Release 0.4.13 (#1016)

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

.github/workflows/lint_on_push_or_pull.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

.github/workflows/preview.yml

@@ -0,0 +1,28 @@
+name: Publish Preview
+on: [pull_request]
+
+jobs:
+  pre_release:
+    name: Pre Release
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout Repo
+        uses: actions/checkout@v4
+
+      - uses: pnpm/action-setup@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: ".nvmrc"
+          cache: "pnpm"
+
+      - name: Install dependencies
+        run: pnpm install
+
+      - name: Build
+        run: pnpm run build
+
+      - name: Pre Release
+        run: pnpx pkg-pr-new publish ./packages/*

.github/workflows/publish_github.yml

@@ -12,7 +12,7 @@ jobs:
       - name: Checkout Repo
         uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4

.github/workflows/release.yml

@@ -15,7 +15,7 @@ jobs:
       - name: Checkout Repo
         uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4

.github/workflows/test.yml

@@ -23,7 +23,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4
@@ -45,7 +45,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -60,7 +60,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -97,7 +97,7 @@ jobs:
     name: Build LlamaIndex Example (${{ matrix.packages }})
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -116,7 +116,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

apps/docs/CHANGELOG.md

@@ -1,5 +1,27 @@
 # docs
 
+## 0.0.39
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.38
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.37
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.0.36
 
 ### Patch Changes

apps/docs/docs/getting_started/concepts.md

@@ -62,7 +62,7 @@ These building blocks can be customized to reflect ranking preferences, as well
 
 [**Retrievers**](../modules/retriever.md):
 A retriever defines how to efficiently retrieve relevant context from a knowledge base (i.e. index) when given a query.
-The specific retrieval logic differs for difference indices, the most popular being dense retrieval against a vector index.
+The specific retrieval logic differs for different indices, the most popular being dense retrieval against a vector index.
 
 [**Response Synthesizers**](../modules/response_synthesizer.md):
 A response synthesizer generates a response from an LLM, using a user query and a given set of retrieved text chunks.

apps/docs/docs/modules/data_loaders/llama_parse/index.mdx

@@ -44,6 +44,8 @@ They can be divided into two groups.
 - `pageSeperator?` Optional. The page seperator to use. Defaults is `\\n---\\n`.
 - `gpt4oMode` set to true to use GPT-4o to extract content. Default is `false`.
 - `gpt4oApiKey?` Optional. Set the GPT-4o API key. Lowers the cost of parsing by using your own API key. Your OpenAI account will be charged. Can also be set in the environment variable `LLAMA_CLOUD_GPT4O_API_KEY`.
+- `boundingBox?` Optional. Specify an area of the document to parse. Expects the bounding box margins as a string in clockwise order, e.g. `boundingBox = "0.1,0,0,0"` to not parse the top 10% of the document.
+- `targetPages?` Optional. Specify which pages to parse by specifying them as a comma-seperated list. First page is `0`.
 - `numWorkers` as in the python version, is set in `SimpleDirectoryReader`. Default is 1.
 
 ### LlamaParse with SimpleDirectoryReader

apps/docs/docs/modules/data_loaders/llama_parse/json_mode.mdx

@@ -8,7 +8,7 @@ In JSON mode, LlamaParse will return a data structure representing the parsed ob
 
 ## Usage
 
-For Json mode, you need to use `loadJson`. The `resultType` is automatically set with this method. Currently it can't be used with `SimpleDirectoryReader`.
+For Json mode, you need to use `loadJson`. The `resultType` is automatically set with this method.
 More information about indexing the results on the next page.
 
 ```ts

apps/docs/docs/modules/llms/available_llms/bedrock.md

@@ -15,7 +15,7 @@ Settings.llm = new Bedrock({
 });
 ```
 
-Currently only supports Anthropic models:
+Currently only supports Anthropic and Meta models:
 
 ```ts
 ANTHROPIC_CLAUDE_INSTANT_1 = "anthropic.claude-instant-v1";
@@ -25,6 +25,10 @@ ANTHROPIC_CLAUDE_3_SONNET = "anthropic.claude-3-sonnet-20240229-v1:0";
 ANTHROPIC_CLAUDE_3_HAIKU = "anthropic.claude-3-haiku-20240307-v1:0";
 ANTHROPIC_CLAUDE_3_OPUS = "anthropic.claude-3-opus-20240229-v1:0"; // available on us-west-2
 ANTHROPIC_CLAUDE_3_5_SONNET = "anthropic.claude-3-5-sonnet-20240620-v1:0";
+META_LLAMA2_13B_CHAT = "meta.llama2-13b-chat-v1";
+META_LLAMA2_70B_CHAT = "meta.llama2-70b-chat-v1";
+META_LLAMA3_8B_INSTRUCT = "meta.llama3-8b-instruct-v1:0";
+META_LLAMA3_70B_INSTRUCT = "meta.llama3-70b-instruct-v1:0";
 ```
 
 Sonnet, Haiku and Opus are multimodal, image_url only supports base64 data url format, e.g. `data:image/jpeg;base64,SGVsbG8sIFdvcmxkIQ==`

apps/docs/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docs",
-  "version": "0.0.36",
+  "version": "0.0.39",
   "private": true,
   "scripts": {
     "docusaurus": "docusaurus",

packages/autotool/examples/02_nextjs/CHANGELOG.md

@@ -1,5 +1,30 @@
 # @llamaindex/autotool-02-next-example
 
+## 0.1.23
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.22
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.21
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+  - @llamaindex/autotool@1.0.0
+
 ## 0.1.20
 
 ### Patch Changes

packages/autotool/examples/02_nextjs/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool-02-next-example",
   "private": true,
-  "version": "0.1.20",
+  "version": "0.1.23",
   "scripts": {
     "dev": "next dev",
     "build": "next build",

packages/autotool/package.json

@@ -51,7 +51,7 @@
     "unplugin": "^1.10.1"
   },
   "peerDependencies": {
-    "llamaindex": "^0.4.10",
+    "llamaindex": "^0.4.13",
     "openai": "^4",
     "typescript": "^4"
   },

packages/cloud/CHANGELOG.md

@@ -1,5 +1,11 @@
 # @llamaindex/cloud
 
+## 0.1.2
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+
 ## 0.1.1
 
 ### Patch Changes

packages/cloud/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloud",
-  "version": "0.1.1",
+  "version": "0.1.2",
   "type": "module",
   "license": "MIT",
   "scripts": {

packages/community/CHANGELOG.md

@@ -1,5 +1,28 @@
 # @llamaindex/community
 
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.16
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.15
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.0.14
 
 ### Patch Changes

packages/community/README.md

@@ -5,6 +5,7 @@
 ## Current Features:
 
 - Bedrock support for the Anthropic Claude Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
+- Bedrock support for the Meta LLama 2 and 3 Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
 
 ## LICENSE
 

packages/community/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/community",
   "description": "Community package for LlamaIndexTS",
-  "version": "0.0.14",
+  "version": "0.0.17",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/community/src/llm/bedrock/base.ts

@@ -1,53 +1,34 @@
 import {
   BedrockRuntimeClient,
+  type BedrockRuntimeClientConfig,
   InvokeModelCommand,
   InvokeModelWithResponseStreamCommand,
-  ResponseStream,
-  type BedrockRuntimeClientConfig,
-  type InvokeModelCommandInput,
-  type InvokeModelWithResponseStreamCommandInput,
 } from "@aws-sdk/client-bedrock-runtime";
 import type {
-  BaseTool,
   ChatMessage,
   ChatResponse,
-  ChatResponseChunk,
   CompletionResponse,
   LLMChatParamsNonStreaming,
   LLMChatParamsStreaming,
   LLMCompletionParamsNonStreaming,
   LLMCompletionParamsStreaming,
   LLMMetadata,
-  PartialToolCall,
-  ToolCall,
   ToolCallLLMMessageOptions,
 } from "llamaindex";
-import { ToolCallLLM, streamConverter, wrapLLMEvent } from "llamaindex";
-import type {
-  AnthropicNoneStreamingResponse,
-  AnthropicTextContent,
-  StreamEvent,
-  ToolBlock,
-  ToolChoice,
-} from "./types.js";
+import { streamConverter, ToolCallLLM, wrapLLMEvent } from "llamaindex";
 import {
-  mapBaseToolsToAnthropicTools,
-  mapChatMessagesToAnthropicMessages,
-  mapMessageContentToMessageContentDetails,
-  toUtf8,
-} from "./utils.js";
-
-export type BedrockAdditionalChatOptions = { toolChoice: ToolChoice };
+  type BedrockAdditionalChatOptions,
+  type BedrockChatStreamResponse,
+  Provider,
+} from "./provider";
+import { PROVIDERS } from "./providers";
+import { mapMessageContentToMessageContentDetails } from "./utils.js";
 
 export type BedrockChatParamsStreaming = LLMChatParamsStreaming<
   BedrockAdditionalChatOptions,
   ToolCallLLMMessageOptions
 >;
 
-export type BedrockChatStreamResponse = AsyncIterable<
-  ChatResponseChunk<ToolCallLLMMessageOptions>
->;
-
 export type BedrockChatParamsNonStreaming = LLMChatParamsNonStreaming<
   BedrockAdditionalChatOptions,
   ToolCallLLMMessageOptions
@@ -151,174 +132,6 @@ export const TOOL_CALL_MODELS = [
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET,
 ];
 
-abstract class Provider<ProviderStreamEvent extends {} = {}> {
-  abstract getTextFromResponse(response: Record<string, any>): string;
-
-  abstract getToolsFromResponse<T extends {} = {}>(
-    response: Record<string, any>,
-  ): T[];
-
-  getStreamingEventResponse(
-    response: Record<string, any>,
-  ): ProviderStreamEvent | undefined {
-    return response.chunk?.bytes
-      ? (JSON.parse(toUtf8(response.chunk?.bytes)) as ProviderStreamEvent)
-      : undefined;
-  }
-
-  async *reduceStream(
-    stream: AsyncIterable<ResponseStream>,
-  ): BedrockChatStreamResponse {
-    yield* streamConverter(stream, (response) => {
-      return {
-        delta: this.getTextFromStreamResponse(response),
-        raw: response,
-      };
-    });
-  }
-
-  getTextFromStreamResponse(response: Record<string, any>): string {
-    return this.getTextFromResponse(response);
-  }
-
-  abstract getRequestBody<T extends ChatMessage>(
-    metadata: LLMMetadata,
-    messages: T[],
-    tools?: BaseTool[],
-    options?: BedrockAdditionalChatOptions,
-  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput;
-}
-
-class AnthropicProvider extends Provider<StreamEvent> {
-  getResultFromResponse(
-    response: Record<string, any>,
-  ): AnthropicNoneStreamingResponse {
-    return JSON.parse(toUtf8(response.body));
-  }
-
-  getToolsFromResponse<AnthropicToolContent>(
-    response: Record<string, any>,
-  ): AnthropicToolContent[] {
-    const result = this.getResultFromResponse(response);
-    return result.content
-      .filter((item) => item.type === "tool_use")
-      .map((item) => item as AnthropicToolContent);
-  }
-
-  getTextFromResponse(response: Record<string, any>): string {
-    const result = this.getResultFromResponse(response);
-    return result.content
-      .filter((item) => item.type === "text")
-      .map((item) => (item as AnthropicTextContent).text)
-      .join(" ");
-  }
-
-  getTextFromStreamResponse(response: Record<string, any>): string {
-    const event = this.getStreamingEventResponse(response);
-    if (event?.type === "content_block_delta") {
-      if (event.delta.type === "text_delta") return event.delta.text;
-      if (event.delta.type === "input_json_delta")
-        return event.delta.partial_json;
-    }
-    return "";
-  }
-
-  async *reduceStream(
-    stream: AsyncIterable<ResponseStream>,
-  ): BedrockChatStreamResponse {
-    let collecting = [];
-    let tool: ToolBlock | undefined = undefined;
-    // #TODO this should be broken down into a separate consumer
-    for await (const response of stream) {
-      const event = this.getStreamingEventResponse(response);
-      if (
-        event?.type === "content_block_start" &&
-        event.content_block.type === "tool_use"
-      ) {
-        tool = event.content_block;
-        continue;
-      }
-
-      if (
-        event?.type === "content_block_delta" &&
-        event.delta.type === "input_json_delta"
-      ) {
-        collecting.push(event.delta.partial_json);
-      }
-
-      let options: undefined | ToolCallLLMMessageOptions = undefined;
-      if (tool && collecting.length) {
-        const input = collecting.filter((item) => item).join("");
-        // We have all we need to parse the tool_use json
-        if (event?.type === "content_block_stop") {
-          options = {
-            toolCall: [
-              {
-                id: tool.id,
-                name: tool.name,
-                input: JSON.parse(input),
-              } as ToolCall,
-            ],
-          };
-          // reset the collection/tool
-          collecting = [];
-          tool = undefined;
-        } else {
-          options = {
-            toolCall: [
-              {
-                id: tool.id,
-                name: tool.name,
-                input,
-              } as PartialToolCall,
-            ],
-          };
-        }
-      }
-      const delta = this.getTextFromStreamResponse(response);
-      if (!delta && !options) continue;
-
-      yield {
-        delta,
-        options,
-        raw: response,
-      };
-    }
-  }
-
-  getRequestBody<T extends ChatMessage<ToolCallLLMMessageOptions>>(
-    metadata: LLMMetadata,
-    messages: T[],
-    tools?: BaseTool[],
-    options?: BedrockAdditionalChatOptions,
-  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
-    const extra: Record<string, unknown> = {};
-    if (options?.toolChoice) {
-      extra["tool_choice"] = options?.toolChoice;
-    }
-    const mapped = mapChatMessagesToAnthropicMessages(messages);
-    return {
-      modelId: metadata.model,
-      contentType: "application/json",
-      accept: "application/json",
-      body: JSON.stringify({
-        anthropic_version: "bedrock-2023-05-31",
-        messages: mapped,
-        tools: mapBaseToolsToAnthropicTools(tools),
-        max_tokens: metadata.maxTokens,
-        temperature: metadata.temperature,
-        top_p: metadata.topP,
-        ...extra,
-      }),
-    };
-  }
-}
-
-// Other providers could go here
-const PROVIDERS: { [key: string]: Provider } = {
-  anthropic: new AnthropicProvider(),
-};
-
 const getProvider = (model: string): Provider => {
   const providerName = model.split(".")[0];
   if (!(providerName in PROVIDERS)) {
@@ -373,6 +186,10 @@ export class Bedrock extends ToolCallLLM<BedrockAdditionalChatOptions> {
     this.temperature = temperature ?? DEFAULT_BEDROCK_PARAMS.temperature;
     this.topP = topP ?? DEFAULT_BEDROCK_PARAMS.topP;
     this.client = new BedrockRuntimeClient(params);
+
+    if (!this.supportToolCall) {
+      console.warn(`The model "${this.model}" doesn't support ToolCall`);
+    }
   }
 
   get supportToolCall(): boolean {
@@ -402,10 +219,13 @@ export class Bedrock extends ToolCallLLM<BedrockAdditionalChatOptions> {
     );
     const command = new InvokeModelCommand(input);
     const response = await this.client.send(command);
-    const tools = this.provider.getToolsFromResponse(response);
-    const options: ToolCallLLMMessageOptions = tools.length
-      ? { toolCall: tools }
-      : {};
+    let options: ToolCallLLMMessageOptions = {};
+    if (this.supportToolCall) {
+      const tools = this.provider.getToolsFromResponse(response);
+      if (tools.length) {
+        options = { toolCall: tools };
+      }
+    }
     return {
       raw: response,
       message: {

packages/community/src/llm/bedrock/provider.ts

@@ -0,0 +1,59 @@
+import {
+  type InvokeModelCommandInput,
+  type InvokeModelWithResponseStreamCommandInput,
+  ResponseStream,
+} from "@aws-sdk/client-bedrock-runtime";
+import {
+  type BaseTool,
+  type ChatMessage,
+  type ChatResponseChunk,
+  type LLMMetadata,
+  streamConverter,
+  type ToolCallLLMMessageOptions,
+} from "llamaindex";
+import type { ToolChoice } from "./types";
+import { toUtf8 } from "./utils";
+
+export type BedrockAdditionalChatOptions = { toolChoice: ToolChoice };
+
+export type BedrockChatStreamResponse = AsyncIterable<
+  ChatResponseChunk<ToolCallLLMMessageOptions>
+>;
+
+export abstract class Provider<ProviderStreamEvent extends {} = {}> {
+  abstract getTextFromResponse(response: Record<string, any>): string;
+
+  abstract getToolsFromResponse<T extends {} = {}>(
+    response: Record<string, any>,
+  ): T[];
+
+  getStreamingEventResponse(
+    response: Record<string, any>,
+  ): ProviderStreamEvent | undefined {
+    return response.chunk?.bytes
+      ? (JSON.parse(toUtf8(response.chunk?.bytes)) as ProviderStreamEvent)
+      : undefined;
+  }
+
+  async *reduceStream(
+    stream: AsyncIterable<ResponseStream>,
+  ): BedrockChatStreamResponse {
+    yield* streamConverter(stream, (response) => {
+      return {
+        delta: this.getTextFromStreamResponse(response),
+        raw: response,
+      };
+    });
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    return this.getTextFromResponse(response);
+  }
+
+  abstract getRequestBody<T extends ChatMessage>(
+    metadata: LLMMetadata,
+    messages: T[],
+    tools?: BaseTool[],
+    options?: BedrockAdditionalChatOptions,
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput;
+}

packages/community/src/llm/bedrock/providers/anthropic.ts

@@ -0,0 +1,154 @@
+import {
+  type InvokeModelCommandInput,
+  type InvokeModelWithResponseStreamCommandInput,
+  ResponseStream,
+} from "@aws-sdk/client-bedrock-runtime";
+import type {
+  BaseTool,
+  ChatMessage,
+  LLMMetadata,
+  PartialToolCall,
+  ToolCall,
+  ToolCallLLMMessageOptions,
+} from "llamaindex";
+import {
+  type BedrockAdditionalChatOptions,
+  type BedrockChatStreamResponse,
+  Provider,
+} from "../provider";
+import type {
+  AnthropicNoneStreamingResponse,
+  AnthropicStreamEvent,
+  AnthropicTextContent,
+  ToolBlock,
+} from "../types";
+import {
+  mapBaseToolsToAnthropicTools,
+  mapChatMessagesToAnthropicMessages,
+  toUtf8,
+} from "../utils";
+
+export class AnthropicProvider extends Provider<AnthropicStreamEvent> {
+  getResultFromResponse(
+    response: Record<string, any>,
+  ): AnthropicNoneStreamingResponse {
+    return JSON.parse(toUtf8(response.body));
+  }
+
+  getToolsFromResponse<AnthropicToolContent>(
+    response: Record<string, any>,
+  ): AnthropicToolContent[] {
+    const result = this.getResultFromResponse(response);
+    return result.content
+      .filter((item) => item.type === "tool_use")
+      .map((item) => item as AnthropicToolContent);
+  }
+
+  getTextFromResponse(response: Record<string, any>): string {
+    const result = this.getResultFromResponse(response);
+    return result.content
+      .filter((item) => item.type === "text")
+      .map((item) => (item as AnthropicTextContent).text)
+      .join(" ");
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    const event = this.getStreamingEventResponse(response);
+    if (event?.type === "content_block_delta") {
+      if (event.delta.type === "text_delta") return event.delta.text;
+      if (event.delta.type === "input_json_delta")
+        return event.delta.partial_json;
+    }
+    return "";
+  }
+
+  async *reduceStream(
+    stream: AsyncIterable<ResponseStream>,
+  ): BedrockChatStreamResponse {
+    let collecting = [];
+    let tool: ToolBlock | undefined = undefined;
+    // #TODO this should be broken down into a separate consumer
+    for await (const response of stream) {
+      const event = this.getStreamingEventResponse(response);
+      if (
+        event?.type === "content_block_start" &&
+        event.content_block.type === "tool_use"
+      ) {
+        tool = event.content_block;
+        continue;
+      }
+
+      if (
+        event?.type === "content_block_delta" &&
+        event.delta.type === "input_json_delta"
+      ) {
+        collecting.push(event.delta.partial_json);
+      }
+
+      let options: undefined | ToolCallLLMMessageOptions = undefined;
+      if (tool && collecting.length) {
+        const input = collecting.filter((item) => item).join("");
+        // We have all we need to parse the tool_use json
+        if (event?.type === "content_block_stop") {
+          options = {
+            toolCall: [
+              {
+                id: tool.id,
+                name: tool.name,
+                input: JSON.parse(input),
+              } as ToolCall,
+            ],
+          };
+          // reset the collection/tool
+          collecting = [];
+          tool = undefined;
+        } else {
+          options = {
+            toolCall: [
+              {
+                id: tool.id,
+                name: tool.name,
+                input,
+              } as PartialToolCall,
+            ],
+          };
+        }
+      }
+      const delta = this.getTextFromStreamResponse(response);
+      if (!delta && !options) continue;
+
+      yield {
+        delta,
+        options,
+        raw: response,
+      };
+    }
+  }
+
+  getRequestBody<T extends ChatMessage<ToolCallLLMMessageOptions>>(
+    metadata: LLMMetadata,
+    messages: T[],
+    tools?: BaseTool[],
+    options?: BedrockAdditionalChatOptions,
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
+    const extra: Record<string, unknown> = {};
+    if (options?.toolChoice) {
+      extra["tool_choice"] = options?.toolChoice;
+    }
+    const mapped = mapChatMessagesToAnthropicMessages(messages);
+    return {
+      modelId: metadata.model,
+      contentType: "application/json",
+      accept: "application/json",
+      body: JSON.stringify({
+        anthropic_version: "bedrock-2023-05-31",
+        messages: mapped,
+        tools: mapBaseToolsToAnthropicTools(tools),
+        max_tokens: metadata.maxTokens,
+        temperature: metadata.temperature,
+        top_p: metadata.topP,
+        ...extra,
+      }),
+    };
+  }
+}

packages/community/src/llm/bedrock/providers/index.ts

@@ -0,0 +1,9 @@
+import { Provider } from "../provider";
+import { AnthropicProvider } from "./anthropic";
+import { MetaProvider } from "./meta";
+
+// Other providers should go here
+export const PROVIDERS: { [key: string]: Provider } = {
+  anthropic: new AnthropicProvider(),
+  meta: new MetaProvider(),
+};

packages/community/src/llm/bedrock/providers/meta.ts

@@ -0,0 +1,69 @@
+import type {
+  InvokeModelCommandInput,
+  InvokeModelWithResponseStreamCommandInput,
+} from "@aws-sdk/client-bedrock-runtime";
+import type { ChatMessage, LLMMetadata } from "llamaindex";
+import type { MetaNoneStreamingResponse, MetaStreamEvent } from "../types";
+import {
+  mapChatMessagesToMetaLlama2Messages,
+  mapChatMessagesToMetaLlama3Messages,
+  toUtf8,
+} from "../utils";
+
+import { Provider } from "../provider";
+
+export class MetaProvider extends Provider<MetaStreamEvent> {
+  constructor() {
+    super();
+  }
+
+  getResultFromResponse(
+    response: Record<string, any>,
+  ): MetaNoneStreamingResponse {
+    return JSON.parse(toUtf8(response.body));
+  }
+
+  getToolsFromResponse(_response: Record<string, any>): never {
+    throw new Error("Not supported by this provider.");
+  }
+
+  getTextFromResponse(response: Record<string, any>): string {
+    const result = this.getResultFromResponse(response);
+    return result.generation;
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    const event = this.getStreamingEventResponse(response);
+    if (event?.generation) {
+      return event.generation;
+    }
+    return "";
+  }
+
+  getRequestBody<T extends ChatMessage>(
+    metadata: LLMMetadata,
+    messages: T[],
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
+    let promptFunction: (messages: ChatMessage[]) => string;
+
+    if (metadata.model.startsWith("meta.llama3")) {
+      promptFunction = mapChatMessagesToMetaLlama3Messages;
+    } else if (metadata.model.startsWith("meta.llama2")) {
+      promptFunction = mapChatMessagesToMetaLlama2Messages;
+    } else {
+      throw new Error(`Meta model ${metadata.model} is not supported`);
+    }
+
+    return {
+      modelId: metadata.model,
+      contentType: "application/json",
+      accept: "application/json",
+      body: JSON.stringify({
+        prompt: promptFunction(messages),
+        max_gen_len: metadata.maxTokens,
+        temperature: metadata.temperature,
+        top_p: metadata.topP,
+      }),
+    };
+  }
+}

packages/community/src/llm/bedrock/types.ts

@@ -79,7 +79,7 @@ export type ToolChoice =
   | { type: "auto" }
   | { type: "tool"; name: string };
 
-export type StreamEvent =
+export type AnthropicStreamEvent =
   | { type: "message_start"; message: Message }
   | ContentBlockStart
   | ContentBlockDelta
@@ -93,6 +93,8 @@ export type AnthropicContent =
   | AnthropicToolContent
   | AnthropicToolResultContent;
 
+export type MetaTextContent = string;
+
 export type AnthropicTextContent = {
   type: "text";
   text: string;
@@ -133,6 +135,11 @@ export type AnthropicMessage = {
   content: AnthropicContent[];
 };
 
+export type MetaMessage = {
+  role: "user" | "assistant" | "system";
+  content: MetaTextContent;
+};
+
 export type AnthropicNoneStreamingResponse = {
   id: string;
   type: "message";
@@ -143,3 +150,16 @@ export type AnthropicNoneStreamingResponse = {
   stop_sequence?: string;
   usage: { input_tokens: number; output_tokens: number };
 };
+
+type MetaResponse = {
+  generation: string;
+  prompt_token_count: number;
+  generation_token_count: number;
+  stop_reason: "stop" | "length";
+};
+
+export type MetaStreamEvent = MetaResponse & {
+  "amazon-bedrock-invocationMetrics": InvocationMetrics;
+};
+
+export type MetaNoneStreamingResponse = MetaResponse;

packages/community/src/llm/bedrock/utils.ts

@@ -4,6 +4,7 @@ import type {
   JSONObject,
   MessageContent,
   MessageContentDetail,
+  MessageContentTextDetail,
   ToolCallLLMMessageOptions,
   ToolMetadata,
 } from "llamaindex";
@@ -13,6 +14,7 @@ import type {
   AnthropicMediaTypes,
   AnthropicMessage,
   AnthropicTextContent,
+  MetaMessage,
 } from "./types.js";
 
 const ACCEPTED_IMAGE_MIME_TYPES = [
@@ -148,6 +150,85 @@ export const mapChatMessagesToAnthropicMessages = <
   return mergeNeighboringSameRoleMessages(mapped);
 };
 
+export const mapChatMessagesToMetaMessages = <T extends ChatMessage>(
+  messages: T[],
+): MetaMessage[] => {
+  return messages.map((msg) => {
+    let content: string;
+    if (typeof msg.content === "string") {
+      content = msg.content;
+    } else {
+      content = (msg.content[0] as MessageContentTextDetail).text;
+    }
+    return {
+      role:
+        msg.role === "assistant"
+          ? "assistant"
+          : msg.role === "user"
+            ? "user"
+            : "system",
+      content,
+    };
+  });
+};
+
+/**
+ * Documentation at https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-3
+ */
+export const mapChatMessagesToMetaLlama3Messages = <T extends ChatMessage>(
+  messages: T[],
+): string => {
+  const mapped = mapChatMessagesToMetaMessages(messages).map((message) => {
+    const text = message.content;
+    return `<|start_header_id|>${message.role}<|end_header_id|>\n${text}\n<|eot_id|>\n`;
+  });
+  return (
+    "<|begin_of_text|>" +
+    mapped.join("\n") +
+    "\n<|start_header_id|>assistant<|end_header_id|>\n"
+  );
+};
+
+/**
+ * Documentation at https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-2
+ */
+export const mapChatMessagesToMetaLlama2Messages = <T extends ChatMessage>(
+  messages: T[],
+): string => {
+  const mapped = mapChatMessagesToMetaMessages(messages);
+  let output = "<s>";
+  let insideInst = false;
+  let needsStartAgain = false;
+  for (const message of mapped) {
+    if (needsStartAgain) {
+      output += "<s>";
+      needsStartAgain = false;
+    }
+    const text = message.content;
+    if (message.role === "system") {
+      if (!insideInst) {
+        output += "[INST] ";
+        insideInst = true;
+      }
+      output += `<<SYS>>\n${text}\n<</SYS>>\n`;
+    } else if (message.role === "user") {
+      output += text;
+      if (insideInst) {
+        output += " [/INST]";
+        insideInst = false;
+      }
+    } else if (message.role === "assistant") {
+      if (insideInst) {
+        output += " [/INST]";
+        insideInst = false;
+      }
+      output += ` ${text} </s>\n`;
+      needsStartAgain = true;
+    }
+  }
+  return output;
+};
+
 export const mapTextContent = (text: string): AnthropicTextContent => {
   return { type: "text", text };
 };

packages/core/CHANGELOG.md

@@ -1,5 +1,13 @@
 # @llamaindex/core
 
+## 0.0.3
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - @llamaindex/env@0.1.8
+
 ## 0.0.2
 
 ### Patch Changes

packages/core/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/core",
   "type": "module",
-  "version": "0.0.2",
+  "version": "0.0.3",
   "description": "LlamaIndex Core Module",
   "exports": {
     "./llms": {

packages/env/CHANGELOG.md

@@ -1,5 +1,11 @@
 # @llamaindex/env
 
+## 0.1.8
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+
 ## 0.1.7
 
 ### Patch Changes

packages/env/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/env",
   "description": "environment wrapper, supports all JS environment including node, deno, bun, edge runtime, and cloudflare worker",
-  "version": "0.1.7",
+  "version": "0.1.8",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/experimental/CHANGELOG.md

@@ -1,5 +1,27 @@
 # @llamaindex/experimental
 
+## 0.0.48
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.47
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.46
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.0.45
 
 ### Patch Changes

packages/experimental/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/experimental",
   "description": "Experimental package for LlamaIndexTS",
-  "version": "0.0.45",
+  "version": "0.0.48",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/llamaindex/CHANGELOG.md

@@ -1,5 +1,28 @@
 # llamaindex
 
+## 0.4.13
+
+### Patch Changes
+
+- e8f8bea: feat: add boundingBox and targetPages to LlamaParseReader
+- 304484b: feat: add ignoreErrors flag to LlamaParseReader
+
+## 0.4.12
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - @llamaindex/cloud@0.1.2
+  - @llamaindex/core@0.0.3
+  - @llamaindex/env@0.1.8
+
+## 0.4.11
+
+### Patch Changes
+
+- 8bf5b4a: fix: llama parse input spreadsheet
+
 ## 0.4.10
 
 ### Patch Changes

packages/llamaindex/e2e/examples/cloudflare-worker-agent/CHANGELOG.md

@@ -1,5 +1,27 @@
 # @llamaindex/cloudflare-worker-agent-test
 
+## 0.0.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.30
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.0.29
 
 ### Patch Changes

packages/llamaindex/e2e/examples/cloudflare-worker-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloudflare-worker-agent-test",
-  "version": "0.0.29",
+  "version": "0.0.32",
   "type": "module",
   "private": true,
   "scripts": {

packages/llamaindex/e2e/examples/nextjs-agent/CHANGELOG.md

@@ -1,5 +1,27 @@
 # @llamaindex/next-agent-test
 
+## 0.1.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.1.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.1.30
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.1.29
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-agent-test",
-  "version": "0.1.29",
+  "version": "0.1.32",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-edge-runtime/CHANGELOG.md

@@ -1,5 +1,27 @@
 # test-edge-runtime
 
+## 0.1.31
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.1.30
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.1.29
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.1.28
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-edge-runtime/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/nextjs-edge-runtime-test",
-  "version": "0.1.28",
+  "version": "0.1.31",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-node-runtime/CHANGELOG.md

@@ -1,5 +1,27 @@
 # @llamaindex/next-node-runtime
 
+## 0.0.13
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.12
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.11
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.0.10
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-node-runtime/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-node-runtime-test",
-  "version": "0.0.10",
+  "version": "0.0.13",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/waku-query-engine/CHANGELOG.md

@@ -1,5 +1,27 @@
 # @llamaindex/waku-query-engine-test
 
+## 0.0.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.30
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
 ## 0.0.29
 
 ### Patch Changes

packages/llamaindex/e2e/examples/waku-query-engine/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/waku-query-engine-test",
-  "version": "0.0.29",
+  "version": "0.0.32",
   "type": "module",
   "private": true,
   "scripts": {

packages/llamaindex/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llamaindex",
-  "version": "0.4.10",
+  "version": "0.4.13",
   "license": "MIT",
   "type": "module",
   "keywords": [

packages/llamaindex/src/cloud/LlamaCloudRetriever.ts

@@ -8,6 +8,7 @@ import type { NodeWithScore } from "@llamaindex/core/schema";
 import { jsonToNode, ObjectType } from "@llamaindex/core/schema";
 import type { BaseRetriever, RetrieveParams } from "../Retriever.js";
 import { wrapEventCaller } from "../internal/context/EventCaller.js";
+import { getCallbackManager } from "../internal/settings/CallbackManager.js";
 import { extractText } from "../llm/utils.js";
 import type { ClientParams, CloudConstructorParams } from "./constants.js";
 import { DEFAULT_PROJECT_NAME } from "./constants.js";
@@ -28,9 +29,14 @@ export class LlamaCloudRetriever implements BaseRetriever {
     nodes: TextNodeWithScore[],
   ): NodeWithScore[] {
     return nodes.map((node: TextNodeWithScore) => {
+      const textNode = jsonToNode(node.node, ObjectType.TEXT);
+      textNode.metadata = {
+        ...textNode.metadata,
+        ...node.node.extra_info, // append LlamaCloud extra_info to node metadata (file_name, pipeline_id, etc.)
+      };
       return {
         // Currently LlamaCloud only supports text nodes
-        node: jsonToNode(node.node, ObjectType.TEXT),
+        node: textNode,
         score: node.score,
       };
     });
@@ -83,6 +89,15 @@ export class LlamaCloudRetriever implements BaseRetriever {
         },
       });
 
-    return this.resultNodesToNodeWithScore(results.retrieval_nodes);
+    const nodesWithScores = this.resultNodesToNodeWithScore(
+      results.retrieval_nodes,
+    );
+    getCallbackManager().dispatchEvent("retrieve-end", {
+      payload: {
+        query,
+        nodes: nodesWithScores,
+      },
+    });
+    return nodesWithScores;
   }
 }

packages/llamaindex/src/readers/LlamaParseReader.ts

@@ -1,105 +1,100 @@
 import { Document } from "@llamaindex/core/schema";
 import { fs, getEnv } from "@llamaindex/env";
-import { filetypemime } from "magic-bytes.js";
+import { filetypeinfo } from "magic-bytes.js";
 import { FileReader, type Language, type ResultType } from "./type.js";
 
-const SupportedFiles: { [key: string]: string } = {
-  ".pdf": "application/pdf",
-  // Documents and Presentations
-  ".602": "application/x-t602",
-  ".abw": "application/x-abiword",
-  ".cgm": "image/cgm",
-  ".cwk": "application/x-cwk",
-  ".doc": "application/msword",
-  ".docx":
-    "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
-  ".docm": "application/vnd.ms-word.document.macroEnabled.12",
-  ".dot": "application/msword",
-  ".dotm": "application/vnd.ms-word.template.macroEnabled.12",
-  ".dotx":
-    "application/vnd.openxmlformats-officedocument.wordprocessingml.template",
-  ".hwp": "application/x-hwp",
-  ".key": "application/x-iwork-keynote-sffkey",
-  ".lwp": "application/vnd.lotus-wordpro",
-  ".mw": "application/macwriteii",
-  ".mcw": "application/macwriteii",
-  ".pages": "application/x-iwork-pages-sffpages",
-  ".pbd": "application/x-pagemaker",
-  ".ppt": "application/vnd.ms-powerpoint",
-  ".pptm": "application/vnd.ms-powerpoint.presentation.macroEnabled.12",
-  ".pptx":
-    "application/vnd.openxmlformats-officedocument.presentationml.presentation",
-  ".pot": "application/vnd.ms-powerpoint",
-  ".potm": "application/vnd.ms-powerpoint.template.macroEnabled.12",
-  ".potx":
-    "application/vnd.openxmlformats-officedocument.presentationml.template",
-  ".rtf": "application/rtf",
-  ".sda": "application/vnd.stardivision.draw",
-  ".sdd": "application/vnd.stardivision.impress",
-  ".sdp": "application/sdp",
-  ".sdw": "application/vnd.stardivision.writer",
-  ".sgl": "application/vnd.stardivision.writer",
-  ".sti": "application/vnd.sun.xml.impress.template",
-  ".sxi": "application/vnd.sun.xml.impress",
-  ".sxw": "application/vnd.sun.xml.writer",
-  ".stw": "application/vnd.sun.xml.writer.template",
-  ".sxg": "application/vnd.sun.xml.writer.global",
-  ".txt": "text/plain",
-  ".uof": "application/vnd.uoml+xml",
-  ".uop": "application/vnd.openofficeorg.presentation",
-  ".uot": "application/x-uo",
-  ".vor": "application/vnd.stardivision.writer",
-  ".wpd": "application/wordperfect",
-  ".wps": "application/vnd.ms-works",
-  ".xml": "application/xml",
-  ".zabw": "application/x-abiword",
-  // Images
-  ".epub": "application/epub+zip",
-  ".jpg": "image/jpeg",
-  ".jpeg": "image/jpeg",
-  ".png": "image/png",
-  ".gif": "image/gif",
-  ".bmp": "image/bmp",
-  ".svg": "image/svg+xml",
-  ".tiff": "image/tiff",
-  ".webp": "image/webp",
-  // Web
-  ".htm": "text/html",
-  ".html": "text/html",
-  // Spreadsheets
-  ".xlsx": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
-  ".xls": "application/vnd.ms-excel",
-  ".xlsm": "application/vnd.ms-excel.sheet.macroEnabled.12",
-  ".xlsb": "application/vnd.ms-excel.sheet.binary.macroEnabled.12",
-  ".xlw": "application/vnd.ms-excel",
-  ".csv": "text/csv",
-  ".dif": "application/x-dif",
-  ".sylk": "text/vnd.sylk",
-  ".slk": "text/vnd.sylk",
-  ".prn": "application/x-prn",
-  ".numbers": "application/x-iwork-numbers-sffnumbers",
-  ".et": "application/vnd.ms-excel",
-  ".ods": "application/vnd.oasis.opendocument.spreadsheet",
-  ".fods": "application/vnd.oasis.opendocument.spreadsheet",
-  ".uos1": "application/vnd.uoml+xml",
-  ".uos2": "application/vnd.uoml+xml",
-  ".dbf": "application/vnd.dbf",
-  ".wk1": "application/vnd.lotus-1-2-3",
-  ".wk2": "application/vnd.lotus-1-2-3",
-  ".wk3": "application/vnd.lotus-1-2-3",
-  ".wk4": "application/vnd.lotus-1-2-3",
-  ".wks": "application/vnd.lotus-1-2-3",
-  ".123": "application/vnd.lotus-1-2-3",
-  ".wq1": "application/x-lotus",
-  ".wq2": "application/x-lotus",
-  ".wb1": "application/x-quattro-pro",
-  ".wb2": "application/x-quattro-pro",
-  ".wb3": "application/x-quattro-pro",
-  ".qpw": "application/x-quattro-pro",
-  ".xlr": "application/vnd.ms-works",
-  ".eth": "application/ethos",
-  ".tsv": "text/tab-separated-values",
-};
+const SUPPORT_FILE_EXT: string[] = [
+  ".pdf",
+  // document and presentations
+  ".602",
+  ".abw",
+  ".cgm",
+  ".cwk",
+  ".doc",
+  ".docx",
+  ".docm",
+  ".dot",
+  ".dotm",
+  ".hwp",
+  ".key",
+  ".lwp",
+  ".mw",
+  ".mcw",
+  ".pages",
+  ".pbd",
+  ".ppt",
+  ".pptm",
+  ".pptx",
+  ".pot",
+  ".potm",
+  ".potx",
+  ".rtf",
+  ".sda",
+  ".sdd",
+  ".sdp",
+  ".sdw",
+  ".sgl",
+  ".sti",
+  ".sxi",
+  ".sxw",
+  ".stw",
+  ".sxg",
+  ".txt",
+  ".uof",
+  ".uop",
+  ".uot",
+  ".vor",
+  ".wpd",
+  ".wps",
+  ".xml",
+  ".zabw",
+  ".epub",
+  // images
+  ".jpg",
+  ".jpeg",
+  ".png",
+  ".gif",
+  ".bmp",
+  ".svg",
+  ".tiff",
+  ".webp",
+  // web
+  ".htm",
+  ".html",
+  // spreadsheets
+  ".xlsx",
+  ".xls",
+  ".xlsm",
+  ".xlsb",
+  ".xlw",
+  ".csv",
+  ".dif",
+  ".sylk",
+  ".slk",
+  ".prn",
+  ".numbers",
+  ".et",
+  ".ods",
+  ".fods",
+  ".uos1",
+  ".uos2",
+  ".dbf",
+  ".wk1",
+  ".wk2",
+  ".wk3",
+  ".wk4",
+  ".wks",
+  ".123",
+  ".wq1",
+  ".wq2",
+  ".wb1",
+  ".wb2",
+  ".wb3",
+  ".qpw",
+  ".xlr",
+  ".eth",
+  ".tsv",
+];
 
 /**
  * Represents a reader for parsing files using the LlamaParse API.
@@ -138,6 +133,12 @@ export class LlamaParseReader extends FileReader {
   gpt4oMode: boolean = false;
   // The API key for the GPT-4o API. Optional, lowers the cost of parsing. Can be set as an env variable: LLAMA_CLOUD_GPT4O_API_KEY.
   gpt4oApiKey?: string;
+  // The bounding box to use to extract text from documents. Describe as a string containing the bounding box margins.
+  boundingBox?: string;
+  // The target pages to extract text from documents. Describe as a comma separated list of page numbers. The first page of the document is page 0
+  targetPages?: string;
+  // Whether or not to ignore and skip errors raised during parsing.
+  ignoreErrors: boolean = true;
   // numWorkers is implemented in SimpleDirectoryReader
 
   constructor(params: Partial<LlamaParseReader> = {}) {
@@ -165,7 +166,7 @@ export class LlamaParseReader extends FileReader {
     fileName?: string,
   ): Promise<string> {
     // Load data, set the mime type
-    const { mimeType, extension } = await this.getMimeType(data);
+    const { mime, extension } = await LlamaParseReader.getMimeType(data);
 
     if (this.verbose) {
       const name = fileName ? fileName : extension;
@@ -173,7 +174,7 @@ export class LlamaParseReader extends FileReader {
     }
 
     const body = new FormData();
-    body.set("file", new Blob([data], { type: mimeType }), fileName);
+    body.set("file", new Blob([data], { type: mime }), fileName);
 
     const LlamaParseBodyParams = {
       language: this.language,
@@ -186,6 +187,8 @@ export class LlamaParseReader extends FileReader {
       page_seperator: this.pageSeperator,
       gpt4o_mode: this.gpt4oMode?.toString(),
       gpt4o_api_key: this.gpt4oApiKey,
+      bounding_box: this.boundingBox,
+      target_pages: this.targetPages,
     };
 
     // Appends body with any defined LlamaParseBodyParams
@@ -283,19 +286,29 @@ export class LlamaParseReader extends FileReader {
     fileContent: Uint8Array,
     fileName?: string,
   ): Promise<Document[]> {
-    // Creates a job for the file
-    const jobId = await this.createJob(fileContent, fileName);
-    if (this.verbose) {
-      console.log(`Started parsing the file under job id ${jobId}`);
-    }
+    let jobId;
+    try {
+      // Creates a job for the file
+      jobId = await this.createJob(fileContent, fileName);
+      if (this.verbose) {
+        console.log(`Started parsing the file under job id ${jobId}`);
+      }
 
-    // Return results as Document objects
-    const resultJson = await this.getJobResult(jobId, this.resultType);
-    return [
-      new Document({
-        text: resultJson[this.resultType],
-      }),
-    ];
+      // Return results as Document objects
+      const resultJson = await this.getJobResult(jobId, this.resultType);
+      return [
+        new Document({
+          text: resultJson[this.resultType],
+        }),
+      ];
+    } catch (e) {
+      console.error(`Error while parsing file under job id ${jobId}`, e);
+      if (this.ignoreErrors) {
+        return [];
+      } else {
+        throw e;
+      }
+    }
   }
   /**
    * Loads data from a file and returns an array of JSON objects.
@@ -305,18 +318,28 @@ export class LlamaParseReader extends FileReader {
    * @return {Promise<Record<string, any>[]>} A Promise that resolves to an array of JSON objects.
    */
   async loadJson(file: string): Promise<Record<string, any>[]> {
-    const data = await fs.readFile(file);
-    // Creates a job for the file
-    const jobId = await this.createJob(data);
-    if (this.verbose) {
-      console.log(`Started parsing the file under job id ${jobId}`);
-    }
+    let jobId;
+    try {
+      const data = await fs.readFile(file);
+      // Creates a job for the file
+      jobId = await this.createJob(data);
+      if (this.verbose) {
+        console.log(`Started parsing the file under job id ${jobId}`);
+      }
 
-    // Return results as an array of JSON objects (same format as Python version of the reader)
-    const resultJson = await this.getJobResult(jobId, "json");
-    resultJson.job_id = jobId;
-    resultJson.file_path = file;
-    return [resultJson];
+      // Return results as an array of JSON objects (same format as Python version of the reader)
+      const resultJson = await this.getJobResult(jobId, "json");
+      resultJson.job_id = jobId;
+      resultJson.file_path = file;
+      return [resultJson];
+    } catch (e) {
+      console.error(`Error while parsing the file under job id ${jobId}`, e);
+      if (this.ignoreErrors) {
+        return [];
+      } else {
+        throw e;
+      }
+    }
   }
 
   /**
@@ -331,66 +354,100 @@ export class LlamaParseReader extends FileReader {
     jsonResult: Record<string, any>[],
     downloadPath: string,
   ): Promise<Record<string, any>[]> {
-    const headers = { Authorization: `Bearer ${this.apiKey}` };
+    try {
+      // Create download directory if it doesn't exist (Actually check for write access, not existence, since fsPromises does not have a `existsSync` method)
+      try {
+        await fs.access(downloadPath);
+      } catch {
+        await fs.mkdir(downloadPath, { recursive: true });
+      }
 
-    // Create download directory if it doesn't exist (Actually check for write access, not existence, since fsPromises does not have a `existsSync` method)
-    if (!fs.access(downloadPath)) {
-      await fs.mkdir(downloadPath, { recursive: true });
-    }
-
-    const images: Record<string, any>[] = [];
-    for (const result of jsonResult) {
-      const jobId = result.job_id;
-      for (const page of result.pages) {
-        if (this.verbose) {
-          console.log(`> Image for page ${page.page}: ${page.images}`);
-        }
-        for (const image of page.images) {
-          const imageName = image.name;
-          // Get the full path
-          let imagePath = `${downloadPath}/${jobId}-${imageName}`;
-
-          if (!imagePath.endsWith(".png") && !imagePath.endsWith(".jpg")) {
-            imagePath += ".png";
+      const images: Record<string, any>[] = [];
+      for (const result of jsonResult) {
+        const jobId = result.job_id;
+        for (const page of result.pages) {
+          if (this.verbose) {
+            console.log(`> Image for page ${page.page}: ${page.images}`);
           }
-
-          // Get a valid image path
-          image.path = imagePath;
-          image.job_id = jobId;
-          image.original_pdf_path = result.file_path;
-          image.page_number = page.page;
-
-          const imageUrl = `${this.baseUrl}/job/${jobId}/result/image/${imageName}`;
-          const response = await fetch(imageUrl, { headers });
-          if (!response.ok) {
-            throw new Error(
-              `Failed to download image: ${await response.text()}`,
+          for (const image of page.images) {
+            const imageName = image.name;
+            const imagePath = await this.getImagePath(
+              downloadPath,
+              jobId,
+              imageName,
             );
+            await this.fetchAndSaveImage(imageName, imagePath, jobId);
+            // Assign metadata to the image
+            image.path = imagePath;
+            image.job_id = jobId;
+            image.original_pdf_path = result.file_path;
+            image.page_number = page.page;
+            images.push(image);
           }
-          const arrayBuffer = await response.arrayBuffer();
-          const buffer = new Uint8Array(arrayBuffer);
-          await fs.writeFile(imagePath, buffer);
-
-          images.push(image);
         }
       }
+      return images;
+    } catch (e) {
+      console.error(`Error while downloading images from the parsed result`, e);
+      if (this.ignoreErrors) {
+        return [];
+      } else {
+        throw e;
+      }
     }
-    return images;
   }
 
-  private async getMimeType(
+  private async getImagePath(
+    downloadPath: string,
+    jobId: string,
+    imageName: string,
+  ): Promise<string> {
+    // Get the full path
+    let imagePath = `${downloadPath}/${jobId}-${imageName}`;
+    // Get a valid image path
+    if (!imagePath.endsWith(".png") && !imagePath.endsWith(".jpg")) {
+      imagePath += ".png";
+    }
+
+    return imagePath;
+  }
+
+  private async fetchAndSaveImage(
+    imageName: string,
+    imagePath: string,
+    jobId: string,
+  ): Promise<void> {
+    const headers = { Authorization: `Bearer ${this.apiKey}` };
+    // Construct the image URL
+    const imageUrl = `${this.baseUrl}/job/${jobId}/result/image/${imageName}`;
+    const response = await fetch(imageUrl, { headers });
+    if (!response.ok) {
+      throw new Error(`Failed to download image: ${await response.text()}`);
+    }
+    // Convert the response to an ArrayBuffer and then to a Buffer
+    const arrayBuffer = await response.arrayBuffer();
+    const buffer = new Uint8Array(arrayBuffer);
+    // Write the image buffer to the specified imagePath
+    await fs.writeFile(imagePath, buffer);
+  }
+
+  static async getMimeType(
     data: Uint8Array,
-  ): Promise<{ mimeType: string; extension: string }> {
-    const mimes = filetypemime(data); // Get an array of possible MIME types
-    const extension = Object.keys(SupportedFiles).find(
-      (ext) => SupportedFiles[ext] === mimes[0],
-    ); // Find the extension for the first MIME type
-    if (!extension) {
-      const supportedExtensions = Object.keys(SupportedFiles).join(", ");
+  ): Promise<{ mime: string; extension: string }> {
+    const typeinfos = filetypeinfo(data);
+    // find the first type info that matches the supported MIME types
+    // It could be happened that docx file is recognized as zip file, so we need to check the mime type
+    const info = typeinfos.find((info) => {
+      if (info.extension && SUPPORT_FILE_EXT.includes(`.${info.extension}`)) {
+        return info;
+      }
+    });
+    if (!info || !info.mime || !info.extension) {
+      const ext = SUPPORT_FILE_EXT.join(", ");
       throw new Error(
-        `File has type "${mimes[0]}" which does not match supported MIME Types. Supported formats include: ${supportedExtensions}`,
+        `File has type which does not match supported MIME Types. Supported formats include: ${ext}`,
       );
     }
-    return { mimeType: mimes[0], extension }; // Return the first MIME type and its corresponding extension
+    return { mime: info.mime, extension: info.extension };
   }
 }

packages/llamaindex/tests/readers/llama-parser-reader.test.ts

@@ -0,0 +1,15 @@
+import { LlamaParseReader } from "llamaindex";
+import { readFile } from "node:fs/promises";
+import { join } from "node:path";
+import { fileURLToPath } from "node:url";
+import { expect, test } from "vitest";
+
+const fixturesDir = fileURLToPath(new URL("./fixtures", import.meta.url));
+
+test("file type should be detected correctly", async () => {
+  const xlsx = join(fixturesDir, "test.xlsx");
+  const buffer = await readFile(xlsx);
+  const { mime, extension } = await LlamaParseReader.getMimeType(buffer);
+  expect(mime).toBe("application/vnd.oasis.opendocument.spreadsheet");
+  expect(extension).toBe("ods");
+});