Files
reverse-engineering-anything/src/cli/desktopCaptureCommands.ts
T
fe2985a0b4 feat(desktop): check apps whose files REA can't read, through the app's own headless mode or a screen judge (#9217)
* fix(desktop): preserve document URLs and value formats

* fix(desktop): bound added checks and omit noisy text guards

* fix(desktop): exclude unseeded locks from task checks

* fix(desktop): preserve draft bounds and launch URLs

* fix(desktop): render draft values using their source format

* fix(desktop): preserve launch URLs and JSON draft values

* fix(desktop): recognize captured file URL documents

* fix(desktop): render captured file URLs for runtime home

* fix(desktop): distinguish phone numbers from save metadata

* fix(desktop): preserve launch URLs and user text

* fix(desktop): preserve captured URLs and user text

* fix(desktop): preserve local file URL authority

* fix(desktop): preserve local file URL authority

* fix(desktop): preserve local file URL authority

* fix(desktop): render captured file URLs at launch

* fix(desktop): classify numeric timestamps using field context

* fix(desktop): classify numeric timestamps using field context

* fix(desktop): classify numeric timestamps using field context

* fix(desktop): retain context for generated draft values

* fix(desktop): retain context for generated draft values

* fix(desktop): retain context for generated draft values

* test(desktop): cover generated IDs in text guards

* test(desktop): cover generated IDs in text guards

* test(desktop): cover generated IDs in text guards

* fix(desktop): isolate launcher URL encoding

* fix(desktop): isolate launcher URL encoding

* fix(desktop): isolate launcher URL encoding

* fix(desktop): isolate launcher URL encoding

* fix(desktop): retain clean text in noisy record fields

* fix(desktop): classify added line values in one pass

* fix(desktop): apply XML quote rules to text drafts

* fix(desktop): distinguish XML attributes from quoted text

* fix(desktop): preserve parsed XML line context

* Format appTaskChecks.ts with the pinned oxfmt

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix(desktop): retain partial XML syntax facts

* fix(desktop): recognize markdown autolinks in drafts

* fix(desktop): reuse export exclusions in task drafts

* fix(desktop): report unchecked client build paths

* docs(desktop): clarify gzip scrub limits

* fix(desktop): reuse export exclusions in task drafts

* docs(desktop): keep the judge command on one line

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* docs(desktop): clarify gzip scrub limits

* fix(setup): own desktop build process cleanup

* fix(doctor): compare desktop settings with resolved defaults

* style(desktop): format the noisy-record draft files with the pinned oxfmt

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* test(desktop): give the root-array draft test its own describe

The records describe grew past the 150-line lint limit with it.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix(doctor): inspect client-recorded desktop runners

* fix(doctor): reuse matching runner probes

* fix(setup): propagate cancellation to desktop actions

* fix(setup): stop Hopper on cancel, never report a cancelled setup ready, and keep a failed action's client setting

- The Hopper install gets setup's signal: Linux already honoured it, and the
  macOS installer now stops its downloads and stops before mounting or copying.
- runSetup checks the signal after the final doctor and close, so a
  cancellation there raises instead of returning ready.
- A failed pack or agentenv install no longer writes its archive or runner to
  client configs, so the setting the client already had stays.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix(setup): keep the completed setup result in a cancellation after cleanup

A cancellation during the final doctor or host close carried only the applied
actions, dropping each client's result and backup path and the doctor report.
It now carries the whole result as setup_result, as doctor's cancellation
carries its report.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* test(desktop): move launch execution to boundary suite

* fix(desktop): redact build schema failure output

* refactor(desktop): normalize schema failure once

* fix(desktop): omit unsafe schema parser snippets

* fix(desktop): preserve draft change and selected path evidence

* test(desktop): group retained dconf note regressions

* fix(desktop): respect archive entry format in draft checks

* test(desktop): separate unowned-file draft note cases

* fix(desktop): preserve explicitly selected journal documents

* fix(xml): avoid scanning selected ancestors per tag

* fix(desktop): exclude covered text runs from draft counts

* fix: preserve text when drafting desktop checks

* fix: restore nested desktop state aliases

* refactor: share UUID recognition across draft checks

* fix: retain multiline XML attribute context

* fix: retain relationship identifiers in task drafts

* docs: clarify desktop draft identifier handling

* test(desktop): type formatting run cases as pairs

* fix(desktop): carry document format into selector evaluation

* test(desktop): separate markup boundary regression group

* fix(desktop): propagate formats through draft selectors

* fix(desktop): switch selector format within archive entries

* test(desktop): remove duplicate seed filename case

* fix(setup): retain diagnostics when cancellation overlaps failure

* refactor(setup): keep command cancellation at its owner

* test(setup): verify cancellation at the application boundary

* test(desktop): pass JSON format in exported index parity

* docs(desktop): clarify task checks before scratch export

* fix(desktop): respell selected paths in file URLs

* test(desktop): group captured document spelling cases

* test(desktop): pin reader-free verifier to selector parent

* fix(desktop): bound draft preview serialization

* fix(desktop): restore selected state aliases

* fix(desktop): respect entry format in draft noise

* test(desktop): match exported alias evidence at its JSON boundary

* fix(desktop): retain app state logs in exported seeds

* fix: align desktop numeric seed comparisons

* test(desktop): keep one draft exclusion fixture import

* test(desktop): retain log eligibility across draft test groups

* fix(doctor): resolve empty registered desktop runner

* test(desktop): check retained log seed admission

* style: fix desktop launch test formatting

* style: format native command path schema

* fix(desktop): map selected external launch paths

* style(desktop): format seed document regressions

* test(desktop): group launch URL regressions

* fix(desktop): omit redundant nested state aliases

* fix(desktop): compare added free text as a whole

* fix(desktop): check recompressed seeds as shipped

* docs(desktop): clarify recovered seed format

* test(desktop): group shipped gzip seed checks

* fix: retain quoted identifier text in task drafts

* fix: recognize Unicode XML in desktop state readers

* fix(desktop): bound nested draft checks

* fix(desktop): explain missing arrays in truncated views

* fix(desktop): find XML type references in archives

* test(desktop): remove duplicated capture assertions

* fix(desktop): retain nested omitted text boundaries

* style(desktop): format draft value imports

* test(desktop): group model and draft value checks

* fix(desktop): map captured launch paths once

* refactor: remove unused whole-text guard state

* refactor: stop passing unused text children

* test: align free-text checks with whole-note policy

* fix: validate recompressed desktop archive seeds

* style(desktop): format archive seed regression

* fix(desktop): retain placed-item sibling guards

* fix(desktop): preserve archive paths and bound doctor reads

* fix(doctor): refuse non-files before descriptor admission

* fix(setup): track remaining cleanup resources across retries

* test(desktop): specify XML format in heap pairing fixture

* fix(desktop): compare screen task views by value

* fix(desktop): reject inexact reader numbers

* fix(desktop): budget export for both reader passes

* fix(desktop): bound reader runner argument size

* test(desktop): verify chunked reader payloads and parent output

* fix(doctor): query agentenv for model endpoint

* style(desktop): format reader payload assertion

* fix(doctor): preserve uncertainty for empty endpoint override

* fix(desktop): redact runtime schema errors

* fix(desktop): normalize captured file URLs

* fix(desktop): report seeded files by capture path

* fix(desktop): retain startup readiness failure stage

* fix(desktop): retain JSON date checks

* fix(desktop): preserve archive formatting checks

* fix(desktop): preserve explicitly selected journal files

* fix(desktop): report build command transport limits

* docs(desktop): clarify unreadable dconf versions

* test(desktop): remove duplicate history regression

* fix(desktop): align XML names with exported reader

* fix(desktop): create selected state alias targets

* fix(desktop): inspect supported runtime archives

* fix(desktop): retain record identity in draft checks

* fix(desktop): align timestamp selector digits

* fix(desktop): scrub XML independently of reader compatibility

* fix(desktop): preserve file aliases and reader diagnostics

* fix(desktop): retain remote reader cleanup ownership

* style(desktop): format capture review fixes

* fix(desktop): retain reader leases across verifier reloads

* test(desktop): overlap reader cleanup retries

* fix(desktop): run long exports within exec limits

---------

Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-09 23:09:52 +08:00

440 lines
16 KiB
TypeScript

import { createInterface } from "node:readline";
import { z } from "incur";
import type { Logger } from "pino";
import { captureDesktopApplication } from "../application/desktop/DesktopApplicationCapture.js";
import type { DesktopStateReaders } from "../application/desktop/DesktopApplicationState.js";
import { AnalysisError } from "../domain/analysisErrorBase.js";
import { analysisErrorWithCleanupFailure } from "../domain/analysisErrorCleanup.js";
import { ProviderCleanupError } from "../domain/providerCleanupError.js";
import { ArtifactReaderFailure } from "../artifacts/ArtifactReader.js";
import { bindArtifactFileCleanup } from "../artifacts/ArtifactFileOperation.js";
import type { Evidence } from "../domain/evidence.js";
import { localPathStringSchema } from "../domain/localPath.js";
import { err, ok } from "../domain/result.js";
import { finishAnalysisCommand } from "./finishAnalysisCommand.js";
import { projectCliAnalysisResult } from "./projectCliAnalysisResult.js";
import { absoluteCallerPath } from "../filesystem/callerPath.js";
import { cliDecimalField } from "./options.js";
import { CLI_COMMANDS } from "../cliCommandNames.js";
import { logCliCommand } from "../cliLogging.js";
import {
DESKTOP_CAPTURE_DEFAULTS,
type DesktopCaptureProgress,
} from "../domain/desktop/desktopCapture.js";
import { SessionRequestQueue } from "../desktop/DesktopSessionRequests.js";
import { withCommandCancellation } from "./commandCancellation.js";
import type { CliInstance } from "./types.js";
const positive = (fallback: number) =>
cliDecimalField(z.int().positive().default(fallback));
/** Options of `capture-desktop-application`. */
const captureOptions = z.object({
desktopEntry: z
.string()
.min(1)
.optional()
.describe("Desktop entry id (org.xfce.mousepad) or absolute .desktop path"),
executable: z
.string()
.min(1)
.optional()
.describe(
"Executable path or program name on PATH, instead of a desktop entry",
),
// An empty argument reaches the application as is. Path options refuse
// one, because normalizing it would select the current directory.
argument: z
.array(z.string())
.optional()
.describe("Argument appended to the launch command (repeatable)"),
output: localPathStringSchema.describe(
"New or empty directory for capture.json and retained content",
),
statePath: z
.array(localPathStringSchema)
.optional()
.describe(
"File or directory whose starting content is retained in addition to the application's state directories; a file brings its SQLite write-ahead log and rollback journal (repeatable)",
),
exclude: z
.array(localPathStringSchema)
.optional()
.describe("Path left out of every checkpoint walk (repeatable)"),
maxDurationSeconds: positive(
DESKTOP_CAPTURE_DEFAULTS.maxDurationSeconds,
).describe("Stop the session after this many seconds"),
maxEntries: positive(DESKTOP_CAPTURE_DEFAULTS.maxEntries).describe(
"Maximum paths recorded per checkpoint walk",
),
maxFileBytes: positive(DESKTOP_CAPTURE_DEFAULTS.maxFileBytes).describe(
"Largest file whose content is retained",
),
maxContentBytes: positive(DESKTOP_CAPTURE_DEFAULTS.maxContentBytes).describe(
"Total retained content",
),
packageFile: localPathStringSchema
.optional()
.describe(
"The .deb file the application was installed from; its control fields, SHA-256 and size are recorded, never the file",
),
});
/** The checkpoints a bundle's seed and a task's end state are read at. */
const checkpointOptions = {
seedCheckpoint: z
.string()
.optional()
.describe(
"Checkpoint whose state seeds the bundle; the baseline by default",
),
endCheckpoint: z
.string()
.optional()
.describe(
"Checkpoint whose state a task must pass on; the capture's last by default",
),
};
/**
* Run a desktop operation with SQLite and dconf readers, finishing through
* the shared command owner so the result, cancellation and the SQLite
* service's cleanup are all kept. The readers load here, so registering the
* command catalog never loads them.
*/
const withStateReaders = async (
environment: NodeJS.ProcessEnv,
signal: AbortSignal,
operation: string,
run: (readers: DesktopStateReaders) => Promise<Evidence | AnalysisError>,
) => {
const [
{ createSqliteDatabaseService },
{ readSqliteStateView },
{ dumpDconfDatabase },
] = await Promise.all([
import("../composition/sqlite.js"),
import("../application/desktop/DesktopSqliteState.js"),
import("../desktop/DconfDatabaseReader.js"),
]);
const sqlite = createSqliteDatabaseService(environment);
return finishAnalysisCommand(
async () => {
const result = await run({
sqlite: (database, readSignal) =>
readSqliteStateView(sqlite, database, readSignal),
dconf: (database, readSignal) =>
dumpDconfDatabase(database, environment, readSignal),
});
return result instanceof AnalysisError ? err(result) : ok(result);
},
{
close: async () => {
const failures: unknown[] = [];
try {
await sqlite.close();
} catch (cause: unknown) {
failures.push(cause);
}
if (operation === "export_desktop_application") {
try {
const { OwnedWorkerRuntime } =
await import("../process/OwnedWorkerRuntime.js");
await OwnedWorkerRuntime.retryPending();
} catch (cause: unknown) {
failures.push(cause);
}
}
if (failures.length === 1) throw failures[0];
if (failures.length > 1) {
const typed = failures.map((cause) => {
if (cause instanceof AnalysisError) return cause;
const observation =
cause instanceof ArtifactReaderFailure
? ArtifactReaderFailure.cleanupObservation(cause, [])
: {
resources: [],
reason:
cause instanceof Error ? cause.message : String(cause),
};
const failure = new ProviderCleanupError(
"rea-desktop",
observation.resources,
{ reason: observation.reason },
{ operation, cause },
);
bindArtifactFileCleanup(cause, failure);
return failure;
});
const combined = analysisErrorWithCleanupFailure(
typed[0]!,
typed[1]!,
operation,
);
combined.bindCleanupDependencies(typed);
throw combined;
}
},
},
"rea-desktop",
operation,
signal,
).then(projectCliAnalysisResult);
};
/** Register the export of a capture folder as a computer app bundle. */
const registerDesktopExportCommand = (
cli: CliInstance,
logger: Logger,
environment: NodeJS.ProcessEnv,
): void => {
cli.command(CLI_COMMANDS.exportDesktopApplication, {
description:
"Export a desktop application capture folder as an agentenv computer app bundle: class, env.toml, skill and seed",
args: z.object({
capture: localPathStringSchema.describe(
"Folder written by capture-desktop-application",
),
directory: localPathStringSchema.describe(
"New or empty bundle directory",
),
}),
options: z.object({
name: z
.string()
.optional()
.describe(
"Environment name; defaults to one derived from the application",
),
task: z
.array(localPathStringSchema)
.optional()
.describe(
"Task file to validate against the capture and add with its verifier and eval; a capture yields one",
),
packageUrl: z
.string()
.optional()
.describe(
"http or https URL the computer downloads the captured --package-file from when the bundle is put, for an application installed from a .deb file",
),
downloadInstall: z
.string()
.optional()
.describe(
"JSON file naming the AppImage or tarball a build computer unpacked the application from: kind, desktop_entry, url, sha256, architecture, and a tarball's program and name",
),
extraPackage: z
.array(z.string())
.optional()
.describe(
"Ubuntu 24.04 archive package installed beside the application, such as a runtime piece its package does not install; repeat for more",
),
syntheticSeed: z
.boolean()
.default(false)
.describe(
"The seed was made for the task, as in a build computer, not taken from the user's own data; the README says so instead of keeping the bundle local",
),
screens: z
.string()
.optional()
.describe(
"Folder holding a screenshot of each checkpoint as <name>.jpg, which a build computer takes; a task's screen_judge checks ship the seed's and the end's as their judge's examples",
),
...checkpointOptions,
}),
run: ({ args, options }) =>
withCommandCancellation((signal) =>
logCliCommand(
logger,
CLI_COMMANDS.exportDesktopApplication,
async () => {
const { exportDesktopApplication } =
await import("../application/desktop/DesktopApplicationExport.js");
return withStateReaders(
environment,
signal,
"export_desktop_application",
(readers) =>
exportDesktopApplication(
{
capture: absoluteCallerPath(args.capture),
directory: absoluteCallerPath(args.directory),
name: options.name,
tasks: (options.task ?? []).map((path) =>
absoluteCallerPath(path),
),
packageUrl: options.packageUrl,
downloadInstall:
options.downloadInstall === undefined
? undefined
: absoluteCallerPath(options.downloadInstall),
extraPackages: options.extraPackage,
seedCheckpoint: options.seedCheckpoint,
endCheckpoint: options.endCheckpoint,
syntheticSeed: options.syntheticSeed,
screens:
options.screens === undefined
? undefined
: absoluteCallerPath(options.screens),
},
readers,
signal,
),
);
},
),
),
});
};
/** Register the change summary an agent writes a task from. */
const registerDesktopTaskCommand = (
cli: CliInstance,
logger: Logger,
environment: NodeJS.ProcessEnv,
): void => {
cli.command(CLI_COMMANDS.describeDesktopApplicationTask, {
description:
"Describe how a desktop application capture's files changed from the seed checkpoint to the end, with suggested checks and a task draft to complete for export-desktop-application --task",
args: z.object({
capture: localPathStringSchema.describe(
"Folder written by capture-desktop-application",
),
}),
options: z.object(checkpointOptions),
run: ({ args, options }) =>
withCommandCancellation((signal) =>
logCliCommand(
logger,
CLI_COMMANDS.describeDesktopApplicationTask,
async () => {
const { draftDesktopApplicationTask } =
await import("../application/desktop/DesktopTaskDraft.js");
return withStateReaders(
environment,
signal,
"describe_desktop_application_task",
(readers) =>
draftDesktopApplicationTask(
absoluteCallerPath(args.capture),
readers,
signal,
{ seed: options.seedCheckpoint, end: options.endCheckpoint },
),
);
},
),
),
});
};
/** Register the state view over a capture folder. */
const registerDesktopStateCommand = (
cli: CliInstance,
logger: Logger,
environment: NodeJS.ProcessEnv,
): void => {
cli.command(CLI_COMMANDS.inspectDesktopApplicationState, {
description:
"Read a desktop application capture folder into the application's state: roles, typed JSON views of each version and kinds of change",
args: z.object({
capture: localPathStringSchema.describe(
"Folder written by capture-desktop-application",
),
}),
run: ({ args }) =>
withCommandCancellation((signal) =>
logCliCommand(
logger,
CLI_COMMANDS.inspectDesktopApplicationState,
async () => {
const { inspectDesktopApplicationState } =
await import("../application/desktop/DesktopApplicationState.js");
return withStateReaders(
environment,
signal,
"inspect_desktop_application_state",
(readers) =>
inspectDesktopApplicationState(
absoluteCallerPath(args.capture),
readers,
signal,
),
);
},
),
),
});
};
/** Register the interactive Linux desktop application capture. */
export const registerDesktopCaptureCommands = (
cli: CliInstance,
logger: Logger,
environment: NodeJS.ProcessEnv,
): void => {
registerDesktopStateCommand(cli, logger, environment);
registerDesktopExportCommand(cli, logger, environment);
registerDesktopTaskCommand(cli, logger, environment);
cli.command(CLI_COMMANDS.captureDesktopApplication, {
description:
"Launch a Linux desktop application, watch it while it is used and record its windows and state at checkpoints read from stdin",
options: captureOptions,
run: ({ options }) =>
withCommandCancellation((signal) =>
logCliCommand(
logger,
CLI_COMMANDS.captureDesktopApplication,
async () => {
const requests = new SessionRequestQueue(
createInterface({ input: process.stdin, terminal: false }),
);
try {
return await captureDesktopApplication(
{
desktopEntry: options.desktopEntry,
executable: options.executable,
arguments: options.argument ?? [],
output: options.output,
statePaths: options.statePath ?? [],
excludes: options.exclude ?? [],
maxDurationSeconds: options.maxDurationSeconds,
maxEntries: options.maxEntries,
maxFileBytes: options.maxFileBytes,
maxContentBytes: options.maxContentBytes,
packageFile:
options.packageFile === undefined
? undefined
: absoluteCallerPath(options.packageFile),
},
{
platform: process.platform,
environment,
callerDirectory: process.cwd(),
nextRequest: () => requests.next(),
report: async (message, event) => {
const progress: DesktopCaptureProgress = {
rea_progress: {
phase: "desktop_capture",
message,
...(event === undefined ? {} : { event }),
},
};
process.stderr.write(`${JSON.stringify(progress)}\n`);
},
},
signal,
);
} finally {
requests.close();
}
},
),
),
});
};