Files
pn-new-crm/src/lib/api/helpers.ts

369 lines
13 KiB
TypeScript
Raw Normal View History

feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
import { randomUUID } from 'node:crypto';
import { and, eq } from 'drizzle-orm';
import { NextRequest, NextResponse } from 'next/server';
import { auth } from '@/lib/auth';
import { db } from '@/lib/db';
import { portRoleOverrides, ports, userPortRoles, userProfiles } from '@/lib/db/schema';
import { type RolePermissions } from '@/lib/db/schema/users';
import { createAuditLog } from '@/lib/audit';
import { errorResponse, ForbiddenError } from '@/lib/errors';
import { logger } from '@/lib/logger';
feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
import { runWithRequestContext, getRequestContext } from '@/lib/request-context';
import {
checkRateLimit,
rateLimiters,
rateLimitHeaders,
type RateLimiterName,
} from '@/lib/rate-limit';
// ─── Types ────────────────────────────────────────────────────────────────────
/**
* Authenticated request context resolved by `withAuth`.
* Passed as the second argument to every wrapped route handler.
*/
export interface AuthContext {
userId: string;
portId: string;
portSlug: string;
/** true for super_admin users - bypasses all permission checks. */
isSuperAdmin: boolean;
/**
* Effective permissions after role + port override deep-merge.
* null when isSuperAdmin is true (super admins bypass permission checks).
*/
permissions: RolePermissions | null;
user: {
email: string;
name: string;
};
/** Client IP extracted from X-Forwarded-For header. */
ipAddress: string;
userAgent: string;
}
fix(types): unblock catch-all routes under stricter Next 15.5 typing + Phase 2B deps Two changes bundled (build was failing on the type fix; deps came along on the same branch). 1. RouteHandler / withAuth / withPermission are now generic over the route's params shape. Default stays `Record<string, string>` for the common `[id]`-style routes (no caller changes needed). Catch-all routes like `[...path]` declare their narrow shape via a type-arg: export const PATCH = withAuth<{ path: string[] }>( withPermission<{ path: string[] }>('files', 'manage_folders', async (req, ctx, params) => { /* params.path: string[] */ } ), ); Without this, Next.js 15.5+'s stricter route-type checking rejected the build because the inferred `params: Promise<{ path: string[] }>` for `[...path]` doesn't satisfy `Promise<Record<string, string>>`. Updated `src/app/api/v1/files/folders/[...path]/route.ts` (the only catch-all in the tree right now) to use the new generic. 2. Phase 2B deps (within-major-jump where the API didn't actually break): - @pdfme/common, @pdfme/generator, @pdfme/schemas: 5.5.10 → 6.1.2 (closes 3 mod XSS/SSRF/decompression-bomb advisories) - lucide-react: 0.460.0 → 1.14.0 - sonner: 1.7.4 → 2.0.7 - tailwind-merge: 2.6.1 → 3.5.0 Tests: 1185/1185 vitest. tsc clean. Local `next build` succeeds. Reverted (deferred to a focused PR): - @hookform/resolvers 5: Resolver<T> typing change requires per-form useForm migration - eslint 10: incompatible with @rushstack/eslint-patch (pulled in by eslint-config-next) - react-day-picker 10: ClassNames removed `table`; needs calendar.tsx migration - zod 4: 94 type errors cascading through drizzle insert types; needs comprehensive migration Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-08 17:07:07 +02:00
/**
* Route params type. Defaults to `Record<string, string>` for the common
* `[id]`-style routes. Catch-all routes (`[...slug]`) need to override
* `TParams` so Next.js 15.5+'s stricter route-type checking accepts the
* exported handler against the inferred `{ slug: string[] }` shape.
*/
export type RouteParams = Record<string, string | string[]>;
export type RouteHandler<TParams extends RouteParams = Record<string, string>, T = unknown> = (
req: NextRequest,
ctx: AuthContext,
params: TParams,
) => Promise<NextResponse<T>>;
// ─── deepMerge ───────────────────────────────────────────────────────────────
/**
* Recursively merges `source` into `target`.
* Used to apply port-level role permission overrides on top of the base role.
*/
export function deepMerge(
target: Record<string, unknown>,
source: Record<string, unknown>,
): Record<string, unknown> {
const result = { ...target };
for (const key of Object.keys(source)) {
const sourceVal = source[key];
const targetVal = result[key];
if (
typeof sourceVal === 'object' &&
sourceVal !== null &&
!Array.isArray(sourceVal) &&
typeof targetVal === 'object' &&
targetVal !== null &&
!Array.isArray(targetVal)
) {
result[key] = deepMerge(
targetVal as Record<string, unknown>,
sourceVal as Record<string, unknown>,
);
} else {
result[key] = sourceVal;
}
}
return result;
}
// ─── withAuth ────────────────────────────────────────────────────────────────
/**
* Validates the session, loads the user profile, resolves port context and
* applies port-level role overrides before calling the inner handler.
*
* Usage:
* ```ts
* export const GET = withAuth(handler);
* export const POST = withAuth(withPermission('clients', 'create', handler));
* ```
*/
fix(types): unblock catch-all routes under stricter Next 15.5 typing + Phase 2B deps Two changes bundled (build was failing on the type fix; deps came along on the same branch). 1. RouteHandler / withAuth / withPermission are now generic over the route's params shape. Default stays `Record<string, string>` for the common `[id]`-style routes (no caller changes needed). Catch-all routes like `[...path]` declare their narrow shape via a type-arg: export const PATCH = withAuth<{ path: string[] }>( withPermission<{ path: string[] }>('files', 'manage_folders', async (req, ctx, params) => { /* params.path: string[] */ } ), ); Without this, Next.js 15.5+'s stricter route-type checking rejected the build because the inferred `params: Promise<{ path: string[] }>` for `[...path]` doesn't satisfy `Promise<Record<string, string>>`. Updated `src/app/api/v1/files/folders/[...path]/route.ts` (the only catch-all in the tree right now) to use the new generic. 2. Phase 2B deps (within-major-jump where the API didn't actually break): - @pdfme/common, @pdfme/generator, @pdfme/schemas: 5.5.10 → 6.1.2 (closes 3 mod XSS/SSRF/decompression-bomb advisories) - lucide-react: 0.460.0 → 1.14.0 - sonner: 1.7.4 → 2.0.7 - tailwind-merge: 2.6.1 → 3.5.0 Tests: 1185/1185 vitest. tsc clean. Local `next build` succeeds. Reverted (deferred to a focused PR): - @hookform/resolvers 5: Resolver<T> typing change requires per-form useForm migration - eslint 10: incompatible with @rushstack/eslint-patch (pulled in by eslint-config-next) - react-day-picker 10: ClassNames removed `table`; needs calendar.tsx migration - zod 4: 94 type errors cascading through drizzle insert types; needs comprehensive migration Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-08 17:07:07 +02:00
export function withAuth<TParams extends RouteParams = Record<string, string>>(
handler: RouteHandler<TParams>,
): (req: NextRequest, routeContext: { params: Promise<TParams> }) => Promise<NextResponse> {
return async (req, routeContext) => {
feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
// Mint or accept a request id BEFORE entering the ALS frame so every
// log line + the response header reference the same value. Clients
// (or upstream proxies) may pre-supply via X-Request-Id; otherwise
// generate a fresh UUID. Pattern-validated so a crafted header can't
// smuggle log-injection chars.
const incomingId = req.headers.get('x-request-id');
const requestId =
incomingId && /^[A-Za-z0-9-]{8,64}$/.test(incomingId) ? incomingId : randomUUID();
feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
/** Stamp `X-Request-Id` onto every response leaving the wrapper. */
const tag = (res: NextResponse): NextResponse => {
res.headers.set('X-Request-Id', requestId);
return res;
};
return runWithRequestContext(
{
requestId,
portId: '',
userId: '',
method: req.method,
path: new URL(req.url).pathname,
startedAt: Date.now(),
},
async () => {
try {
// 1. Validate session via Better Auth.
const session = await auth.api.getSession({ headers: req.headers });
if (!session?.user) {
return tag(NextResponse.json({ error: 'Authentication required' }, { status: 401 }));
}
// 2. Load the CRM user profile (keyed on Better Auth user ID).
const profile = await db.query.userProfiles.findFirst({
where: eq(userProfiles.userId, session.user.id),
});
if (!profile || !profile.isActive) {
return tag(NextResponse.json({ error: 'Account disabled' }, { status: 403 }));
}
// 3. Resolve port context. Port id comes from the X-Port-Id
// header (set by the client after port selection), falling
// back to the user's default port preference. NEVER from the
// request body — SECURITY-GUIDELINES.md §2.1.
const portIdFromHeader = req.headers.get('X-Port-Id');
const portId =
portIdFromHeader ??
(profile.preferences as { defaultPortId?: string } | null)?.defaultPortId ??
null;
if (!portId && !profile.isSuperAdmin) {
return tag(NextResponse.json({ error: 'Port context required' }, { status: 400 }));
}
// 4. Resolve effective permissions.
let permissions: RolePermissions | null = null;
let portSlug = '';
if (!profile.isSuperAdmin && portId) {
const portRole = await db.query.userPortRoles.findFirst({
where: and(
eq(userPortRoles.userId, profile.userId),
eq(userPortRoles.portId, portId),
),
with: {
role: true,
port: true,
},
});
feat(platform): residential module + admin UI + reliability fixes Residential platform - New schema: residentialClients, residentialInterests (separate from marina/yacht clients) with migration 0010 - Service layer with CRUD + audit + sockets + per-port portal toggle - v1 + public API routes (/api/v1/residential/*, /api/public/residential-inquiries) - List + detail pages with inline editing for clients and interests - Per-user residentialAccess toggle on userPortRoles (migration 0011) - Permission keys: residential_clients, residential_interests - Sidebar nav + role form integration - Smoke spec covering page loads, UI create flow, public endpoint Admin & shared UI - Admin → Forms (form templates CRUD) with validators + service - Notification preferences page (in-app + email per type) - Email composition + accounts list + threads view - Branded auth shell shared across CRM + portal auth surfaces - Inline editing extended to yacht/company/interest detail pages - InlineTagEditor + per-entity tags endpoints (yachts, companies) - Notes service polymorphic across clients/interests/yachts/companies - Client list columns: yachtCount + companyCount badges - Reservation file-download via presigned URL (replaces stale <a href>) Route handler refactor - Extracted yachts/companies/berths reservation handlers to sibling handlers.ts files (Next.js 15 route.ts only allows specific exports) Reliability fixes - apiFetch double-stringify bug fixed across 13 components (apiFetch already JSON.stringifies its body; passing a stringified body produced double-encoded JSON which failed zod validation) - SocketProvider gated behind useSyncExternalStore-based mount check to avoid useSession() SSR crashes under React 19 + Next 15 - apiFetch falls back to URL-pathname → port-id resolution when the Zustand store hasn't hydrated yet (fresh contexts, e2e tests) - CRM invite flow (schema, service, route, email, dev script) - Dashboard route → [portSlug]/dashboard/page.tsx + redirect - Document the dev-server restart-after-migration gotcha in CLAUDE.md Tests - 5-case residential smoke spec - Integration test updates for new service signatures Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 21:54:32 +02:00
feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
if (!portRole) {
return tag(NextResponse.json({ error: 'No access to this port' }, { status: 403 }));
}
permissions = { ...(portRole.role.permissions as RolePermissions) };
portSlug = (portRole.port as { slug: string } | null)?.slug ?? '';
// Apply port-specific role overrides (deep-merge on top of base role).
const override = await db.query.portRoleOverrides.findFirst({
where: and(
eq(portRoleOverrides.portId, portId),
eq(portRoleOverrides.roleId, portRole.roleId),
),
});
if (override?.permissionOverrides) {
permissions = deepMerge(
permissions as unknown as Record<string, unknown>,
override.permissionOverrides as Record<string, unknown>,
) as RolePermissions;
}
// Per-user residential toggle.
if (portRole.residentialAccess && permissions) {
permissions = {
...permissions,
residential_clients: { view: true, create: true, edit: true, delete: true },
residential_interests: {
view: true,
create: true,
edit: true,
delete: true,
change_stage: true,
},
};
}
} else if (profile.isSuperAdmin && portId) {
const port = await db.query.ports.findFirst({
where: eq(ports.id, portId),
});
if (!port) {
return tag(NextResponse.json({ error: 'Port not found' }, { status: 404 }));
}
portSlug = port.slug;
}
// Now that the user + port are resolved, enrich the ALS frame
// so log lines + error_events rows pick up the identifiers.
const frame = getRequestContext();
if (frame) {
frame.userId = profile.userId;
frame.portId = portId ?? '';
}
const ctx: AuthContext = {
userId: profile.userId,
portId: portId ?? '',
portSlug,
isSuperAdmin: profile.isSuperAdmin,
permissions,
user: {
email: session.user.email,
name: session.user.name,
feat(platform): residential module + admin UI + reliability fixes Residential platform - New schema: residentialClients, residentialInterests (separate from marina/yacht clients) with migration 0010 - Service layer with CRUD + audit + sockets + per-port portal toggle - v1 + public API routes (/api/v1/residential/*, /api/public/residential-inquiries) - List + detail pages with inline editing for clients and interests - Per-user residentialAccess toggle on userPortRoles (migration 0011) - Permission keys: residential_clients, residential_interests - Sidebar nav + role form integration - Smoke spec covering page loads, UI create flow, public endpoint Admin & shared UI - Admin → Forms (form templates CRUD) with validators + service - Notification preferences page (in-app + email per type) - Email composition + accounts list + threads view - Branded auth shell shared across CRM + portal auth surfaces - Inline editing extended to yacht/company/interest detail pages - InlineTagEditor + per-entity tags endpoints (yachts, companies) - Notes service polymorphic across clients/interests/yachts/companies - Client list columns: yachtCount + companyCount badges - Reservation file-download via presigned URL (replaces stale <a href>) Route handler refactor - Extracted yachts/companies/berths reservation handlers to sibling handlers.ts files (Next.js 15 route.ts only allows specific exports) Reliability fixes - apiFetch double-stringify bug fixed across 13 components (apiFetch already JSON.stringifies its body; passing a stringified body produced double-encoded JSON which failed zod validation) - SocketProvider gated behind useSyncExternalStore-based mount check to avoid useSession() SSR crashes under React 19 + Next 15 - apiFetch falls back to URL-pathname → port-id resolution when the Zustand store hasn't hydrated yet (fresh contexts, e2e tests) - CRM invite flow (schema, service, route, email, dev script) - Dashboard route → [portSlug]/dashboard/page.tsx + redirect - Document the dev-server restart-after-migration gotcha in CLAUDE.md Tests - 5-case residential smoke spec - Integration test updates for new service signatures Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 21:54:32 +02:00
},
feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
ipAddress: req.headers.get('x-forwarded-for')?.split(',')[0]?.trim() ?? 'unknown',
userAgent: req.headers.get('user-agent') ?? 'unknown',
feat(platform): residential module + admin UI + reliability fixes Residential platform - New schema: residentialClients, residentialInterests (separate from marina/yacht clients) with migration 0010 - Service layer with CRUD + audit + sockets + per-port portal toggle - v1 + public API routes (/api/v1/residential/*, /api/public/residential-inquiries) - List + detail pages with inline editing for clients and interests - Per-user residentialAccess toggle on userPortRoles (migration 0011) - Permission keys: residential_clients, residential_interests - Sidebar nav + role form integration - Smoke spec covering page loads, UI create flow, public endpoint Admin & shared UI - Admin → Forms (form templates CRUD) with validators + service - Notification preferences page (in-app + email per type) - Email composition + accounts list + threads view - Branded auth shell shared across CRM + portal auth surfaces - Inline editing extended to yacht/company/interest detail pages - InlineTagEditor + per-entity tags endpoints (yachts, companies) - Notes service polymorphic across clients/interests/yachts/companies - Client list columns: yachtCount + companyCount badges - Reservation file-download via presigned URL (replaces stale <a href>) Route handler refactor - Extracted yachts/companies/berths reservation handlers to sibling handlers.ts files (Next.js 15 route.ts only allows specific exports) Reliability fixes - apiFetch double-stringify bug fixed across 13 components (apiFetch already JSON.stringifies its body; passing a stringified body produced double-encoded JSON which failed zod validation) - SocketProvider gated behind useSyncExternalStore-based mount check to avoid useSession() SSR crashes under React 19 + Next 15 - apiFetch falls back to URL-pathname → port-id resolution when the Zustand store hasn't hydrated yet (fresh contexts, e2e tests) - CRM invite flow (schema, service, route, email, dev script) - Dashboard route → [portSlug]/dashboard/page.tsx + redirect - Document the dev-server restart-after-migration gotcha in CLAUDE.md Tests - 5-case residential smoke spec - Integration test updates for new service signatures Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 21:54:32 +02:00
};
feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
const params = await routeContext.params;
return tag(await handler(req, ctx, params));
} catch (error) {
return tag(errorResponse(error));
feat(platform): residential module + admin UI + reliability fixes Residential platform - New schema: residentialClients, residentialInterests (separate from marina/yacht clients) with migration 0010 - Service layer with CRUD + audit + sockets + per-port portal toggle - v1 + public API routes (/api/v1/residential/*, /api/public/residential-inquiries) - List + detail pages with inline editing for clients and interests - Per-user residentialAccess toggle on userPortRoles (migration 0011) - Permission keys: residential_clients, residential_interests - Sidebar nav + role form integration - Smoke spec covering page loads, UI create flow, public endpoint Admin & shared UI - Admin → Forms (form templates CRUD) with validators + service - Notification preferences page (in-app + email per type) - Email composition + accounts list + threads view - Branded auth shell shared across CRM + portal auth surfaces - Inline editing extended to yacht/company/interest detail pages - InlineTagEditor + per-entity tags endpoints (yachts, companies) - Notes service polymorphic across clients/interests/yachts/companies - Client list columns: yachtCount + companyCount badges - Reservation file-download via presigned URL (replaces stale <a href>) Route handler refactor - Extracted yachts/companies/berths reservation handlers to sibling handlers.ts files (Next.js 15 route.ts only allows specific exports) Reliability fixes - apiFetch double-stringify bug fixed across 13 components (apiFetch already JSON.stringifies its body; passing a stringified body produced double-encoded JSON which failed zod validation) - SocketProvider gated behind useSyncExternalStore-based mount check to avoid useSession() SSR crashes under React 19 + Next 15 - apiFetch falls back to URL-pathname → port-id resolution when the Zustand store hasn't hydrated yet (fresh contexts, e2e tests) - CRM invite flow (schema, service, route, email, dev script) - Dashboard route → [portSlug]/dashboard/page.tsx + redirect - Document the dev-server restart-after-migration gotcha in CLAUDE.md Tests - 5-case residential smoke spec - Integration test updates for new service signatures Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 21:54:32 +02:00
}
feat(errors): platform-wide request ids + error codes + admin inspector End-to-end error-handling overhaul. A user hitting any failure now sees a plain-text message + stable error code + reference id. A super admin can paste the id into /admin/errors/<id> for the full request shape, sanitized body, error stack, and a heuristic likely-cause hint. REQUEST CONTEXT (AsyncLocalStorage) - src/lib/request-context.ts mints a per-request frame carrying requestId + portId + userId + method + path + start timestamp. - withAuth wraps every authenticated handler in runWithRequestContext and accepts an upstream X-Request-Id header (validated shape) or generates a fresh UUID. The id ALWAYS leaves on the X-Request-Id response header, including early-return 401/403/4xx paths. - Pino logger reads from the same context via mixin — every log line emitted during the request automatically carries the ids with no per-call threading. ERROR CODE REGISTRY - src/lib/error-codes.ts defines stable DOMAIN_REASON codes with HTTP status + plain-text user-facing message (no jargon, written for the rep on the phone with a customer). - New CodedError class wraps a registered code + optional internalMessage (admin-only — never sent to client). - Existing AppError subclasses got plain-text default rewrites so legacy throw sites improve immediately without migration. - High-impact services migrated to specific codes: expenses (RECEIPT_REQUIRED, INVOICE_LINKED), interest-berths (CROSS_PORT_LINK_REJECTED), berth-pdf (PDF_MAGIC_BYTE / PDF_EMPTY / PDF_TOO_LARGE / VERSION_ALREADY_CURRENT), recommender (INTEREST_PORT_MISMATCH). ERROR ENVELOPE - errorResponse always sets X-Request-Id header + requestId field. - 5xx responses include a "Quote error ID …" friendly line. - 4xx kept clean (validation, permission, not-found don't pollute the inspector — they're already in audit log). PERSISTENCE (error_events table, migration 0040) - One row per 5xx, keyed on requestId, with method/path/status/error name+message/stack head (4KB cap)/sanitized body excerpt (1KB cap; password/token/secret/etc keys redacted)/duration/IP/UA/metadata. - captureErrorEvent extracts Postgres SQLSTATE/severity/cause.code so the classifier can recognize FK / unique / NOT NULL / schema- drift violations. - Failure to persist is logged-not-thrown. LIKELY-CULPRIT CLASSIFIER (src/lib/error-classifier.ts) - 4-pass heuristic (first match wins): 1. Postgres SQLSTATE → human reason (23503 FK, 23505 unique, 42703 schema drift, 53300 connection limit, …) 2. Error class name (AbortError, TimeoutError, FetchError, ZodError) 3. Stack-path patterns (/lib/storage/, /lib/email/, documenso, openai|claude, /queue/workers/) 4. Free-text message keywords (econnrefused, rate limit, timeout, unauthorized|invalid api key) - Returns { label, hint, subsystem } for the inspector badge. CLIENT SIDE - apiFetch throws structured ApiError with message + code + requestId + details + retryAfter. - toastError() helper renders the standard 3-line toast: plain message / Error code: X / Reference ID: Y [Copy ID]. ADMIN INSPECTOR - /<port>/admin/errors lists captured 5xx with status badge + path + likely-culprit badge + truncated message + reference id. Filter by status code; auto-refresh via TanStack Query. - /<port>/admin/errors/<requestId> deep-dive: request shape, full error name+message+stack, sanitized body excerpt, raw metadata, registered-code lookup (so admin can compare to what user saw), likely-culprit hint with subsystem tag. - /<port>/admin/errors/codes is the in-app code reference page — every registered code grouped by domain prefix, searchable, with HTTP status + user message inline. Linked from inspector header so admins can flip to it while triaging. - Permission: admin.view_audit_log. Super admins see all ports; regular admins port-scoped. - system-monitoring dashboard now surfaces error_events alongside permission_denied audit + queue failed jobs (RecentError gains source: 'request' variant). DOCS - docs/error-handling.md walks through coded errors, plain-text message guidelines, client toasting, admin inspector usage, persistence rules, classifier internals, pruning, and the legacy → CodedError migration path. MIGRATION SAFETY - Audit confirmed all 41 migrations (0000-0040) apply cleanly in journal order against an empty DB. 0040 references ports(id) which exists from 0000. 0035/0038 don't deadlock under sequential psql -f. Removed redundant idx_ds_sent_by from 0038 (created in 0037). Tests: 1168/1168 vitest passing. tsc clean. - security-error-responses tests updated for plain-text messages + new optional response keys (code/requestId/message). - berth-pdf-versions tests assert stable error codes via toMatchObject({ code }) rather than message regex. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-05 14:12:59 +02:00
},
);
};
}
// ─── requireSuperAdmin ───────────────────────────────────────────────────────
/**
* Throws ForbiddenError when the caller is not a super-admin. Use inside
* route handlers (after withAuth) for endpoints that mutate global, cross-
* tenant state global roles, cross-port migrations, system jobs.
*
* Logs the denied attempt to the audit trail (mirrors withPermission).
*/
export function requireSuperAdmin(ctx: AuthContext, attemptedAction = 'super_admin_only'): void {
if (ctx.isSuperAdmin) return;
logger.warn({ userId: ctx.userId, attemptedAction }, 'Super-admin gate denied');
void createAuditLog({
userId: ctx.userId,
portId: ctx.portId,
action: 'permission_denied',
entityType: 'super_admin',
entityId: '',
metadata: { attemptedAction },
ipAddress: ctx.ipAddress,
userAgent: ctx.userAgent,
});
throw new ForbiddenError('Super admin access required');
}
// ─── withPermission ──────────────────────────────────────────────────────────
/**
* Wraps a route handler with a permission gate.
* Denied attempts are logged to the audit trail.
*
* Compose inside withAuth:
* ```ts
* export const DELETE = withAuth(withPermission('clients', 'delete', handler));
* ```
*/
fix(types): unblock catch-all routes under stricter Next 15.5 typing + Phase 2B deps Two changes bundled (build was failing on the type fix; deps came along on the same branch). 1. RouteHandler / withAuth / withPermission are now generic over the route's params shape. Default stays `Record<string, string>` for the common `[id]`-style routes (no caller changes needed). Catch-all routes like `[...path]` declare their narrow shape via a type-arg: export const PATCH = withAuth<{ path: string[] }>( withPermission<{ path: string[] }>('files', 'manage_folders', async (req, ctx, params) => { /* params.path: string[] */ } ), ); Without this, Next.js 15.5+'s stricter route-type checking rejected the build because the inferred `params: Promise<{ path: string[] }>` for `[...path]` doesn't satisfy `Promise<Record<string, string>>`. Updated `src/app/api/v1/files/folders/[...path]/route.ts` (the only catch-all in the tree right now) to use the new generic. 2. Phase 2B deps (within-major-jump where the API didn't actually break): - @pdfme/common, @pdfme/generator, @pdfme/schemas: 5.5.10 → 6.1.2 (closes 3 mod XSS/SSRF/decompression-bomb advisories) - lucide-react: 0.460.0 → 1.14.0 - sonner: 1.7.4 → 2.0.7 - tailwind-merge: 2.6.1 → 3.5.0 Tests: 1185/1185 vitest. tsc clean. Local `next build` succeeds. Reverted (deferred to a focused PR): - @hookform/resolvers 5: Resolver<T> typing change requires per-form useForm migration - eslint 10: incompatible with @rushstack/eslint-patch (pulled in by eslint-config-next) - react-day-picker 10: ClassNames removed `table`; needs calendar.tsx migration - zod 4: 94 type errors cascading through drizzle insert types; needs comprehensive migration Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-08 17:07:07 +02:00
export function withPermission<TParams extends RouteParams = Record<string, string>>(
resource: keyof RolePermissions,
action: string,
fix(types): unblock catch-all routes under stricter Next 15.5 typing + Phase 2B deps Two changes bundled (build was failing on the type fix; deps came along on the same branch). 1. RouteHandler / withAuth / withPermission are now generic over the route's params shape. Default stays `Record<string, string>` for the common `[id]`-style routes (no caller changes needed). Catch-all routes like `[...path]` declare their narrow shape via a type-arg: export const PATCH = withAuth<{ path: string[] }>( withPermission<{ path: string[] }>('files', 'manage_folders', async (req, ctx, params) => { /* params.path: string[] */ } ), ); Without this, Next.js 15.5+'s stricter route-type checking rejected the build because the inferred `params: Promise<{ path: string[] }>` for `[...path]` doesn't satisfy `Promise<Record<string, string>>`. Updated `src/app/api/v1/files/folders/[...path]/route.ts` (the only catch-all in the tree right now) to use the new generic. 2. Phase 2B deps (within-major-jump where the API didn't actually break): - @pdfme/common, @pdfme/generator, @pdfme/schemas: 5.5.10 → 6.1.2 (closes 3 mod XSS/SSRF/decompression-bomb advisories) - lucide-react: 0.460.0 → 1.14.0 - sonner: 1.7.4 → 2.0.7 - tailwind-merge: 2.6.1 → 3.5.0 Tests: 1185/1185 vitest. tsc clean. Local `next build` succeeds. Reverted (deferred to a focused PR): - @hookform/resolvers 5: Resolver<T> typing change requires per-form useForm migration - eslint 10: incompatible with @rushstack/eslint-patch (pulled in by eslint-config-next) - react-day-picker 10: ClassNames removed `table`; needs calendar.tsx migration - zod 4: 94 type errors cascading through drizzle insert types; needs comprehensive migration Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-08 17:07:07 +02:00
handler: RouteHandler<TParams>,
): RouteHandler<TParams> {
return async (req, ctx, params) => {
if (!ctx.isSuperAdmin) {
const resourcePerms = ctx.permissions?.[resource] as Record<string, boolean> | undefined;
if (!resourcePerms || !resourcePerms[action]) {
logger.warn({ userId: ctx.userId, resource, action }, 'Permission denied');
// Log the denied attempt - fire-and-forget; audit must never block response.
void createAuditLog({
userId: ctx.userId,
portId: ctx.portId,
action: 'permission_denied',
entityType: resource,
entityId: '',
metadata: { attemptedAction: action },
ipAddress: ctx.ipAddress,
userAgent: ctx.userAgent,
});
return NextResponse.json({ error: 'Insufficient permissions' }, { status: 403 });
}
}
return handler(req, ctx, params);
};
}
// ─── withRateLimit ───────────────────────────────────────────────────────────
/**
* Wraps a route handler with a per-user rate-limit gate. Compose inside
* withAuth so the userId is available - falls back to IP for anonymous
* routes (we don't currently expose any).
*
* 429 responses include `X-RateLimit-Limit` / `Remaining` / `Reset` headers
* and a `Retry-After` hint.
*
* ```ts
* export const POST = withAuth(
* withPermission('expenses', 'create',
* withRateLimit('ocr', handler)
* )
* );
* ```
*/
export function withRateLimit(name: RateLimiterName, handler: RouteHandler): RouteHandler {
const config = rateLimiters[name];
return async (req, ctx, params) => {
const identifier = `${ctx.userId}`;
const result = await checkRateLimit(identifier, config);
if (!result.allowed) {
const retryAfterSec = Math.max(1, Math.ceil((result.resetAt - Date.now()) / 1000));
logger.warn(
{ userId: ctx.userId, limiter: name, limit: result.limit },
'Rate limit exceeded',
);
return NextResponse.json(
{ error: 'Rate limit exceeded', retryAfter: retryAfterSec },
{
status: 429,
headers: {
...rateLimitHeaders(result),
'Retry-After': String(retryAfterSec),
},
},
);
}
return handler(req, ctx, params);
};
}