mirror of
https://github.com/ChatGPTNextWeb/ChatGPT-Next-Web.git
synced 2025-11-12 20:23:45 +08:00
优化bedrock流式消息处理的缓冲机制,简化app和后台api调用逻辑判断和处理
This commit is contained in:
477
app/utils/aws.ts
477
app/utils/aws.ts
@@ -67,8 +67,9 @@ export interface SignParams {
|
||||
region: string;
|
||||
accessKeyId: string;
|
||||
secretAccessKey: string;
|
||||
body: string;
|
||||
body: string | object;
|
||||
service: string;
|
||||
headers?: Record<string, string>;
|
||||
isStreaming?: boolean;
|
||||
}
|
||||
|
||||
@@ -99,7 +100,7 @@ function normalizeHeaderValue(value: string): string {
|
||||
return value.replace(/\s+/g, " ").trim();
|
||||
}
|
||||
|
||||
function encodeURIComponent_RFC3986(str: string): string {
|
||||
function encodeRFC3986(str: string): string {
|
||||
return encodeURIComponent(str)
|
||||
.replace(
|
||||
/[!'()*]/g,
|
||||
@@ -108,41 +109,36 @@ function encodeURIComponent_RFC3986(str: string): string {
|
||||
.replace(/[-_.~]/g, (c) => c);
|
||||
}
|
||||
|
||||
function encodeURI_RFC3986(uri: string): string {
|
||||
if (!uri || uri === "/") return "";
|
||||
function getCanonicalUri(path: string): string {
|
||||
if (!path || path === "/") return "/";
|
||||
|
||||
const segments = uri.split("/");
|
||||
return (
|
||||
"/" +
|
||||
path
|
||||
.split("/")
|
||||
.map((segment) => {
|
||||
if (!segment) return "";
|
||||
if (segment === "invoke-with-response-stream") return segment;
|
||||
|
||||
return segments
|
||||
.map((segment) => {
|
||||
if (!segment) return "";
|
||||
|
||||
if (segment.includes("model/")) {
|
||||
const parts = segment.split(/(model\/)/);
|
||||
return parts
|
||||
.map((part) => {
|
||||
if (part === "model/") return part;
|
||||
if (part.includes(".") || part.includes(":")) {
|
||||
if (segment.includes("model/")) {
|
||||
return segment
|
||||
.split(/(model\/)/)
|
||||
.map((part) => {
|
||||
if (part === "model/") return part;
|
||||
return part
|
||||
.split(/([.:])/g)
|
||||
.map((subpart, i) => {
|
||||
if (i % 2 === 1) return subpart;
|
||||
return encodeURIComponent_RFC3986(subpart);
|
||||
})
|
||||
.map((subpart, i) =>
|
||||
i % 2 === 1 ? subpart : encodeRFC3986(subpart),
|
||||
)
|
||||
.join("");
|
||||
}
|
||||
return encodeURIComponent_RFC3986(part);
|
||||
})
|
||||
.join("");
|
||||
}
|
||||
})
|
||||
.join("");
|
||||
}
|
||||
|
||||
if (segment === "invoke-with-response-stream") {
|
||||
return segment;
|
||||
}
|
||||
|
||||
return encodeURIComponent_RFC3986(segment);
|
||||
})
|
||||
.join("/");
|
||||
return encodeRFC3986(segment);
|
||||
})
|
||||
.join("/")
|
||||
);
|
||||
}
|
||||
|
||||
export async function sign({
|
||||
@@ -153,18 +149,20 @@ export async function sign({
|
||||
secretAccessKey,
|
||||
body,
|
||||
service,
|
||||
headers: customHeaders = {},
|
||||
isStreaming = true,
|
||||
}: SignParams): Promise<Record<string, string>> {
|
||||
try {
|
||||
const endpoint = new URL(url);
|
||||
const canonicalUri = "/" + encodeURI_RFC3986(endpoint.pathname.slice(1));
|
||||
const canonicalUri = getCanonicalUri(endpoint.pathname.slice(1));
|
||||
const canonicalQueryString = endpoint.search.slice(1);
|
||||
|
||||
const now = new Date();
|
||||
const amzDate = now.toISOString().replace(/[:-]|\.\d{3}/g, "");
|
||||
const dateStamp = amzDate.slice(0, 8);
|
||||
|
||||
const payloadHash = SHA256(body).toString(Hex);
|
||||
const bodyString = typeof body === "string" ? body : JSON.stringify(body);
|
||||
const payloadHash = SHA256(bodyString).toString(Hex);
|
||||
|
||||
const headers: Record<string, string> = {
|
||||
accept: isStreaming
|
||||
@@ -174,6 +172,7 @@ export async function sign({
|
||||
host: endpoint.host,
|
||||
"x-amz-content-sha256": payloadHash,
|
||||
"x-amz-date": amzDate,
|
||||
...customHeaders,
|
||||
};
|
||||
|
||||
if (isStreaming) {
|
||||
@@ -237,54 +236,274 @@ export async function sign({
|
||||
}
|
||||
|
||||
// Bedrock utilities
|
||||
function decodeBase64(base64String: string): string {
|
||||
try {
|
||||
return Buffer.from(base64String, "base64").toString("utf-8");
|
||||
} catch (e) {
|
||||
console.error("[Base64 Decode Error]:", e);
|
||||
return "";
|
||||
}
|
||||
}
|
||||
|
||||
export function parseEventData(chunk: Uint8Array): any {
|
||||
const decoder = new TextDecoder();
|
||||
const text = decoder.decode(chunk);
|
||||
const results = [];
|
||||
|
||||
try {
|
||||
// First try to parse as JSON
|
||||
const parsed = JSON.parse(text);
|
||||
// AWS Bedrock wraps the response in a 'body' field
|
||||
|
||||
// Handle bytes field in the response
|
||||
if (parsed.bytes) {
|
||||
const decoded = decodeBase64(parsed.bytes);
|
||||
try {
|
||||
results.push(JSON.parse(decoded));
|
||||
} catch (e) {
|
||||
results.push({ output: decoded });
|
||||
}
|
||||
return results;
|
||||
}
|
||||
|
||||
// Handle body field
|
||||
if (typeof parsed.body === "string") {
|
||||
try {
|
||||
return JSON.parse(parsed.body);
|
||||
results.push(JSON.parse(parsed.body));
|
||||
} catch (e) {
|
||||
return { output: parsed.body };
|
||||
results.push({ output: parsed.body });
|
||||
}
|
||||
return results;
|
||||
}
|
||||
return parsed.body || parsed;
|
||||
|
||||
results.push(parsed.body || parsed);
|
||||
return results;
|
||||
} catch (e) {
|
||||
try {
|
||||
// Handle base64 encoded responses
|
||||
const base64Match = text.match(/:"([A-Za-z0-9+/=]+)"/);
|
||||
if (base64Match) {
|
||||
const decoded = Buffer.from(base64Match[1], "base64").toString("utf-8");
|
||||
// Handle event-stream format
|
||||
const eventRegex = /:event-type[^\{]+({.*?})/g;
|
||||
let match;
|
||||
while ((match = eventRegex.exec(text)) !== null) {
|
||||
try {
|
||||
return JSON.parse(decoded);
|
||||
const eventData = match[1];
|
||||
const parsed = JSON.parse(eventData);
|
||||
if (parsed.bytes) {
|
||||
const decoded = decodeBase64(parsed.bytes);
|
||||
try {
|
||||
results.push(JSON.parse(decoded));
|
||||
} catch (e) {
|
||||
results.push({ output: decoded });
|
||||
}
|
||||
} else {
|
||||
results.push(parsed);
|
||||
}
|
||||
} catch (e) {
|
||||
return { output: decoded };
|
||||
results.push({ output: match[1] });
|
||||
}
|
||||
}
|
||||
|
||||
// Handle event-type responses
|
||||
const eventMatch = text.match(/:event-type[^\{]+({.*})/);
|
||||
if (eventMatch) {
|
||||
try {
|
||||
return JSON.parse(eventMatch[1]);
|
||||
} catch (e) {
|
||||
return { output: eventMatch[1] };
|
||||
}
|
||||
if (results.length > 0) {
|
||||
return results;
|
||||
}
|
||||
|
||||
// Handle plain text responses
|
||||
if (text.trim()) {
|
||||
const cleanText = text.replace(/[\x00-\x1F\x7F-\x9F]/g, "");
|
||||
return { output: cleanText };
|
||||
results.push({ output: cleanText.trim() });
|
||||
return results;
|
||||
}
|
||||
} catch (innerError) {
|
||||
console.error("[AWS Parse Error] Inner parsing failed:", innerError);
|
||||
}
|
||||
}
|
||||
return null;
|
||||
return [];
|
||||
}
|
||||
|
||||
export function processMessage(
|
||||
data: any,
|
||||
remainText: string,
|
||||
runTools: any[],
|
||||
index: number,
|
||||
): { remainText: string; index: number } {
|
||||
if (!data) return { remainText, index };
|
||||
|
||||
try {
|
||||
// Handle message_start event
|
||||
if (data.type === "message_start") {
|
||||
// Keep existing text but mark the start of a new message
|
||||
console.debug("[Message Start] Current text:", remainText);
|
||||
return { remainText, index };
|
||||
}
|
||||
|
||||
// Handle content_block_start event
|
||||
if (data.type === "content_block_start") {
|
||||
if (data.content_block?.type === "tool_use") {
|
||||
index += 1;
|
||||
runTools.push({
|
||||
id: data.content_block.id,
|
||||
type: "function",
|
||||
function: {
|
||||
name: data.content_block.name,
|
||||
arguments: "",
|
||||
},
|
||||
});
|
||||
}
|
||||
return { remainText, index };
|
||||
}
|
||||
|
||||
// Handle content_block_delta event
|
||||
if (data.type === "content_block_delta") {
|
||||
if (data.delta?.type === "input_json_delta" && runTools[index]) {
|
||||
runTools[index].function.arguments += data.delta.partial_json;
|
||||
} else if (data.delta?.type === "text_delta") {
|
||||
const newText = data.delta.text || "";
|
||||
// console.debug("[Text Delta] Adding:", newText);
|
||||
remainText += newText;
|
||||
}
|
||||
return { remainText, index };
|
||||
}
|
||||
|
||||
// Handle tool calls
|
||||
if (data.choices?.[0]?.message?.tool_calls) {
|
||||
for (const toolCall of data.choices[0].message.tool_calls) {
|
||||
index += 1;
|
||||
runTools.push({
|
||||
id: toolCall.id || `tool-${Date.now()}`,
|
||||
type: "function",
|
||||
function: {
|
||||
name: toolCall.function?.name,
|
||||
arguments: toolCall.function?.arguments || "",
|
||||
},
|
||||
});
|
||||
}
|
||||
return { remainText, index };
|
||||
}
|
||||
|
||||
// Handle various response formats
|
||||
let newText = "";
|
||||
if (data.delta?.text) {
|
||||
newText = data.delta.text;
|
||||
} else if (data.choices?.[0]?.message?.content) {
|
||||
newText = data.choices[0].message.content;
|
||||
} else if (data.content?.[0]?.text) {
|
||||
newText = data.content[0].text;
|
||||
} else if (data.generation) {
|
||||
newText = data.generation;
|
||||
} else if (data.outputText) {
|
||||
newText = data.outputText;
|
||||
} else if (data.response) {
|
||||
newText = data.response;
|
||||
} else if (data.output) {
|
||||
newText = data.output;
|
||||
}
|
||||
|
||||
// Only append if we have new text
|
||||
if (newText) {
|
||||
// console.debug("[New Text] Adding:", newText);
|
||||
remainText += newText;
|
||||
}
|
||||
} catch (e) {
|
||||
console.error("[Bedrock Request] parse error", e);
|
||||
}
|
||||
|
||||
return { remainText, index };
|
||||
}
|
||||
|
||||
export function processChunks(
|
||||
chunks: Uint8Array[],
|
||||
pendingChunk: Uint8Array | null,
|
||||
remainText: string,
|
||||
runTools: any[],
|
||||
index: number,
|
||||
): {
|
||||
chunks: Uint8Array[];
|
||||
pendingChunk: Uint8Array | null;
|
||||
remainText: string;
|
||||
index: number;
|
||||
} {
|
||||
let currentText = remainText;
|
||||
let currentIndex = index;
|
||||
|
||||
while (chunks.length > 0) {
|
||||
const chunk = chunks[0];
|
||||
try {
|
||||
// Try to process the chunk
|
||||
const parsedEvents = parseEventData(chunk);
|
||||
if (parsedEvents.length > 0) {
|
||||
// Process each event in the chunk
|
||||
for (const parsed of parsedEvents) {
|
||||
const result = processMessage(
|
||||
parsed,
|
||||
currentText,
|
||||
runTools,
|
||||
currentIndex,
|
||||
);
|
||||
currentText = result.remainText;
|
||||
currentIndex = result.index;
|
||||
}
|
||||
chunks.shift(); // Remove processed chunk
|
||||
|
||||
// If there's a pending chunk, try to process it now
|
||||
if (pendingChunk) {
|
||||
const pendingEvents = parseEventData(pendingChunk);
|
||||
if (pendingEvents.length > 0) {
|
||||
for (const pendingParsed of pendingEvents) {
|
||||
const pendingResult = processMessage(
|
||||
pendingParsed,
|
||||
currentText,
|
||||
runTools,
|
||||
currentIndex,
|
||||
);
|
||||
currentText = pendingResult.remainText;
|
||||
currentIndex = pendingResult.index;
|
||||
}
|
||||
pendingChunk = null;
|
||||
}
|
||||
}
|
||||
} else {
|
||||
// If parsing fails, it might be an incomplete chunk
|
||||
if (pendingChunk) {
|
||||
// Merge with pending chunk
|
||||
const mergedChunk = new Uint8Array(
|
||||
pendingChunk.length + chunk.length,
|
||||
);
|
||||
mergedChunk.set(pendingChunk);
|
||||
mergedChunk.set(chunk, pendingChunk.length);
|
||||
pendingChunk = mergedChunk;
|
||||
} else {
|
||||
pendingChunk = chunk;
|
||||
}
|
||||
chunks.shift();
|
||||
}
|
||||
} catch (e) {
|
||||
console.error("[Chunk Process Error]:", e);
|
||||
chunks.shift(); // Remove error chunk
|
||||
}
|
||||
}
|
||||
|
||||
// Try to process any remaining pending chunk one last time
|
||||
if (pendingChunk) {
|
||||
const finalEvents = parseEventData(pendingChunk);
|
||||
if (finalEvents.length > 0) {
|
||||
for (const finalParsed of finalEvents) {
|
||||
const finalResult = processMessage(
|
||||
finalParsed,
|
||||
currentText,
|
||||
runTools,
|
||||
currentIndex,
|
||||
);
|
||||
currentText = finalResult.remainText;
|
||||
currentIndex = finalResult.index;
|
||||
}
|
||||
pendingChunk = null;
|
||||
}
|
||||
}
|
||||
|
||||
return {
|
||||
chunks,
|
||||
pendingChunk,
|
||||
remainText: currentText,
|
||||
index: currentIndex,
|
||||
};
|
||||
}
|
||||
|
||||
export function getBedrockEndpoint(
|
||||
@@ -309,150 +528,32 @@ export function extractMessage(res: any, modelId: string = ""): string {
|
||||
return "";
|
||||
}
|
||||
|
||||
let message = "";
|
||||
|
||||
// Handle Mistral model response format
|
||||
if (modelId.toLowerCase().includes("mistral")) {
|
||||
if (res.choices?.[0]?.message?.content) {
|
||||
return res.choices[0].message.content;
|
||||
message = res.choices[0].message.content;
|
||||
} else {
|
||||
message = res.output || "";
|
||||
}
|
||||
return res.output || "";
|
||||
}
|
||||
|
||||
// Handle Llama model response format
|
||||
if (modelId.toLowerCase().includes("llama")) {
|
||||
return res?.generation || "";
|
||||
else if (modelId.toLowerCase().includes("llama")) {
|
||||
message = res?.generation || "";
|
||||
}
|
||||
|
||||
// Handle Titan model response format
|
||||
if (modelId.toLowerCase().includes("titan")) {
|
||||
return res?.outputText || "";
|
||||
else if (modelId.toLowerCase().includes("titan")) {
|
||||
message = res?.outputText || "";
|
||||
}
|
||||
|
||||
// Handle Claude and other models
|
||||
return res?.content?.[0]?.text || "";
|
||||
}
|
||||
|
||||
export async function* transformBedrockStream(
|
||||
stream: ReadableStream,
|
||||
modelId: string,
|
||||
) {
|
||||
const reader = stream.getReader();
|
||||
|
||||
try {
|
||||
while (true) {
|
||||
const { done, value } = await reader.read();
|
||||
if (done) break;
|
||||
|
||||
const parsed = parseEventData(value);
|
||||
if (!parsed) continue;
|
||||
|
||||
// console.log("parseEventData=========================");
|
||||
// console.log(parsed);
|
||||
// Handle Claude 3 models
|
||||
if (modelId.startsWith("anthropic.claude")) {
|
||||
if (parsed.type === "message_start") {
|
||||
// Initialize message
|
||||
continue;
|
||||
} else if (parsed.type === "content_block_start") {
|
||||
if (parsed.content_block?.type === "tool_use") {
|
||||
yield `data: ${JSON.stringify(parsed)}\n\n`;
|
||||
}
|
||||
continue;
|
||||
} else if (parsed.type === "content_block_delta") {
|
||||
if (parsed.delta?.type === "text_delta") {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { text: parsed.delta.text },
|
||||
})}\n\n`;
|
||||
} else if (parsed.delta?.type === "input_json_delta") {
|
||||
yield `data: ${JSON.stringify(parsed)}\n\n`;
|
||||
}
|
||||
} else if (parsed.type === "content_block_stop") {
|
||||
yield `data: ${JSON.stringify(parsed)}\n\n`;
|
||||
} else if (
|
||||
parsed.type === "message_delta" &&
|
||||
parsed.delta?.stop_reason
|
||||
) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { stop_reason: parsed.delta.stop_reason },
|
||||
})}\n\n`;
|
||||
}
|
||||
}
|
||||
// Handle Mistral models
|
||||
else if (modelId.toLowerCase().includes("mistral")) {
|
||||
if (parsed.choices?.[0]?.message?.tool_calls) {
|
||||
const toolCalls = parsed.choices[0].message.tool_calls;
|
||||
for (const toolCall of toolCalls) {
|
||||
yield `data: ${JSON.stringify({
|
||||
type: "content_block_start",
|
||||
content_block: {
|
||||
type: "tool_use",
|
||||
id: toolCall.id || `tool-${Date.now()}`,
|
||||
name: toolCall.function?.name,
|
||||
},
|
||||
})}\n\n`;
|
||||
|
||||
if (toolCall.function?.arguments) {
|
||||
yield `data: ${JSON.stringify({
|
||||
type: "content_block_delta",
|
||||
delta: {
|
||||
type: "input_json_delta",
|
||||
partial_json: toolCall.function.arguments,
|
||||
},
|
||||
})}\n\n`;
|
||||
}
|
||||
|
||||
yield `data: ${JSON.stringify({
|
||||
type: "content_block_stop",
|
||||
})}\n\n`;
|
||||
}
|
||||
} else if (parsed.choices?.[0]?.message?.content) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { text: parsed.choices[0].message.content },
|
||||
})}\n\n`;
|
||||
}
|
||||
|
||||
if (parsed.choices?.[0]?.finish_reason) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { stop_reason: parsed.choices[0].finish_reason },
|
||||
})}\n\n`;
|
||||
}
|
||||
}
|
||||
// Handle Llama models
|
||||
else if (modelId.toLowerCase().includes("llama")) {
|
||||
if (parsed.generation) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { text: parsed.generation },
|
||||
})}\n\n`;
|
||||
}
|
||||
if (parsed.stop_reason) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { stop_reason: parsed.stop_reason },
|
||||
})}\n\n`;
|
||||
}
|
||||
}
|
||||
// Handle Titan models
|
||||
else if (modelId.toLowerCase().includes("titan")) {
|
||||
if (parsed.outputText) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { text: parsed.outputText },
|
||||
})}\n\n`;
|
||||
}
|
||||
if (parsed.completionReason) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { stop_reason: parsed.completionReason },
|
||||
})}\n\n`;
|
||||
}
|
||||
}
|
||||
// Handle other models with basic text output
|
||||
else {
|
||||
const text = parsed.response || parsed.output || "";
|
||||
if (text) {
|
||||
yield `data: ${JSON.stringify({
|
||||
delta: { text },
|
||||
})}\n\n`;
|
||||
}
|
||||
}
|
||||
}
|
||||
} finally {
|
||||
reader.releaseLock();
|
||||
else if (res.content?.[0]?.text) {
|
||||
message = res.content[0].text;
|
||||
}
|
||||
// Handle other response formats
|
||||
else {
|
||||
message = res.output || res.response || res.message || "";
|
||||
}
|
||||
|
||||
return message;
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user