ZacharyZhang-NY/AgentCompany
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 2 Wiki Activity

Pin work_products identity to (company, workspace, path)

Browse Source 
The natural key used to be (companyId, provider, externalId), so two
workspaces with the same workspace-relative path collided onto one
row. Provenance was also frozen at the first write: `created_by_run_id`
never refreshed, so a run that edited a pre-existing file never
qualified as a deliverable and the DocumentCenter "open" link could
still jump into the first workspace after a repo move.

Migration v45 adds `workspace_id` + `last_touched_by_run_id`, backfills
`workspace_id` via the task/run join, seeds `last_touched_by_run_id =
created_by_run_id`, and installs a partial unique index on
`(company_id, workspace_id, provider, external_id)` scoped to rows
that have a workspace. The artifact projector now records
`workspace_id` on insert and refreshes `last_touched_by_run_id` on
every upsert. The deliverable signal query reads either provenance
column so a run that only edited a pre-existing file still counts.
`resolveWorkProductTarget` prefers the stored `workspace_id` and falls
back to the task/run join for legacy NULL rows.
This commit is contained in:
ZacharyZhang-NY committed 2026-04-19 15:43:48 -04:00
1 parent d1483dcd5d
commit d9b35d73ab
12 files changed
+179 -25
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

src/main/database/documents.ts
+56 -18
View File
@@ -8,19 +8,26 @@ import { buildConfigRevisionMethods } from "./config-revisions";
function resolveWorkspaceRoot(
getDb: () => DatabaseSync,
companyId: string,
workspaceId: string | null,
taskId: string | null,
createdByRunId: string | null,
runId: string | null,
): string | null {
if (workspaceId) {
const row = getDb().prepare(
"SELECT local_path AS localPath FROM workspaces WHERE id = ? AND company_id = ?",
).get(workspaceId, companyId) as { localPath: string } | undefined;
if (row?.localPath) return row.localPath;
}
if (taskId) {
const row = getDb().prepare(
"SELECT w.local_path AS localPath FROM tasks t JOIN workspaces w ON w.id = t.workspace_id WHERE t.id = ? AND t.company_id = ?",
).get(taskId, companyId) as { localPath: string } | undefined;
if (row?.localPath) return row.localPath;
}
if (createdByRunId) {
if (runId) {
const row = getDb().prepare(
"SELECT w.local_path AS localPath FROM runs r JOIN workspaces w ON w.id = r.workspace_id WHERE r.id = ? AND r.company_id = ?",
).get(createdByRunId, companyId) as { localPath: string } | undefined;
).get(runId, companyId) as { localPath: string } | undefined;
if (row?.localPath) return row.localPath;
}
return null;
@@ -222,6 +229,7 @@ export function buildDocumentMethods(
createWorkProduct(input: {
companyId: string; taskId: string | null; projectId?: string | null;
workspaceId?: string | null;
type: string; provider: string; externalId?: string | null;
title: string; url?: string | null; status?: string; reviewState?: string;
isPrimary?: boolean; summary?: string | null; metadataJson?: string | null;
@@ -235,35 +243,47 @@ export function buildDocumentMethods(
).run(ts, input.companyId, input.taskId, input.type);
}
getDb().prepare(
`INSERT INTO work_products (id, company_id, project_id, task_id, type, provider, external_id, title, url, status, review_state, is_primary, summary, metadata_json, created_by_run_id, created_at, updated_at)
VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)`,
`INSERT INTO work_products (id, company_id, project_id, task_id, workspace_id, type, provider, external_id, title, url, status, review_state, is_primary, summary, metadata_json, created_by_run_id, last_touched_by_run_id, created_at, updated_at)
VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)`,
).run(
id, input.companyId, input.projectId ?? null, input.taskId ?? null,
input.workspaceId ?? null,
input.type, input.provider, input.externalId ?? null,
input.title, input.url ?? null, input.status ?? "active",
input.reviewState ?? "none", input.isPrimary ? 1 : 0,
input.summary ?? null, input.metadataJson ?? null,
input.createdByRunId ?? null, ts, ts,
input.createdByRunId ?? null, input.createdByRunId ?? null, ts, ts,
);
return id;
},
/**
* Look up a work product by its natural identity — the tuple
* (companyId, provider, externalId) uniquely names a registered
* artifact. Returns null when not found. Used by the artifact
* projector to make tool-output registration idempotent: the first
* write inserts, subsequent writes (same path, same provider) go
* through `updateWorkProduct` instead of spamming duplicates.
* Look up a work product by its natural identity. The v45 schema
* pinned identity to `(companyId, workspaceId, provider, externalId)`
* so two workspaces with the same relative path no longer collide.
* When `workspaceId` is null we fall back to the legacy
* `(companyId, provider, externalId)` scope so historical rows that
* pre-date the workspace backfill can still be found and updated.
*/
findWorkProductByExternalId(
companyId: string,
provider: string,
externalId: string,
workspaceId: string | null,
): { id: string; taskId: string | null } | null {
if (workspaceId) {
const row = getDb()
.prepare(
"SELECT id, task_id as taskId FROM work_products WHERE company_id = ? AND workspace_id = ? AND provider = ? AND external_id = ? LIMIT 1",
)
.get(companyId, workspaceId, provider, externalId) as
| { id: string; taskId: string | null }
| undefined;
return row ?? null;
}
const row = getDb()
.prepare(
"SELECT id, task_id as taskId FROM work_products WHERE company_id = ? AND provider = ? AND external_id = ? LIMIT 1",
"SELECT id, task_id as taskId FROM work_products WHERE company_id = ? AND workspace_id IS NULL AND provider = ? AND external_id = ? LIMIT 1",
)
.get(companyId, provider, externalId) as
| { id: string; taskId: string | null }
@@ -278,6 +298,7 @@ export function buildDocumentMethods(
type?: string;
provider?: string;
externalId?: string | null;
workspaceId?: string | null;
title?: string;
url?: string | null;
status?: string;
@@ -285,6 +306,7 @@ export function buildDocumentMethods(
isPrimary?: boolean;
summary?: string | null;
metadataJson?: string | null;
lastTouchedByRunId?: string | null;
},
): number {
const ts = nowIso();
@@ -294,6 +316,7 @@ export function buildDocumentMethods(
["type", "type", (v) => (v as string) ?? null],
["provider", "provider", (v) => (v as string) ?? null],
["externalId", "external_id", (v) => (v as string | null) ?? null],
["workspaceId", "workspace_id", (v) => (v as string | null) ?? null],
["title", "title", (v) => (v as string) ?? null],
["url", "url", (v) => (v as string | null) ?? null],
["status", "status", (v) => (v as string) ?? null],
@@ -301,6 +324,7 @@ export function buildDocumentMethods(
["isPrimary", "is_primary", (v) => (v ? 1 : 0)],
["summary", "summary", (v) => (v as string | null) ?? null],
["metadataJson", "metadata_json", (v) => (v as string | null) ?? null],
["lastTouchedByRunId", "last_touched_by_run_id", (v) => (v as string | null) ?? null],
];
for (const [key, col, coerce] of map) {
if (key in patch) {
@@ -329,9 +353,9 @@ export function buildDocumentMethods(
*/
resolveWorkProductTarget(workProductId: string, companyId: string): { kind: "url" | "path"; value: string } | null {
const row = getDb().prepare(
"SELECT provider, external_id, url, task_id, created_by_run_id FROM work_products WHERE id = ? AND company_id = ?",
"SELECT provider, external_id, url, task_id, created_by_run_id, last_touched_by_run_id, workspace_id FROM work_products WHERE id = ? AND company_id = ?",
).get(workProductId, companyId) as
| { provider: string; external_id: string | null; url: string | null; task_id: string | null; created_by_run_id: string | null }
| { provider: string; external_id: string | null; url: string | null; task_id: string | null; created_by_run_id: string | null; last_touched_by_run_id: string | null; workspace_id: string | null }
| undefined;
if (!row) return null;
@@ -340,7 +364,13 @@ export function buildDocumentMethods(
}
if (row.provider !== "filesystem" || !row.external_id) return null;
const workspaceRoot = resolveWorkspaceRoot(getDb, companyId, row.task_id, row.created_by_run_id);
const workspaceRoot = resolveWorkspaceRoot(
getDb,
companyId,
row.workspace_id,
row.task_id,
row.last_touched_by_run_id ?? row.created_by_run_id,
);
if (!workspaceRoot) return null;
const absolute = joinWorkspaceRelative(workspaceRoot, row.external_id);
@@ -351,19 +381,23 @@ export function buildDocumentMethods(
listWorkProductsForTask(taskId: string, companyId: string) {
return readAll<{
id: string; company_id: string; project_id: string | null; task_id: string;
workspace_id: string | null;
type: string; provider: string; external_id: string | null; title: string;
url: string | null; status: string; review_state: string; is_primary: number;
summary: string | null; metadata_json: string | null; created_by_run_id: string | null;
summary: string | null; metadata_json: string | null;
created_by_run_id: string | null; last_touched_by_run_id: string | null;
created_at: string; updated_at: string;
}>(
"SELECT * FROM work_products WHERE task_id = ? AND company_id = ? ORDER BY is_primary DESC, updated_at DESC",
taskId, companyId,
).map((r) => ({
id: r.id, companyId: r.company_id, projectId: r.project_id, taskId: r.task_id,
workspaceId: r.workspace_id,
type: r.type, provider: r.provider, externalId: r.external_id, title: r.title,
url: r.url, status: r.status, reviewState: r.review_state,
isPrimary: Boolean(r.is_primary), summary: r.summary,
metadataJson: r.metadata_json, createdByRunId: r.created_by_run_id,
lastTouchedByRunId: r.last_touched_by_run_id,
createdAt: r.created_at, updatedAt: r.updated_at,
}));
},
@@ -379,19 +413,23 @@ export function buildDocumentMethods(
listWorkProductsForCompany(companyId: string, limit = 500) {
return readAll<{
id: string; company_id: string; project_id: string | null; task_id: string | null;
workspace_id: string | null;
type: string; provider: string; external_id: string | null; title: string;
url: string | null; status: string; review_state: string; is_primary: number;
summary: string | null; metadata_json: string | null; created_by_run_id: string | null;
summary: string | null; metadata_json: string | null;
created_by_run_id: string | null; last_touched_by_run_id: string | null;
created_at: string; updated_at: string;
}>(
"SELECT * FROM work_products WHERE company_id = ? ORDER BY updated_at DESC LIMIT ?",
companyId, limit,
).map((r) => ({
id: r.id, companyId: r.company_id, projectId: r.project_id, taskId: r.task_id,
workspaceId: r.workspace_id,
type: r.type, provider: r.provider, externalId: r.external_id, title: r.title,
url: r.url, status: r.status, reviewState: r.review_state,
isPrimary: Boolean(r.is_primary), summary: r.summary,
metadataJson: r.metadata_json, createdByRunId: r.created_by_run_id,
lastTouchedByRunId: r.last_touched_by_run_id,
createdAt: r.created_at, updatedAt: r.updated_at,
}));
},