4rt_dcli/tom_d4rt_cli_api.dart, because the bridge generator failed to root a relative source path before mapping it to a package URI. registerLibraryReExport matches that key against a script's import, which is always a package: URI — so every symbol reached only through a re-export of the CLI API barrel was invisible to interpreted scripts.

Fixed upstream in tom_d4rt_generator 1.15.3 and picked up here by regeneration; no hand edits to any *.b.dart.

Fixed — .start-execute blocks of bare statements failed to parse (tccc5) #

D4rt.execute(source:) parses its argument as a compilation unit and calls the top-level main. A .start-execute block was handed straight to it, so a block of bare statements

.start-execute
var testVar = 100;
verify(testVar == 100, '.start-execute variable works');
.end

parsed as a top-level variable followed by a function declaration with no body and failed with Expected to find ')' / A function body must be provided.

The three-case rule that fixes this already existed — but only on the stdin path. It is now the top-level prepareProgramSource, applied at every entry point that means "run this as a fresh program": stdin, the exec command, .start-execute, and CliController.execute.

• imports + main → run unchanged
• main without imports → the bridge import block is prefixed
• bare statements → prefixed and wrapped in a generated main

The generated wrapper is async when the statements await, matching what .start-script already did; await inside a string or comment does not count. .start-execute also awaits a Future result now instead of printing it.

Requires tom_d4rt >= 1.29.0, without which a run whose assertions all pass is still failed by phantom extension-registration errors. Routing .start-execute through the bridge import block — which it needs, because a bridged library's top-level functions (verify, verifyEquals, …) only become visible through its import — is what made those errors reachable from this path.

• Fix duplicate-export collisions that broke AOT compilation in hosted consumers (e.g. tom_core_d4rt's d4rt binary). The VS Code integration re-export now hides PermissionResult, ConversationExchange, ProjectInfo and WorkspaceInfo so the package's own bot_mode types win, and telegram_bot_server hides the VS Code ConversationExchange so the local conversation_trail type is used. No public API change for the package's own types; the VS Code variants of those four names are no longer re-exported (import tom_vscode_scripting_api directly if you need them).

• Housekeeping: test artifacts now live in a gitignored testlog/ folder; doc/ no longer ships machine-generated baselines or last_testrun.json. No code changes.

Maintenance #

• Regenerated dcli bridges against the current tom_d4rt_generator 1.9.0 (summary-backed extraction, GEN-095 and follow-up fixes).
• Pinned dependency constraints to current releases (tom_d4rt ^1.8.20, tom_vscode_scripting_api ^1.0.1, tom_chattools ^1.0.2).

Maintenance #

• Renamed version.g.dart → version.versioner.dart.
• Updated barrel import in tom_d4rt_dcli.dart.

Bug Fixes #

• GEN-070 follow-up: Find class now properly bridged via generator fix (multi-chain barrel re-export)
• Removed dcli_missing_bridges.dart supplementary bridge (no longer needed)
• Removed lastModified/setLastModifed tests (not exported from dcli barrel)
• Replaced deprecated symlink() tests with createSymLink() tests

Tests #

• All 389 tests pass, 0 failures, 0 skips

Bug Fixes #

• DCLI-GEN-001: Added supplementary bridge for missing global functions (lastModified, setLastModifed, symlink)
• DCLI-GEN-002: Added Find class bridge with static getters (file, directory, link)
• DCLI-VSCODE-001: Fixed VS Code bridge import path and test constructor arguments
• DCLI-LOCK-001: Updated tests for deprecated NamedLock.withLock (dcli 8.4.2), added withLockAsync tests
• DCLI-API-001: Fixed expandDefine test prefix ($ → @)
• Symlink bridge uses createSymLink internally (avoids deprecated symlink() warning)

Tests #

• All 391 tests pass, 0 failures, 0 skips

• Full DCli scripting support now
• Updated tom_d4rt dependency to ^1.8.1
• Regenerated bridges with latest generator (multi-barrel registration, extension filtering)

• Initial version.