i/next-ai-draw-io
1
0
Fork 0
mirror of https://github.com/DayuanJiang/next-ai-draw-io.git synced 2026-10-10 19:49:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: older defects (batch C) and the second batch's review

Browse Source 
Chats:
- New Chat right after an answer saves that chat once. Saves run one at
  a time and read the chat on screen when their turn comes; a save
  scheduled for a chat that is no longer on screen is dropped. A chat
  whose id was still on its way to the URL no longer comes back after
  New Chat (the next answer went into it).
- Crossing the 768 px breakpoint keeps the chat panel: a streaming answer,
  unsaved messages and attachments stay. The panel gets the sizes of each
  side, and a panel collapsed on desktop opens on mobile.
- The chat's export waits for its own reply: an edit's history export
  still on its way no longer answers it with the older diagram, and two
  file saves at once no longer swap results.
- A second edit in one answer is previewed on the first edit's result.
- Stop also ends a running screenshot check; a chat that cannot be saved
  (storage full) can be left with "Continue without saving".
- Small diagrams with shapes count as diagrams; the tool card no longer
  crashes on malformed operations.

Quota and providers:
- Requests that reach the server's own endpoints count toward the quota:
  EdgeOne (always its own endpoint now), a private base URL whatever key
  header is sent, keyless Ollama without a URL. With the quota on, a
  redirect is followed only to a public address. The output cap applies
  to these requests too.
- Stop records the tokens of the steps that finished; the screenshot check
  counts its tokens without counting a request.
- EdgeOne configured only by AI_PROVIDER works, also in the admin Test,
  which forwards the access code. Azure set up only in the admin panel
  works in chat. The Test sends a Bedrock session token.
- The admin panel's Test of an entry without a URL uses the server's URL
  as the server does (no private address check for it); the admin panel
  no longer writes an Ollama URL.

MCP server:
- Write tools and start_session run one at a time, so two at once never
  drop each other's change; a cancelled call waiting its turn is skipped.
  get_diagram and export_diagram keep the session they started with.
- Export to .drawio first gets the user's latest edits from the browser.
- History thumbnails: one that arrives after the next AI write is
  dropped; a sync reply keeps the image; a version that changed only page
  settings is its own entry.
- A diagram over the 10 MB limit is saved without its image, or the user
  is told to download it (the server now answers 413 instead of cutting
  the connection).
- Labels holding text like id='1' or parent='1' are no longer read as
  attributes (a layer or a parent was deleted). A broken bare
  <mxGraphModel> file is refused.
- After a sync reply the tab no longer sends its autosave copy again.

Desktop and files:
- A newer switch of the same preset is not rolled back by an older one
  that failed. .env values with escaped quotes are read whole.
- MCP saved files: a file that could not be read stays protected while a
  folder without permission hides it, and is saved again once deleted.
- The desktop app reports "no chats" only when the count was read and no
  model settings are stored.
This commit is contained in:
dayuan.jiang
2026-10-05 18:57:02 +09:00
parent c75f74a5a0
commit c0fa997186
67 changed files with 3152 additions and 531 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/api/admin/test-model/route.ts
+12 -2
View File
@@ -48,6 +48,7 @@ export async function POST(req: Request) {
sameEndpoint && stored ? [stored] : [],
)
const serverUrl = globalBaseUrl(resolved.provider)
return validateModel(
new Request(new URL("/api/validate-model", req.url), {
method: "POST",
@@ -55,14 +56,23 @@ export async function POST(req: Request) {
"Content-Type": "application/json",
// Checked again there, in place of an access code
"x-admin-password": req.headers.get("x-admin-password") || "",
// The EdgeOne function checks the access code and Pages
// cookies, and its URL is built from the page's origin
"x-access-code": req.headers.get("x-access-code") || "",
cookie: req.headers.get("cookie") || "",
...(req.headers.get("origin") && {
origin: req.headers.get("origin") as string,
}),
},
body: JSON.stringify({
provider: resolved.provider,
apiKey: resolved.apiKey,
// Without a URL of its own, chat sends the entry's key to
// the server's <P>_BASE_URL: test that endpoint, not
// another one
baseUrl: resolved.baseUrl || globalBaseUrl(resolved.provider),
// another one. It is the server's own, which chat uses
// without the checks for a URL a user typed.
baseUrl: resolved.baseUrl || serverUrl,
...(!resolved.baseUrl && serverUrl && { serverBaseUrl: true }),
modelId: body.modelId,
awsAccessKeyId: resolved.awsAccessKeyId,
awsSecretAccessKey: resolved.awsSecretAccessKey,
app/api/chat/route.ts
+60 -25
View File
@@ -13,6 +13,7 @@ import { z } from "zod"
import { checkAccessCode, rejectCrossSite } from "@/lib/access-code"
import {
CACHE_POINT,
edgeOneEndpoint,
getAIModel,
getServerProvider,
SINGLE_SYSTEM_PROVIDERS,
@@ -189,15 +190,16 @@ async function handleChatRequest(req: Request): Promise<Response> {
}
// A server model's provider comes from its config: for one set up in
// the admin panel the header holds the provider name's slug
const isEdgeOne = (serverModelConfig.provider || provider) === "edgeone"
// the admin panel the header holds the provider name's slug. Without
// either, the server's own AI_PROVIDER.
const isEdgeOne =
(serverModelConfig.provider || provider || getServerProvider()) ===
"edgeone"
// For EdgeOne provider, construct full URL from request origin
// because createOpenAI needs absolute URL, not relative path
if (isEdgeOne && !baseUrl) {
const origin = req.headers.get("origin") || new URL(req.url).origin
baseUrl = `${origin}/api/edgeai`
}
// EdgeOne is this deployment's own function, whatever URL the request
// names: another host would get the user's EdgeOne cookies, and the
// quota counts it. Absolute, as the SDK needs.
if (isEdgeOne) baseUrl = edgeOneEndpoint(req)
// Same rule as validate-model: with ALLOW_PRIVATE_URLS=false a request may
// not point the server at a private or internal address
@@ -212,8 +214,12 @@ async function handleChatRequest(req: Request): Promise<Response> {
const cookieHeader = req.headers.get("cookie")
const clientOverrides = {
// Server model provider takes precedence over client header
provider: serverModelConfig.provider || provider,
// Server model provider takes precedence over client header; EdgeOne
// named only in AI_PROVIDER is named here, for its own base URL
provider:
serverModelConfig.provider ||
provider ||
(isEdgeOne ? "edgeone" : null),
baseUrl,
apiKey: req.headers.get("x-ai-api-key"),
// A server model runs the model it was configured with, whatever the header says
@@ -274,18 +280,24 @@ async function handleChatRequest(req: Request): Promise<Response> {
// === SERVER-SIDE QUOTA CHECK START ===
// Quota is opt-in (DYNAMODB_QUOTA_TABLE) and counts what runs on the
// server's keys, or on its keyless Ollama or EdgeOne. Decided by the key
// actually used: a key header the provider never reads must not skip it.
// EdgeOne never reads one; keyless Ollama at a private address is the
// server's own network.
// server's keys, or on the server's own endpoints: EdgeOne, its keyless
// Ollama, and anything at a private address (the server's network,
// which ignores a dummy key header). Bedrock and EdgeOne never use the
// base URL header. In the desktop app every endpoint is the user's.
const clientBaseUrl = normalizeBaseUrl(
req.headers.get("x-ai-base-url") ?? "",
)
const usesClientBaseUrl =
resolvedProvider !== "bedrock" && resolvedProvider !== "edgeone"
const onServerEndpoint =
(resolvedProvider === "edgeone" && !clientBaseUrl) ||
(resolvedProvider === "ollama" &&
!clientOverrides.apiKey &&
(!clientBaseUrl || (await isPrivateUrl(clientBaseUrl))))
process.env.NEXT_AI_DRAWIO_DESKTOP !== "1" &&
(resolvedProvider === "edgeone" ||
(resolvedProvider === "ollama" &&
!clientBaseUrl &&
!clientOverrides.apiKey) ||
(usesClientBaseUrl &&
!!clientBaseUrl &&
(await isPrivateUrl(clientBaseUrl))))
const countsQuota =
isQuotaEnabled() &&
(onServerCredentials || onServerEndpoint) &&
@@ -317,11 +329,11 @@ async function handleChatRequest(req: Request): Promise<Response> {
)
// The user setting can raise the budget only on their own key (in the
// desktop app every key is the user's); on the server's keys it can only
// lower it
// desktop app every key is the user's); on the server's keys or own
// endpoints it can only lower it
const maxOutputTokens = resolveMaxOutputTokens(
req.headers.get("x-max-output-tokens"),
onServerCredentials,
onServerCredentials || onServerEndpoint,
)
console.log(`[maxOutputTokens] ${maxOutputTokens}`)
@@ -353,8 +365,13 @@ async function handleChatRequest(req: Request): Promise<Response> {
${userInputText}
"""`
// Convert UIMessages to ModelMessages and add system message
const modelMessages = await convertToModelMessages(messages)
// Convert UIMessages to ModelMessages and add system message. A tool
// call that never got its result (the user stopped while it ran) is
// left out: the SDK would refuse this and every later request of the
// chat (MissingToolResultsError)
const modelMessages = await convertToModelMessages(messages, {
ignoreIncompleteToolCalls: true,
})
// DEBUG_LLM_PAYLOAD=true logs the incoming message structure
if (DEBUG_LLM_PAYLOAD) {
@@ -541,6 +558,8 @@ IMPORTANT: The "Current diagram XML" is the SINGLE SOURCE OF TRUTH for what's on
const allMessages = [...systemMessages, ...enhancedMessages]
// Set by onAbort, which records the finished steps' tokens itself
let stopped = false
const result = streamText({
model,
// The system messages carry cache points, so they go in messages.
@@ -606,7 +625,7 @@ IMPORTANT: The "Current diagram XML" is the SINGLE SOURCE OF TRUTH for what's on
// Record token usage for server-side quota tracking (if enabled)
// Use totalUsage (cumulative across all steps) instead of usage (final step only)
// inputTokens already includes cache reads and writes in AI SDK 6
if (countsQuota && totalUsage) {
if (countsQuota && totalUsage && !stopped) {
const totalTokens =
(totalUsage.inputTokens || 0) +
(totalUsage.outputTokens || 0)
@@ -618,7 +637,23 @@ IMPORTANT: The "Current diagram XML" is the SINGLE SOURCE OF TRUTH for what's on
console.error(error) // what AI SDK does without an onError
endTrace()
},
onAbort: () => endTrace(),
onAbort: ({ steps }) => {
stopped = true
endTrace()
// Stopped (or disconnected) after some steps finished: their
// tokens were used, or stopping every request after a costly
// first step would get around the token limits
if (countsQuota) {
const tokens = steps.reduce(
(sum, step) =>
sum +
(step.usage.inputTokens || 0) +
(step.usage.outputTokens || 0),
0,
)
if (tokens > 0) recordTokenUsage(userId, tokens)
}
},
tools: {
// Client-side tool that will be executed on the client
display_diagram: {
app/api/validate-diagram/route.ts
+43 -1
View File
@@ -6,6 +6,12 @@
import { Output, streamText } from "ai"
import { checkAccessCode, rejectCrossSite } from "@/lib/access-code"
import { getValidationModel } from "@/lib/ai-providers"
import {
checkAndIncrementRequest,
isQuotaEnabled,
recordTokenUsage,
} from "@/lib/dynamo-quota-manager"
import { getUserIdFromRequest } from "@/lib/user-id"
import { VALIDATION_SYSTEM_PROMPT } from "@/lib/validation-prompts"
import {
type ValidationResult,
@@ -78,6 +84,35 @@ export async function POST(req: Request): Promise<Response> {
)
}
// It runs the server's vision model: with the quota on, the daily
// and per-minute token limits apply, and its tokens are counted. Not
// the request limit, which is for chats: the day's last chat still
// gets its check, and a check does not count as a chat.
const userId = getUserIdFromRequest(req)
const countsQuota = isQuotaEnabled() && userId !== "anonymous"
if (countsQuota) {
const quotaCheck = await checkAndIncrementRequest(
userId,
{
requests: 0,
tokens: Number(process.env.DAILY_TOKEN_LIMIT) || 200000,
tpm: Number(process.env.TPM_LIMIT) || 20000,
},
0,
)
if (!quotaCheck.allowed) {
return Response.json(
{
error: quotaCheck.error,
type: quotaCheck.type,
used: quotaCheck.used,
limit: quotaCheck.limit,
},
{ status: 429 },
)
}
}
// Get the validation model
let model
try {
@@ -120,7 +155,14 @@ export async function POST(req: Request): Promise<Response> {
],
maxOutputTokens: 1024,
abortSignal: AbortSignal.timeout(timeout),
onFinish: ({ output }) => {
onFinish: ({ output, totalUsage }) => {
if (countsQuota && totalUsage) {
recordTokenUsage(
userId,
(totalUsage.inputTokens || 0) +
(totalUsage.outputTokens || 0),
)
}
if (sessionId && output) {
console.log(
`[validate-diagram] Session ${sessionId}: valid=${output.valid}, issues=${output.issues?.length ?? 0}`,
app/api/validate-model/route.ts
+31 -3
View File
@@ -3,7 +3,12 @@ import { NextResponse } from "next/server"
import { z } from "zod"
import { checkAccessCode, rejectCrossSite } from "@/lib/access-code"
import { checkAdminAuth } from "@/lib/admin/auth"
import { getAIModel, usesServerCredentials } from "@/lib/ai-providers"
import {
edgeOneEndpoint,
getAIModel,
globalBaseUrl,
usesServerCredentials,
} from "@/lib/ai-providers"
import { classifyLLMError } from "@/lib/llm-errors"
import { allowPrivateUrls, isPrivateUrl } from "@/lib/ssrf-protection"
import type { ProviderName } from "@/lib/types/model-config"
@@ -19,8 +24,11 @@ interface ValidateRequest {
awsAccessKeyId?: string
awsSecretAccessKey?: string
awsRegion?: string
awsSessionToken?: string
// Vertex AI specific
vertexApiKey?: string // Express Mode API key
// Set by the admin panel's Test: baseUrl is the server's <P>_BASE_URL
serverBaseUrl?: boolean
}
const TEST_TIMEOUT_MS = 15_000
@@ -47,11 +55,11 @@ export async function POST(req: Request) {
const {
provider,
apiKey,
baseUrl,
modelId,
awsAccessKeyId,
awsSecretAccessKey,
awsRegion,
awsSessionToken,
// Note: Express Mode only needs vertexApiKey
vertexApiKey,
} = body
@@ -62,9 +70,26 @@ export async function POST(req: Request) {
{ status: 400 },
)
}
// EdgeOne is this site's own function, as in the chat; the admin
// panel's Test sends no URL, and a relative one cannot be fetched
const baseUrl =
provider === "edgeone" ? edgeOneEndpoint(req) : body.baseUrl
// The admin panel's Test of an entry without a URL sends the
// server's own <P>_BASE_URL, which chat uses as it is: not a URL a
// user chose, so no private-address or redirect rules
const serverUrl =
body.serverBaseUrl === true &&
!!baseUrl &&
baseUrl === globalBaseUrl(provider) &&
!checkAdminAuth(req)
// SECURITY: Block SSRF attacks via custom baseUrl
if (baseUrl && !allowPrivateUrls() && (await isPrivateUrl(baseUrl))) {
if (
baseUrl &&
!serverUrl &&
!allowPrivateUrls() &&
(await isPrivateUrl(baseUrl))
) {
return NextResponse.json(
{ valid: false, error: "Invalid base URL" },
{ status: 400 },
@@ -122,9 +147,12 @@ export async function POST(req: Request) {
modelId,
apiKey,
baseUrl,
trustedBaseUrl: serverUrl,
awsAccessKeyId,
awsSecretAccessKey,
awsRegion,
// Temporary AWS credentials need it, as in the chat
awsSessionToken,
vertexApiKey,
// EdgeOne checks the Pages cookies and the access code
...(provider === "edgeone" && {