mirror of
https://github.com/run-llama/llama_cloud_services.git
synced 2026-07-21 12:05:23 -04:00
Compare commits
3 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| fefaa7de80 | |||
| d0649ece6e | |||
| 5d4cabd843 |
@@ -21,7 +21,7 @@ jobs:
|
||||
- uses: pnpm/action-setup@v4
|
||||
|
||||
- name: Setup Node.js
|
||||
uses: actions/setup-node@v5
|
||||
uses: actions/setup-node@v6
|
||||
with:
|
||||
node-version-file: "ts/llama_cloud_services/.nvmrc"
|
||||
|
||||
|
||||
@@ -31,7 +31,7 @@ jobs:
|
||||
|
||||
- uses: pnpm/action-setup@v4
|
||||
- name: Setup Node.js
|
||||
uses: actions/setup-node@v5
|
||||
uses: actions/setup-node@v6
|
||||
with:
|
||||
node-version-file: "ts/llama_cloud_services/.nvmrc"
|
||||
- name: Install dependencies
|
||||
|
||||
@@ -24,7 +24,7 @@ jobs:
|
||||
- uses: actions/checkout@v5
|
||||
- uses: pnpm/action-setup@v4
|
||||
- name: Setup Node.js
|
||||
uses: actions/setup-node@v5
|
||||
uses: actions/setup-node@v6
|
||||
with:
|
||||
node-version-file: "ts/llama_cloud_services/.nvmrc"
|
||||
- name: Install dependencies
|
||||
|
||||
@@ -20,7 +20,7 @@ jobs:
|
||||
- uses: pnpm/action-setup@v4
|
||||
|
||||
- name: Setup Node.js
|
||||
uses: actions/setup-node@v5
|
||||
uses: actions/setup-node@v6
|
||||
with:
|
||||
node-version: "22"
|
||||
cache: "pnpm"
|
||||
|
||||
@@ -9,10 +9,12 @@ test("LlamaIndex module resolution test", async (t) => {
|
||||
const index = new LlamaCloudIndex({
|
||||
name: "test-index",
|
||||
projectName: "Default",
|
||||
apiKey: process.env.LLAMA_CLOUD_API_KEY || "test-key",
|
||||
});
|
||||
const reader = new LlamaParseReader({
|
||||
resultType: "markdown",
|
||||
verbose: false,
|
||||
apiKey: process.env.LLAMA_CLOUD_API_KEY || "test-key",
|
||||
});
|
||||
ok(index !== undefined);
|
||||
ok(reader !== undefined);
|
||||
@@ -24,6 +26,7 @@ test("LlamaIndex module resolution test", async (t) => {
|
||||
const index = new mod.LlamaCloudIndex({
|
||||
name: "test-index",
|
||||
projectName: "Default",
|
||||
apiKey: process.env.LLAMA_CLOUD_API_KEY || "test-key",
|
||||
});
|
||||
ok(index !== undefined);
|
||||
});
|
||||
|
||||
@@ -1,5 +1,11 @@
|
||||
# llama-cloud-services
|
||||
|
||||
## 0.3.9
|
||||
|
||||
### Patch Changes
|
||||
|
||||
- 5d4cabd: Add ImageNode support in TypeScript
|
||||
|
||||
## 0.3.8
|
||||
|
||||
### Patch Changes
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
{
|
||||
"name": "llama-cloud-services",
|
||||
"version": "0.3.8",
|
||||
"version": "0.3.9",
|
||||
"type": "module",
|
||||
"license": "MIT",
|
||||
"scripts": {
|
||||
|
||||
@@ -9,10 +9,16 @@ import { DEFAULT_PROJECT_NAME } from "@llamaindex/core/global";
|
||||
import type { QueryBundle } from "@llamaindex/core/query-engine";
|
||||
import { BaseRetriever } from "@llamaindex/core/retriever";
|
||||
import type { NodeWithScore } from "@llamaindex/core/schema";
|
||||
import { jsonToNode, ObjectType } from "@llamaindex/core/schema";
|
||||
import { jsonToNode, ObjectType, ImageNode } from "@llamaindex/core/schema";
|
||||
import { extractText } from "@llamaindex/core/utils";
|
||||
import type { ClientParams, CloudConstructorParams } from "./type.js";
|
||||
import { getPipelineId, initService } from "./utils.js";
|
||||
import { getPipelineId, getProjectId, initService } from "./utils.js";
|
||||
import {
|
||||
type PageScreenshotNodeWithScore,
|
||||
type PageFigureNodeWithScore,
|
||||
generateFilePageScreenshotPresignedUrlApiV1FilesIdPageScreenshotsPageIndexPresignedUrlPost,
|
||||
generateFilePageFigurePresignedUrlApiV1FilesIdPageFiguresPageIndexFigureNamePresignedUrlPost,
|
||||
} from "./api";
|
||||
|
||||
export type CloudRetrieveParams = Omit<
|
||||
RetrievalParams,
|
||||
@@ -43,6 +49,95 @@ export class LlamaCloudRetriever extends BaseRetriever {
|
||||
});
|
||||
}
|
||||
|
||||
private async fetchBase64FromPresignedUrl(url: string): Promise<string> {
|
||||
const response = await fetch(url);
|
||||
if (!response.ok) {
|
||||
throw new Error(
|
||||
`Failed to fetch media from presigned URL: ${response.status} ${response.statusText}`,
|
||||
);
|
||||
}
|
||||
const buffer = Buffer.from(await response.arrayBuffer());
|
||||
return buffer.toString("base64");
|
||||
}
|
||||
|
||||
private async pageScreenshotNodesToNodeWithScore(
|
||||
nodes: PageScreenshotNodeWithScore[] | undefined,
|
||||
projectId: string,
|
||||
): Promise<NodeWithScore[]> {
|
||||
if (!nodes || nodes.length === 0) return [];
|
||||
|
||||
const results = await Promise.all(
|
||||
nodes.map(async (n) => {
|
||||
const { data: presigned } =
|
||||
await generateFilePageScreenshotPresignedUrlApiV1FilesIdPageScreenshotsPageIndexPresignedUrlPost(
|
||||
{
|
||||
throwOnError: true,
|
||||
path: {
|
||||
id: n.node.file_id,
|
||||
page_index: n.node.page_index,
|
||||
},
|
||||
query: {
|
||||
project_id: projectId,
|
||||
organization_id: this.organizationId ?? null,
|
||||
},
|
||||
},
|
||||
);
|
||||
const base64 = await this.fetchBase64FromPresignedUrl(presigned.url);
|
||||
const imageNode = new ImageNode({
|
||||
image: base64,
|
||||
metadata: {
|
||||
...(n.node.metadata ?? {}),
|
||||
file_id: n.node.file_id,
|
||||
page_index: n.node.page_index,
|
||||
},
|
||||
});
|
||||
return { node: imageNode, score: n.score } satisfies NodeWithScore;
|
||||
}),
|
||||
);
|
||||
|
||||
return results;
|
||||
}
|
||||
|
||||
private async pageFigureNodesToNodeWithScore(
|
||||
nodes: PageFigureNodeWithScore[] | undefined,
|
||||
projectId: string,
|
||||
): Promise<NodeWithScore[]> {
|
||||
if (!nodes || nodes.length === 0) return [];
|
||||
|
||||
const results = await Promise.all(
|
||||
nodes.map(async (n) => {
|
||||
const { data: presigned } =
|
||||
await generateFilePageFigurePresignedUrlApiV1FilesIdPageFiguresPageIndexFigureNamePresignedUrlPost(
|
||||
{
|
||||
throwOnError: true,
|
||||
path: {
|
||||
id: n.node.file_id,
|
||||
page_index: n.node.page_index,
|
||||
figure_name: n.node.figure_name,
|
||||
},
|
||||
query: {
|
||||
project_id: projectId,
|
||||
organization_id: this.organizationId ?? null,
|
||||
},
|
||||
},
|
||||
);
|
||||
const base64 = await this.fetchBase64FromPresignedUrl(presigned.url);
|
||||
const imageNode = new ImageNode({
|
||||
image: base64,
|
||||
metadata: {
|
||||
...(n.node.metadata ?? {}),
|
||||
file_id: n.node.file_id,
|
||||
page_index: n.node.page_index,
|
||||
figure_name: n.node.figure_name,
|
||||
},
|
||||
});
|
||||
return { node: imageNode, score: n.score } satisfies NodeWithScore;
|
||||
}),
|
||||
);
|
||||
|
||||
return results;
|
||||
}
|
||||
|
||||
// LlamaCloud expects null values for filters, but LlamaIndexTS uses undefined for empty values
|
||||
// This function converts the undefined values to null
|
||||
private convertFilter(filters?: MetadataFilters): MetadataFilters | null {
|
||||
@@ -76,6 +171,35 @@ export class LlamaCloudRetriever extends BaseRetriever {
|
||||
}
|
||||
|
||||
async _retrieve(query: QueryBundle): Promise<NodeWithScore[]> {
|
||||
// Handle deprecated image retrieval flag
|
||||
const retrieveImageNodes = (this.retrieveParams as RetrievalParams)
|
||||
.retrieve_image_nodes;
|
||||
if (typeof retrieveImageNodes !== "undefined") {
|
||||
console.warn(
|
||||
"The `retrieve_image_nodes` parameter is deprecated. Use `retrieve_page_screenshot_nodes` and `retrieve_page_figure_nodes` instead.",
|
||||
);
|
||||
}
|
||||
|
||||
const retrievePageScreenshotNodes = (this.retrieveParams as RetrievalParams)
|
||||
.retrieve_page_screenshot_nodes;
|
||||
const retrievePageFigureNodes = (this.retrieveParams as RetrievalParams)
|
||||
.retrieve_page_figure_nodes;
|
||||
|
||||
if (retrieveImageNodes) {
|
||||
if (
|
||||
retrievePageScreenshotNodes === false ||
|
||||
retrievePageFigureNodes === false
|
||||
) {
|
||||
throw new Error(
|
||||
"If `retrieve_image_nodes` is set to true, both `retrieve_page_screenshot_nodes` and `retrieve_page_figure_nodes` must also be set to true or omitted.",
|
||||
);
|
||||
}
|
||||
(this.retrieveParams as RetrievalParams).retrieve_page_screenshot_nodes =
|
||||
true;
|
||||
(this.retrieveParams as RetrievalParams).retrieve_page_figure_nodes =
|
||||
true;
|
||||
}
|
||||
|
||||
const pipelineId = await getPipelineId(
|
||||
this.pipelineName,
|
||||
this.projectName,
|
||||
@@ -98,6 +222,34 @@ export class LlamaCloudRetriever extends BaseRetriever {
|
||||
},
|
||||
});
|
||||
|
||||
return this.resultNodesToNodeWithScore(results.retrieval_nodes);
|
||||
const textNodes = this.resultNodesToNodeWithScore(results.retrieval_nodes);
|
||||
|
||||
const needScreenshots = (this.retrieveParams as RetrievalParams)
|
||||
.retrieve_page_screenshot_nodes;
|
||||
const needFigures = (this.retrieveParams as RetrievalParams)
|
||||
.retrieve_page_figure_nodes;
|
||||
|
||||
if (!needScreenshots && !needFigures) {
|
||||
return textNodes;
|
||||
}
|
||||
|
||||
const projectId = await getProjectId(this.projectName, this.organizationId);
|
||||
|
||||
const [screenshotNodes, figureNodes] = await Promise.all([
|
||||
needScreenshots
|
||||
? this.pageScreenshotNodesToNodeWithScore(
|
||||
results.image_nodes,
|
||||
projectId,
|
||||
)
|
||||
: Promise.resolve([] as NodeWithScore[]),
|
||||
needFigures
|
||||
? this.pageFigureNodesToNodeWithScore(
|
||||
results.page_figure_nodes,
|
||||
projectId,
|
||||
)
|
||||
: Promise.resolve([] as NodeWithScore[]),
|
||||
]);
|
||||
|
||||
return [...textNodes, ...screenshotNodes, ...figureNodes];
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user