Policies: enforce input on uploads only; badge follows edited files (#6627)

This commit is contained in:
Reece Browne
2026-06-11 21:27:44 +01:00
committed by GitHub
parent ef65e6b015
commit 6f1c19c179
8 changed files with 466 additions and 25 deletions
@@ -83,6 +83,11 @@ export function usePolicyAutoRun(): void {
);
for (const [categoryId, s] of active) {
for (const stub of fileStubs) {
// Input-mode policies enforce only on files that actually entered the
// system as an upload — not on files a tool/automation produced in-app
// (versioned edits or independent artifacts like convert/split/merge).
// Those are enforced only by export-mode policies, at export time.
if (stub.derivedFromTool) continue;
const key = dispatchKey(categoryId, stub.id);
// Skip if already run (persisted) or a dispatch is in flight — the
// in-memory guard prevents double-firing during the async wait.
@@ -0,0 +1,113 @@
import { describe, it, expect } from "vitest";
import { buildPolicyBadgeMap } from "@app/hooks/usePolicyFileBadges";
import type { PolicyRunRecord } from "@app/components/policies/policyRunStore";
const NOW = 1_000_000;
const labels = new Map([
["security", "Security"],
["watermark", "Watermark"],
]);
function run(overrides: Partial<PolicyRunRecord>): PolicyRunRecord {
return {
runId: "r",
categoryId: "security",
fileId: "in",
fileName: "in.pdf",
fileSize: 1,
status: "COMPLETED",
outputs: [],
outputFileIds: ["out"],
error: null,
startedAt: NOW - 1_000, // recent by default
...overrides,
};
}
describe("buildPolicyBadgeMap — badge follows the document onto derived files", () => {
it("badges a policy's direct output, and marks it recent within the window", () => {
const map = buildPolicyBadgeMap([run({})], [{ id: "out" }], labels, NOW);
const badges = map.get("out") ?? [];
expect(badges.map((b) => b.id)).toEqual(["security"]);
expect(badges[0].recent).toBe(true);
});
it("a versioned edit inherits the badge via parentFileId (never glows)", () => {
const map = buildPolicyBadgeMap(
[run({})],
[{ id: "out" }, { id: "edit", parentFileId: "out" }],
labels,
NOW,
);
const edit = map.get("edit") ?? [];
expect(edit.map((b) => b.id)).toEqual(["security"]);
expect(edit[0].recent).toBe(false);
});
it("SPLIT parts inherit the badge via sourceFileIds, though they have no parent", () => {
// Split consumes the policy output "out" → two fresh roots, no parentFileId,
// each recording sourceFileIds=["out"]. "out" itself is gone from the
// workbench (consumed) but still lives in the run store.
const map = buildPolicyBadgeMap(
[run({})],
[
{ id: "part1", sourceFileIds: ["out"] },
{ id: "part2", sourceFileIds: ["out"] },
],
labels,
NOW,
);
expect((map.get("part1") ?? []).map((b) => b.id)).toEqual(["security"]);
expect((map.get("part2") ?? []).map((b) => b.id)).toEqual(["security"]);
expect((map.get("part1") ?? [])[0].recent).toBe(false);
});
it("resolves transitively when an intermediate edit was consumed/removed", () => {
// redact → edit (consumed) → split. The split part's sourceFileIds carries
// the original output id directly, so the badge still resolves.
const map = buildPolicyBadgeMap(
[run({})],
[{ id: "part", sourceFileIds: ["editGone", "out"] }],
labels,
NOW,
);
expect((map.get("part") ?? []).map((b) => b.id)).toEqual(["security"]);
});
it("MERGE output inherits every input's badge", () => {
const map = buildPolicyBadgeMap(
[
run({ runId: "r1", categoryId: "security", outputFileIds: ["a"] }),
run({ runId: "r2", categoryId: "watermark", outputFileIds: ["b"] }),
],
[{ id: "merged", sourceFileIds: ["a", "b"] }],
labels,
NOW,
);
expect((map.get("merged") ?? []).map((b) => b.id).sort()).toEqual([
"security",
"watermark",
]);
});
it("a file with no policy provenance gets no badge", () => {
const map = buildPolicyBadgeMap(
[run({})],
[{ id: "out" }, { id: "unrelated", sourceFileIds: ["someUpload"] }],
labels,
NOW,
);
expect(map.has("unrelated")).toBe(false);
});
it("inherited badges never glow even when the source run is recent", () => {
const map = buildPolicyBadgeMap(
[run({ startedAt: NOW })], // maximally recent
[{ id: "out" }, { id: "part", sourceFileIds: ["out"] }],
labels,
NOW,
);
expect((map.get("out") ?? [])[0].recent).toBe(true);
expect((map.get("part") ?? [])[0].recent).toBe(false);
});
});
@@ -1,5 +1,7 @@
import { useMemo } from "react";
import { usePolicyRuns } from "@app/components/policies/policyRunStore";
import type { PolicyRunRecord } from "@app/components/policies/policyRunStore";
import { useAllFiles } from "@app/contexts/FileContext";
import { loadPolicyCatalog } from "@app/services/policyCatalog";
import { ROW_ACCENT } from "@app/components/policies/policyStatus";
import type { FileItemPolicyRef } from "@app/components/shared/FileSidebarFileItem";
@@ -18,38 +20,110 @@ const ACCENT_VAR: Record<string, string> = {
red: "var(--color-red)",
};
/** Minimal provenance shape needed to resolve a file's inherited badges. */
type LineageStub = {
id: string;
parentFileId?: string;
sourceFileIds?: string[];
};
/** Merge a ref into a list, deduping by policy id. A direct (recent) hit wins
* the glow over an inherited one for the same policy. */
function mergeRef(list: FileItemPolicyRef[], ref: FileItemPolicyRef): void {
const existing = list.find((p) => p.id === ref.id);
if (!existing) {
list.push(ref);
} else if (ref.recent) {
existing.recent = true;
}
}
/**
* Pure core of {@link usePolicyFileBadges} (no React/storage deps, so it's
* unit-testable). Returns the policies that have run on each file, keyed by
* fileId.
*
* A policy run is pinned to a specific output fileId, so a later tool edit
* produces a NEW file that wouldn't carry the badge — "been through a policy"
* would vanish the moment you edit the file. To keep the badge with the
* document, every file also INHERITS the badges of the files it was derived
* from: its transitive `sourceFileIds` (recorded at the consume boundary, so it
* covers split/merge/convert too) plus, defensively, its `parentFileId`.
* Because `sourceFileIds` is transitive, a flat lookup suffices — no chain walk,
* and it survives a consumed intermediate. Inherited badges never glow
* (recent=false): only the original application does.
*/
export function buildPolicyBadgeMap(
runs: ReadonlyArray<PolicyRunRecord>,
stubs: ReadonlyArray<LineageStub>,
labelById: ReadonlyMap<string, string>,
now: number,
): Map<string, FileItemPolicyRef[]> {
// Direct badges: a file that IS a policy run's output.
const directByFile = new Map<string, FileItemPolicyRef[]>();
for (const run of runs) {
const name = labelById.get(run.categoryId);
if (!name) continue;
const recent = now - run.startedAt < RECENT_MS;
for (const fileId of run.outputFileIds ?? []) {
const list = directByFile.get(fileId) ?? [];
if (!list.some((p) => p.id === run.categoryId)) {
list.push({
id: run.categoryId,
name,
accentColor: ACCENT_VAR[ROW_ACCENT[run.categoryId] ?? "blue"],
recent,
});
directByFile.set(fileId, list);
}
}
}
// Seed the result with the direct badges (cloning the arrays so the lineage
// pass can mutate freely without touching the direct map).
const result = new Map<string, FileItemPolicyRef[]>();
for (const [id, refs] of directByFile) {
if (refs.length)
result.set(
id,
refs.map((r) => ({ ...r })),
);
}
// Inheritance pass: a derived file carries the badges of every file it came
// from. `sourceFileIds` is the transitive provenance set (so a flat lookup
// catches even ancestors whose intermediate edits were consumed), and
// `parentFileId` is included defensively for any child not created via a
// consume. Inherited badges are marked recent=false (carried, not applied).
for (const stub of stubs) {
const sources = new Set<string>(stub.sourceFileIds ?? []);
if (stub.parentFileId) sources.add(stub.parentFileId);
for (const src of sources) {
const srcBadges = directByFile.get(src);
if (!srcBadges?.length) continue;
const list = result.get(stub.id) ?? [];
for (const ref of srcBadges) mergeRef(list, { ...ref, recent: false });
result.set(stub.id, list);
}
}
return result;
}
/**
* Distinct policies that have produced each file, keyed by fileId, derived from
* the reactive policy run store. Drives the file sidebar's shield badges. The
* badge marks a policy's OUTPUT (the versioned/added result), not the input it
* ran on — so it keys off each run's imported output fileIds. Shadows the core
* stub via the {@code @app/*} alias cascade.
* badge follows a document down its tool-edit chain — see
* {@link buildPolicyBadgeMap}. Shadows the core stub via the {@code @app/*}
* alias cascade.
*/
export function usePolicyFileBadges(): Map<string, FileItemPolicyRef[]> {
const runs = usePolicyRuns();
const { fileStubs } = useAllFiles();
return useMemo(() => {
const labelById = new Map(
loadPolicyCatalog().categories.map((c) => [c.id, c.label]),
);
const now = Date.now();
const byFile = new Map<string, FileItemPolicyRef[]>();
for (const run of runs) {
const name = labelById.get(run.categoryId);
if (!name) continue;
const recent = now - run.startedAt < RECENT_MS;
for (const fileId of run.outputFileIds ?? []) {
const list = byFile.get(fileId) ?? [];
if (!list.some((p) => p.id === run.categoryId)) {
list.push({
id: run.categoryId,
name,
accentColor: ACCENT_VAR[ROW_ACCENT[run.categoryId] ?? "blue"],
recent,
});
byFile.set(fileId, list);
}
}
}
return byFile;
}, [runs]);
return buildPolicyBadgeMap(runs, fileStubs, labelById, Date.now());
}, [runs, fileStubs]);
}