feat(ai): move API implementations to src/api with lazy wrappers (phase 2)
Stream implementations move from src/providers/ to src/api/, renamed by API id (anthropic.ts -> anthropic-messages.ts, google.ts -> google-generative-ai.ts, mistral.ts -> mistral-conversations.ts, amazon-bedrock.ts -> bedrock-converse-stream.ts). Every module now exports exactly stream/streamSimple; shared helpers move alongside. New ProviderStreams dispatch contract in types.ts, lazyApi() wrapper in api/lazy.ts, and one .lazy.ts wrapper per API. Bedrock's wrapper keeps the node-only variable-specifier import and setBedrockProviderModule() (now taking ProviderStreams). providers/register-builtins.ts deleted; interim until the compat entrypoint lands, builtin api-registry registration lives in stream.ts and lazy wrappers are exported from the root barrel. Old per-API lazy exports (streamAnthropic, ...) are gone; package.json subpaths retarget to dist/api/.
This commit is contained in:
Mario Zechner
@@ -0,0 +1,292 @@
|
||||
import { AzureOpenAI } from "openai";
|
||||
import type { ResponseCreateParamsStreaming } from "openai/resources/responses/responses.js";
|
||||
import { clampThinkingLevel } from "../models.ts";
|
||||
import type {
|
||||
Api,
|
||||
AssistantMessage,
|
||||
Context,
|
||||
Model,
|
||||
SimpleStreamOptions,
|
||||
StreamFunction,
|
||||
StreamOptions,
|
||||
} from "../types.ts";
|
||||
import { AssistantMessageEventStream } from "../utils/event-stream.ts";
|
||||
import { headersToRecord } from "../utils/headers.ts";
|
||||
import { clampOpenAIPromptCacheKey } from "./openai-prompt-cache.ts";
|
||||
import { convertResponsesMessages, convertResponsesTools, processResponsesStream } from "./openai-responses-shared.ts";
|
||||
import { buildBaseOptions } from "./simple-options.ts";
|
||||
|
||||
const DEFAULT_AZURE_API_VERSION = "v1";
|
||||
const AZURE_TOOL_CALL_PROVIDERS = new Set(["openai", "openai-codex", "opencode", "azure-openai-responses"]);
|
||||
|
||||
function parseDeploymentNameMap(value: string | undefined): Map<string, string> {
|
||||
const map = new Map<string, string>();
|
||||
if (!value) return map;
|
||||
for (const entry of value.split(",")) {
|
||||
const trimmed = entry.trim();
|
||||
if (!trimmed) continue;
|
||||
const [modelId, deploymentName] = trimmed.split("=", 2);
|
||||
if (!modelId || !deploymentName) continue;
|
||||
map.set(modelId.trim(), deploymentName.trim());
|
||||
}
|
||||
return map;
|
||||
}
|
||||
|
||||
function resolveDeploymentName(model: Model<"azure-openai-responses">, options?: AzureOpenAIResponsesOptions): string {
|
||||
if (options?.azureDeploymentName) {
|
||||
return options.azureDeploymentName;
|
||||
}
|
||||
const mappedDeployment = parseDeploymentNameMap(process.env.AZURE_OPENAI_DEPLOYMENT_NAME_MAP).get(model.id);
|
||||
return mappedDeployment || model.id;
|
||||
}
|
||||
|
||||
function formatAzureOpenAIError(error: unknown): string {
|
||||
if (error instanceof Error) {
|
||||
const status = (error as Error & { status?: unknown }).status;
|
||||
const statusCode = typeof status === "number" ? status : undefined;
|
||||
if (statusCode !== undefined) {
|
||||
return `Azure OpenAI API error (${statusCode}): ${error.message}`;
|
||||
}
|
||||
return error.message;
|
||||
}
|
||||
try {
|
||||
return JSON.stringify(error);
|
||||
} catch {
|
||||
return String(error);
|
||||
}
|
||||
}
|
||||
|
||||
// Azure OpenAI Responses-specific options
|
||||
export interface AzureOpenAIResponsesOptions extends StreamOptions {
|
||||
reasoningEffort?: "minimal" | "low" | "medium" | "high" | "xhigh";
|
||||
reasoningSummary?: "auto" | "detailed" | "concise" | null;
|
||||
azureApiVersion?: string;
|
||||
azureResourceName?: string;
|
||||
azureBaseUrl?: string;
|
||||
azureDeploymentName?: string;
|
||||
}
|
||||
|
||||
/**
|
||||
* Generate function for Azure OpenAI Responses API
|
||||
*/
|
||||
export const stream: StreamFunction<"azure-openai-responses", AzureOpenAIResponsesOptions> = (
|
||||
model: Model<"azure-openai-responses">,
|
||||
context: Context,
|
||||
options?: AzureOpenAIResponsesOptions,
|
||||
): AssistantMessageEventStream => {
|
||||
const stream = new AssistantMessageEventStream();
|
||||
|
||||
// Start async processing
|
||||
(async () => {
|
||||
const deploymentName = resolveDeploymentName(model, options);
|
||||
|
||||
const output: AssistantMessage = {
|
||||
role: "assistant",
|
||||
content: [],
|
||||
api: "azure-openai-responses" as Api,
|
||||
provider: model.provider,
|
||||
model: model.id,
|
||||
usage: {
|
||||
input: 0,
|
||||
output: 0,
|
||||
cacheRead: 0,
|
||||
cacheWrite: 0,
|
||||
totalTokens: 0,
|
||||
cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0, total: 0 },
|
||||
},
|
||||
stopReason: "stop",
|
||||
timestamp: Date.now(),
|
||||
};
|
||||
|
||||
try {
|
||||
// Create Azure OpenAI client
|
||||
const apiKey = options?.apiKey;
|
||||
if (!apiKey) {
|
||||
throw new Error(`No API key for provider: ${model.provider}`);
|
||||
}
|
||||
const client = createClient(model, apiKey, options);
|
||||
let params = buildParams(model, context, options, deploymentName);
|
||||
const nextParams = await options?.onPayload?.(params, model);
|
||||
if (nextParams !== undefined) {
|
||||
params = nextParams as ResponseCreateParamsStreaming;
|
||||
}
|
||||
const requestOptions = {
|
||||
...(options?.signal ? { signal: options.signal } : {}),
|
||||
...(options?.timeoutMs !== undefined ? { timeout: options.timeoutMs } : {}),
|
||||
maxRetries: options?.maxRetries ?? 0,
|
||||
};
|
||||
const { data: openaiStream, response } = await client.responses.create(params, requestOptions).withResponse();
|
||||
await options?.onResponse?.({ status: response.status, headers: headersToRecord(response.headers) }, model);
|
||||
stream.push({ type: "start", partial: output });
|
||||
|
||||
await processResponsesStream(openaiStream, output, stream, model);
|
||||
|
||||
if (options?.signal?.aborted) {
|
||||
throw new Error("Request was aborted");
|
||||
}
|
||||
|
||||
if (output.stopReason === "aborted" || output.stopReason === "error") {
|
||||
throw new Error("An unknown error occurred");
|
||||
}
|
||||
|
||||
stream.push({ type: "done", reason: output.stopReason, message: output });
|
||||
stream.end();
|
||||
} catch (error) {
|
||||
for (const block of output.content) {
|
||||
delete (block as { index?: number }).index;
|
||||
// partialJson is only a streaming scratch buffer; never persist it.
|
||||
delete (block as { partialJson?: string }).partialJson;
|
||||
}
|
||||
output.stopReason = options?.signal?.aborted ? "aborted" : "error";
|
||||
output.errorMessage = formatAzureOpenAIError(error);
|
||||
stream.push({ type: "error", reason: output.stopReason, error: output });
|
||||
stream.end();
|
||||
}
|
||||
})();
|
||||
|
||||
return stream;
|
||||
};
|
||||
|
||||
export const streamSimple: StreamFunction<"azure-openai-responses", SimpleStreamOptions> = (
|
||||
model: Model<"azure-openai-responses">,
|
||||
context: Context,
|
||||
options?: SimpleStreamOptions,
|
||||
): AssistantMessageEventStream => {
|
||||
const apiKey = options?.apiKey;
|
||||
if (!apiKey) {
|
||||
throw new Error(`No API key for provider: ${model.provider}`);
|
||||
}
|
||||
|
||||
const base = buildBaseOptions(model, options, apiKey);
|
||||
const clampedReasoning = options?.reasoning ? clampThinkingLevel(model, options.reasoning) : undefined;
|
||||
const reasoningEffort = clampedReasoning === "off" ? undefined : clampedReasoning;
|
||||
|
||||
return stream(model, context, {
|
||||
...base,
|
||||
reasoningEffort,
|
||||
} satisfies AzureOpenAIResponsesOptions);
|
||||
};
|
||||
|
||||
function normalizeAzureBaseUrl(baseUrl: string): string {
|
||||
const trimmed = baseUrl.trim().replace(/\/+$/, "");
|
||||
let url: URL;
|
||||
try {
|
||||
url = new URL(trimmed);
|
||||
} catch {
|
||||
throw new Error(`Invalid Azure OpenAI base URL: ${baseUrl}`);
|
||||
}
|
||||
|
||||
const isAzureHost =
|
||||
url.hostname.endsWith(".openai.azure.com") || url.hostname.endsWith(".cognitiveservices.azure.com");
|
||||
const normalizedPath = url.pathname.replace(/\/+$/, "");
|
||||
|
||||
// Ensure Azure hosts have /openai/v1 as base path so the AzureOpenAI SDK
|
||||
// can append /deployments/<model>/... and ?api-version=v1 correctly.
|
||||
if (isAzureHost && (normalizedPath === "" || normalizedPath === "/" || normalizedPath === "/openai")) {
|
||||
url.pathname = "/openai/v1";
|
||||
url.search = "";
|
||||
}
|
||||
|
||||
return url.toString().replace(/\/+$/, "");
|
||||
}
|
||||
|
||||
function buildDefaultBaseUrl(resourceName: string): string {
|
||||
return `https://${resourceName}.openai.azure.com/openai/v1`;
|
||||
}
|
||||
|
||||
function resolveAzureConfig(
|
||||
model: Model<"azure-openai-responses">,
|
||||
options?: AzureOpenAIResponsesOptions,
|
||||
): { baseUrl: string; apiVersion: string } {
|
||||
const apiVersion = options?.azureApiVersion || process.env.AZURE_OPENAI_API_VERSION || DEFAULT_AZURE_API_VERSION;
|
||||
|
||||
const baseUrl = options?.azureBaseUrl?.trim() || process.env.AZURE_OPENAI_BASE_URL?.trim() || undefined;
|
||||
const resourceName = options?.azureResourceName || process.env.AZURE_OPENAI_RESOURCE_NAME;
|
||||
|
||||
let resolvedBaseUrl = baseUrl;
|
||||
|
||||
if (!resolvedBaseUrl && resourceName) {
|
||||
resolvedBaseUrl = buildDefaultBaseUrl(resourceName);
|
||||
}
|
||||
|
||||
if (!resolvedBaseUrl && model.baseUrl) {
|
||||
resolvedBaseUrl = model.baseUrl;
|
||||
}
|
||||
|
||||
if (!resolvedBaseUrl) {
|
||||
throw new Error(
|
||||
"Azure OpenAI base URL is required. Set AZURE_OPENAI_BASE_URL or AZURE_OPENAI_RESOURCE_NAME, or pass azureBaseUrl, azureResourceName, or model.baseUrl.",
|
||||
);
|
||||
}
|
||||
|
||||
return {
|
||||
baseUrl: normalizeAzureBaseUrl(resolvedBaseUrl),
|
||||
apiVersion,
|
||||
};
|
||||
}
|
||||
|
||||
function createClient(model: Model<"azure-openai-responses">, apiKey: string, options?: AzureOpenAIResponsesOptions) {
|
||||
const headers = { ...model.headers };
|
||||
|
||||
if (options?.headers) {
|
||||
Object.assign(headers, options.headers);
|
||||
}
|
||||
|
||||
const { baseUrl, apiVersion } = resolveAzureConfig(model, options);
|
||||
|
||||
return new AzureOpenAI({
|
||||
apiKey,
|
||||
apiVersion,
|
||||
dangerouslyAllowBrowser: true,
|
||||
defaultHeaders: headers,
|
||||
baseURL: baseUrl,
|
||||
});
|
||||
}
|
||||
|
||||
function buildParams(
|
||||
model: Model<"azure-openai-responses">,
|
||||
context: Context,
|
||||
options: AzureOpenAIResponsesOptions | undefined,
|
||||
deploymentName: string,
|
||||
) {
|
||||
const messages = convertResponsesMessages(model, context, AZURE_TOOL_CALL_PROVIDERS);
|
||||
|
||||
const params: ResponseCreateParamsStreaming = {
|
||||
model: deploymentName,
|
||||
input: messages,
|
||||
stream: true,
|
||||
prompt_cache_key: clampOpenAIPromptCacheKey(options?.sessionId),
|
||||
store: false,
|
||||
};
|
||||
|
||||
if (options?.maxTokens) {
|
||||
params.max_output_tokens = options?.maxTokens;
|
||||
}
|
||||
|
||||
if (options?.temperature !== undefined) {
|
||||
params.temperature = options?.temperature;
|
||||
}
|
||||
|
||||
if (context.tools && context.tools.length > 0) {
|
||||
params.tools = convertResponsesTools(context.tools);
|
||||
}
|
||||
|
||||
if (model.reasoning) {
|
||||
if (options?.reasoningEffort || options?.reasoningSummary) {
|
||||
const effort = options?.reasoningEffort
|
||||
? (model.thinkingLevelMap?.[options.reasoningEffort] ?? options.reasoningEffort)
|
||||
: "medium";
|
||||
params.reasoning = {
|
||||
effort: effort as NonNullable<typeof params.reasoning>["effort"],
|
||||
summary: options?.reasoningSummary || "auto",
|
||||
};
|
||||
params.include = ["reasoning.encrypted_content"];
|
||||
} else if (model.thinkingLevelMap?.off !== null) {
|
||||
params.reasoning = {
|
||||
effort: (model.thinkingLevelMap?.off ?? "none") as NonNullable<typeof params.reasoning>["effort"],
|
||||
};
|
||||
}
|
||||
}
|
||||
|
||||
return params;
|
||||
}
|
||||
Reference in New Issue
Block a user