merge with master
This commit is contained in:
commit
51765cfe97
|
@ -5,11 +5,8 @@ labels: [possible bug]
|
|||
body:
|
||||
- type: markdown
|
||||
attributes:
|
||||
value: |
|
||||
Use this template to file a bug report for AnythingLLM. Please be as descriptive as possible to allow everyone to replicate and solve your issue.
|
||||
|
||||
Want help contributing a PR? Use our repo chatbot by OnboardAI! https://learnthisrepo.com/anythingllm
|
||||
|
||||
value: |
|
||||
Use this template to file a bug report for AnythingLLM. Please be as descriptive as possible to allow everyone to replicate and solve your issue. Want help contributing a PR? Use our repo chatbot by OnboardAI! https://learnthisrepo.com/anythingllm"
|
||||
- type: dropdown
|
||||
id: runtime
|
||||
attributes:
|
||||
|
|
|
@ -1,6 +1,7 @@
|
|||
{
|
||||
"cSpell.words": [
|
||||
"anythingllm",
|
||||
"Astra",
|
||||
"Dockerized",
|
||||
"Embeddable",
|
||||
"hljs",
|
||||
|
|
|
@ -23,7 +23,12 @@ Here you can find the scripts and known working process to run AnythingLLM outsi
|
|||
|
||||
2. `cd anything-llm` and run `yarn setup`. This will install all dependencies to run in production as well as debug the application.
|
||||
|
||||
3. `cp server/.env.example server/.env` to create the basic ENV file for where instance settings will be read from on service start. This file is automatically managed and should not be edited manually.
|
||||
3. `cp server/.env.example server/.env` to create the basic ENV file for where instance settings will be read from on service start.
|
||||
|
||||
4. Ensure that the `server/.env` file has _at least_ these keys to start. These values will persist and this file will be automatically written and managed after your first successful boot.
|
||||
```
|
||||
STORAGE_DIR="/your/absolute/path/to/server/.env"
|
||||
```
|
||||
|
||||
## To start the application
|
||||
|
||||
|
@ -45,10 +50,10 @@ cd server && npx prisma migrate deploy --schema=./prisma/schema.prisma
|
|||
```
|
||||
|
||||
4. Boot the server in production
|
||||
`cd server && NODE_ENV=production index.js &`
|
||||
`cd server && NODE_ENV=production node index.js &`
|
||||
|
||||
5. Boot the collection in another process
|
||||
`cd collector && NODE_ENV=production index.js &`
|
||||
`cd collector && NODE_ENV=production node index.js &`
|
||||
|
||||
AnythingLLM should now be running on `http://localhost:3001`!
|
||||
|
||||
|
|
|
@ -84,6 +84,7 @@ Some cool features of AnythingLLM
|
|||
**Supported Vector Databases:**
|
||||
|
||||
- [LanceDB](https://github.com/lancedb/lancedb) (default)
|
||||
- [Astra DB](https://www.datastax.com/products/datastax-astra)
|
||||
- [Pinecone](https://pinecone.io)
|
||||
- [Chroma](https://trychroma.com)
|
||||
- [Weaviate](https://weaviate.io)
|
||||
|
|
|
@ -54,6 +54,7 @@ GID='1000'
|
|||
# Only used if you are using an LLM that does not natively support embedding (openai or Azure)
|
||||
# EMBEDDING_ENGINE='openai'
|
||||
# OPEN_AI_KEY=sk-xxxx
|
||||
# EMBEDDING_MODEL_PREF='text-embedding-ada-002'
|
||||
|
||||
# EMBEDDING_ENGINE='azure'
|
||||
# AZURE_OPENAI_ENDPOINT=
|
||||
|
@ -103,6 +104,11 @@ GID='1000'
|
|||
# ZILLIZ_ENDPOINT="https://sample.api.gcp-us-west1.zillizcloud.com"
|
||||
# ZILLIZ_API_TOKEN=api-token-here
|
||||
|
||||
# Enable all below if you are using vector database: Astra DB.
|
||||
# VECTOR_DB="astra"
|
||||
# ASTRA_DB_APPLICATION_TOKEN=
|
||||
# ASTRA_DB_ENDPOINT=
|
||||
|
||||
# CLOUD DEPLOYMENT VARIRABLES ONLY
|
||||
# AUTH_TOKEN="hunter2" # This is the password to your application if remote hosting.
|
||||
|
||||
|
|
|
@ -22,12 +22,27 @@ export default function OpenAiOptions({ settings }) {
|
|||
Model Preference
|
||||
</label>
|
||||
<select
|
||||
disabled={true}
|
||||
className="cursor-not-allowed bg-zinc-900 border border-gray-500 text-white text-sm rounded-lg block w-full p-2.5"
|
||||
name="EmbeddingModelPref"
|
||||
required={true}
|
||||
className="bg-zinc-900 border border-gray-500 text-white text-sm rounded-lg block w-full p-2.5"
|
||||
>
|
||||
<option disabled={true} selected={true}>
|
||||
text-embedding-ada-002
|
||||
</option>
|
||||
<optgroup label="Available embedding models">
|
||||
{[
|
||||
"text-embedding-ada-002",
|
||||
"text-embedding-3-small",
|
||||
"text-embedding-3-large",
|
||||
].map((model) => {
|
||||
return (
|
||||
<option
|
||||
key={model}
|
||||
value={model}
|
||||
selected={settings?.EmbeddingModelPref === model}
|
||||
>
|
||||
{model}
|
||||
</option>
|
||||
);
|
||||
})}
|
||||
</optgroup>
|
||||
</select>
|
||||
</div>
|
||||
</div>
|
||||
|
|
|
@ -85,6 +85,7 @@ function OpenAIModelSelection({ apiKey, settings }) {
|
|||
"gpt-3.5-turbo",
|
||||
"gpt-3.5-turbo-1106",
|
||||
"gpt-4",
|
||||
"gpt-4-turbo-preview",
|
||||
"gpt-4-1106-preview",
|
||||
"gpt-4-32k",
|
||||
].map((model) => {
|
||||
|
|
|
@ -6,9 +6,14 @@ import Directory from "./Directory";
|
|||
import showToast from "../../../../utils/toast";
|
||||
import WorkspaceDirectory from "./WorkspaceDirectory";
|
||||
|
||||
// OpenAI Cost per token for text-ada-embedding
|
||||
// OpenAI Cost per token
|
||||
// ref: https://openai.com/pricing#:~:text=%C2%A0/%201K%20tokens-,Embedding%20models,-Build%20advanced%20search
|
||||
const COST_PER_TOKEN = 0.0000001; // $0.0001 / 1K tokens
|
||||
|
||||
const MODEL_COSTS = {
|
||||
"text-embedding-ada-002": 0.0000001, // $0.0001 / 1K tokens
|
||||
"text-embedding-3-small": 0.00000002, // $0.00002 / 1K tokens
|
||||
"text-embedding-3-large": 0.00000013, // $0.00013 / 1K tokens
|
||||
};
|
||||
|
||||
export default function DocumentSettings({
|
||||
workspace,
|
||||
|
@ -142,10 +147,12 @@ export default function DocumentSettings({
|
|||
});
|
||||
|
||||
// Do not do cost estimation unless the embedding engine is OpenAi.
|
||||
if (
|
||||
!systemSettings?.EmbeddingEngine ||
|
||||
systemSettings.EmbeddingEngine === "openai"
|
||||
) {
|
||||
if (systemSettings?.EmbeddingEngine === "openai") {
|
||||
const COST_PER_TOKEN =
|
||||
MODEL_COSTS[
|
||||
systemSettings?.EmbeddingModelPref || "text-embedding-ada-002"
|
||||
];
|
||||
|
||||
const dollarAmount = (totalTokenCount / 1000) * COST_PER_TOKEN;
|
||||
setEmbeddingsCost(dollarAmount);
|
||||
}
|
||||
|
|
|
@ -8,6 +8,7 @@ const PROVIDER_DEFAULT_MODELS = {
|
|||
"gpt-3.5-turbo",
|
||||
"gpt-3.5-turbo-1106",
|
||||
"gpt-4",
|
||||
"gpt-4-turbo-preview",
|
||||
"gpt-4-1106-preview",
|
||||
"gpt-4-32k",
|
||||
],
|
||||
|
|
|
@ -44,6 +44,7 @@ export default function WorkspaceSettings({ active, workspace, settings }) {
|
|||
const formEl = useRef(null);
|
||||
const [saving, setSaving] = useState(false);
|
||||
const [hasChanges, setHasChanges] = useState(false);
|
||||
const [deleting, setDeleting] = useState(false);
|
||||
const defaults = recommendedSettings(settings?.LLMProvider);
|
||||
|
||||
const handleUpdate = async (e) => {
|
||||
|
@ -72,7 +73,15 @@ export default function WorkspaceSettings({ active, workspace, settings }) {
|
|||
)
|
||||
)
|
||||
return false;
|
||||
await Workspace.delete(workspace.slug);
|
||||
|
||||
setDeleting(true);
|
||||
const success = await Workspace.delete(workspace.slug);
|
||||
if (!success) {
|
||||
showToast("Workspace could not be deleted!", "error", { clear: true });
|
||||
setDeleting(false);
|
||||
return;
|
||||
}
|
||||
|
||||
workspace.slug === slug
|
||||
? (window.location = paths.home())
|
||||
: window.location.reload();
|
||||
|
@ -310,7 +319,11 @@ export default function WorkspaceSettings({ active, workspace, settings }) {
|
|||
</div>
|
||||
</div>
|
||||
<div className="flex items-center justify-between p-2 md:p-6 space-x-2 border-t rounded-b border-gray-600">
|
||||
<DeleteWorkspace workspace={workspace} onClick={deleteWorkspace} />
|
||||
<DeleteWorkspace
|
||||
deleting={deleting}
|
||||
workspace={workspace}
|
||||
onClick={deleteWorkspace}
|
||||
/>
|
||||
{hasChanges && (
|
||||
<button
|
||||
type="submit"
|
||||
|
@ -324,7 +337,7 @@ export default function WorkspaceSettings({ active, workspace, settings }) {
|
|||
);
|
||||
}
|
||||
|
||||
function DeleteWorkspace({ workspace, onClick }) {
|
||||
function DeleteWorkspace({ deleting, workspace, onClick }) {
|
||||
const [canDelete, setCanDelete] = useState(false);
|
||||
useEffect(() => {
|
||||
async function fetchKeys() {
|
||||
|
@ -337,11 +350,12 @@ function DeleteWorkspace({ workspace, onClick }) {
|
|||
if (!canDelete) return null;
|
||||
return (
|
||||
<button
|
||||
disabled={deleting}
|
||||
onClick={onClick}
|
||||
type="button"
|
||||
className="transition-all duration-300 border border-transparent rounded-lg whitespace-nowrap text-sm px-5 py-2.5 focus:z-10 bg-transparent text-white hover:text-white hover:bg-red-600"
|
||||
className="transition-all duration-300 border border-transparent rounded-lg whitespace-nowrap text-sm px-5 py-2.5 focus:z-10 bg-transparent text-white hover:text-white hover:bg-red-600 disabled:bg-red-600 disabled:text-red-200 disabled:animate-pulse"
|
||||
>
|
||||
Delete Workspace
|
||||
{deleting ? "Deleting Workspace..." : "Delete Workspace"}
|
||||
</button>
|
||||
);
|
||||
}
|
||||
|
|
|
@ -0,0 +1,41 @@
|
|||
export default function AstraDBOptions({ settings }) {
|
||||
return (
|
||||
<div className="w-full flex flex-col gap-y-4">
|
||||
<div className="w-full flex items-center gap-4">
|
||||
<div className="flex flex-col w-60">
|
||||
<label className="text-white text-sm font-semibold block mb-4">
|
||||
Astra DB Endpoint
|
||||
</label>
|
||||
<input
|
||||
type="url"
|
||||
name="AstraDBEndpoint"
|
||||
className="bg-zinc-900 text-white placeholder-white placeholder-opacity-60 text-sm rounded-lg focus:border-white block w-full p-2.5"
|
||||
placeholder="Astra DB API endpoint"
|
||||
defaultValue={settings?.AstraDBEndpoint}
|
||||
required={true}
|
||||
autoComplete="off"
|
||||
spellCheck={false}
|
||||
/>
|
||||
</div>
|
||||
|
||||
<div className="flex flex-col w-60">
|
||||
<label className="text-white text-sm font-semibold block mb-4">
|
||||
Astra DB Application Token
|
||||
</label>
|
||||
<input
|
||||
type="password"
|
||||
name="AstraDBApplicationToken"
|
||||
className="bg-zinc-900 text-white placeholder-white placeholder-opacity-60 text-sm rounded-lg focus:border-white block w-full p-2.5"
|
||||
placeholder="AstraCS:..."
|
||||
defaultValue={
|
||||
settings?.AstraDBApplicationToken ? "*".repeat(20) : ""
|
||||
}
|
||||
required={true}
|
||||
autoComplete="off"
|
||||
spellCheck={false}
|
||||
/>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
);
|
||||
}
|
Binary file not shown.
After Width: | Height: | Size: 1.5 KiB |
|
@ -10,6 +10,7 @@ import WeaviateLogo from "@/media/vectordbs/weaviate.png";
|
|||
import QDrantLogo from "@/media/vectordbs/qdrant.png";
|
||||
import MilvusLogo from "@/media/vectordbs/milvus.png";
|
||||
import ZillizLogo from "@/media/vectordbs/zilliz.png";
|
||||
import AstraDBLogo from "@/media/vectordbs/astraDB.png";
|
||||
import PreLoader from "@/components/Preloader";
|
||||
import ChangeWarningModal from "@/components/ChangeWarning";
|
||||
import { MagnifyingGlass } from "@phosphor-icons/react";
|
||||
|
@ -23,6 +24,7 @@ import MilvusDBOptions from "@/components/VectorDBSelection/MilvusDBOptions";
|
|||
import ZillizCloudOptions from "@/components/VectorDBSelection/ZillizCloudOptions";
|
||||
import { useModal } from "@/hooks/useModal";
|
||||
import ModalWrapper from "@/components/ModalWrapper";
|
||||
import AstraDBOptions from "@/components/VectorDBSelection/AstraDBOptions";
|
||||
|
||||
export default function GeneralVectorDatabase() {
|
||||
const [saving, setSaving] = useState(false);
|
||||
|
@ -100,6 +102,13 @@ export default function GeneralVectorDatabase() {
|
|||
options: <MilvusDBOptions settings={settings} />,
|
||||
description: "Open-source, highly scalable, and blazing fast.",
|
||||
},
|
||||
{
|
||||
name: "AstraDB",
|
||||
value: "astra",
|
||||
logo: AstraDBLogo,
|
||||
options: <AstraDBOptions settings={settings} />,
|
||||
description: "Vector Search for Real-world GenAI.",
|
||||
},
|
||||
];
|
||||
|
||||
const updateVectorChoice = (selection) => {
|
||||
|
|
|
@ -11,6 +11,7 @@ import LMStudioLogo from "@/media/llmprovider/lmstudio.png";
|
|||
import LocalAiLogo from "@/media/llmprovider/localai.png";
|
||||
import MistralLogo from "@/media/llmprovider/mistral.jpeg";
|
||||
import ZillizLogo from "@/media/vectordbs/zilliz.png";
|
||||
import AstraDBLogo from "@/media/vectordbs/astraDB.png";
|
||||
import ChromaLogo from "@/media/vectordbs/chroma.png";
|
||||
import PineconeLogo from "@/media/vectordbs/pinecone.png";
|
||||
import LanceDbLogo from "@/media/vectordbs/lancedb.png";
|
||||
|
@ -147,6 +148,13 @@ const VECTOR_DB_PRIVACY = {
|
|||
],
|
||||
logo: ZillizLogo,
|
||||
},
|
||||
astra: {
|
||||
name: "AstraDB",
|
||||
description: [
|
||||
"Your vectors and document text are stored on your cloud AstraDB database.",
|
||||
],
|
||||
logo: AstraDBLogo,
|
||||
},
|
||||
lancedb: {
|
||||
name: "LanceDB",
|
||||
description: [
|
||||
|
|
|
@ -7,6 +7,7 @@ import WeaviateLogo from "@/media/vectordbs/weaviate.png";
|
|||
import QDrantLogo from "@/media/vectordbs/qdrant.png";
|
||||
import MilvusLogo from "@/media/vectordbs/milvus.png";
|
||||
import ZillizLogo from "@/media/vectordbs/zilliz.png";
|
||||
import AstraDBLogo from "@/media/vectordbs/astraDB.png";
|
||||
import System from "@/models/system";
|
||||
import paths from "@/utils/paths";
|
||||
import PineconeDBOptions from "@/components/VectorDBSelection/PineconeDBOptions";
|
||||
|
@ -16,6 +17,7 @@ import WeaviateDBOptions from "@/components/VectorDBSelection/WeaviateDBOptions"
|
|||
import LanceDBOptions from "@/components/VectorDBSelection/LanceDBOptions";
|
||||
import MilvusOptions from "@/components/VectorDBSelection/MilvusDBOptions";
|
||||
import ZillizCloudOptions from "@/components/VectorDBSelection/ZillizCloudOptions";
|
||||
import AstraDBOptions from "@/components/VectorDBSelection/AstraDBOptions";
|
||||
import showToast from "@/utils/toast";
|
||||
import { useNavigate } from "react-router-dom";
|
||||
import VectorDBItem from "@/components/VectorDBSelection/VectorDBItem";
|
||||
|
@ -100,6 +102,13 @@ export default function VectorDatabaseConnection({
|
|||
options: <MilvusOptions settings={settings} />,
|
||||
description: "Open-source, highly scalable, and blazing fast.",
|
||||
},
|
||||
{
|
||||
name: "AstraDB",
|
||||
value: "astra",
|
||||
logo: AstraDBLogo,
|
||||
options: <AstraDBOptions settings={settings} />,
|
||||
description: "Vector Search for Real-world GenAI.",
|
||||
},
|
||||
];
|
||||
|
||||
function handleForward() {
|
||||
|
|
|
@ -51,6 +51,7 @@ JWT_SECRET="my-random-string-for-seeding" # Please generate random string at lea
|
|||
# Only used if you are using an LLM that does not natively support embedding (openai or Azure)
|
||||
# EMBEDDING_ENGINE='openai'
|
||||
# OPEN_AI_KEY=sk-xxxx
|
||||
# EMBEDDING_MODEL_PREF='text-embedding-ada-002'
|
||||
|
||||
# EMBEDDING_ENGINE='azure'
|
||||
# AZURE_OPENAI_ENDPOINT=
|
||||
|
@ -76,6 +77,11 @@ JWT_SECRET="my-random-string-for-seeding" # Please generate random string at lea
|
|||
# PINECONE_API_KEY=
|
||||
# PINECONE_INDEX=
|
||||
|
||||
# Enable all below if you are using vector database: Astra DB.
|
||||
# VECTOR_DB="astra"
|
||||
# ASTRA_DB_APPLICATION_TOKEN=
|
||||
# ASTRA_DB_ENDPOINT=
|
||||
|
||||
# Enable all below if you are using vector database: LanceDB.
|
||||
VECTOR_DB="lancedb"
|
||||
|
||||
|
|
|
@ -106,6 +106,9 @@ const Document = {
|
|||
await prisma.workspace_documents.delete({
|
||||
where: { id: document.id, workspaceId: workspace.id },
|
||||
});
|
||||
await prisma.document_vectors.deleteMany({
|
||||
where: { docId: document.docId },
|
||||
});
|
||||
} catch (error) {
|
||||
console.error(error.message);
|
||||
}
|
||||
|
|
|
@ -68,6 +68,12 @@ const SystemSettings = {
|
|||
ZillizApiToken: process.env.ZILLIZ_API_TOKEN,
|
||||
}
|
||||
: {}),
|
||||
...(vectorDB === "astra"
|
||||
? {
|
||||
AstraDBApplicationToken: process?.env?.ASTRA_DB_APPLICATION_TOKEN,
|
||||
AstraDBEndpoint: process?.env?.ASTRA_DB_ENDPOINT,
|
||||
}
|
||||
: {}),
|
||||
LLMProvider: llmProvider,
|
||||
...(llmProvider === "openai"
|
||||
? {
|
||||
|
|
|
@ -3,6 +3,7 @@ const slugify = require("slugify");
|
|||
const { Document } = require("./documents");
|
||||
const { WorkspaceUser } = require("./workspaceUsers");
|
||||
const { ROLES } = require("../utils/middleware/multiUserProtected");
|
||||
const { v4: uuidv4 } = require("uuid");
|
||||
|
||||
const Workspace = {
|
||||
writable: [
|
||||
|
@ -22,6 +23,7 @@ const Workspace = {
|
|||
new: async function (name = null, creatorId = null) {
|
||||
if (!name) return { result: null, message: "name cannot be null" };
|
||||
var slug = slugify(name, { lower: true });
|
||||
slug = slug || uuidv4();
|
||||
|
||||
const existingBySlug = await this.get({ slug });
|
||||
if (existingBySlug !== null) {
|
||||
|
|
|
@ -22,6 +22,7 @@
|
|||
"dependencies": {
|
||||
"@anthropic-ai/sdk": "^0.8.1",
|
||||
"@azure/openai": "1.0.0-beta.10",
|
||||
"@datastax/astra-db-ts": "^0.1.3",
|
||||
"@google/generative-ai": "^0.1.3",
|
||||
"@googleapis/youtube": "^9.0.0",
|
||||
"@pinecone-database/pinecone": "^2.0.1",
|
||||
|
|
|
@ -52,6 +52,8 @@ class OpenAiLLM {
|
|||
return 8192;
|
||||
case "gpt-4-1106-preview":
|
||||
return 128000;
|
||||
case "gpt-4-turbo-preview":
|
||||
return 128000;
|
||||
case "gpt-4-32k":
|
||||
return 32000;
|
||||
default:
|
||||
|
@ -65,6 +67,7 @@ class OpenAiLLM {
|
|||
"gpt-3.5-turbo",
|
||||
"gpt-3.5-turbo-1106",
|
||||
"gpt-4-1106-preview",
|
||||
"gpt-4-turbo-preview",
|
||||
"gpt-4-32k",
|
||||
];
|
||||
const isPreset = validModels.some((model) => modelName === model);
|
||||
|
|
|
@ -9,6 +9,7 @@ class OpenAiEmbedder {
|
|||
});
|
||||
const openai = new OpenAIApi(config);
|
||||
this.openai = openai;
|
||||
this.model = process.env.EMBEDDING_MODEL_PREF || "text-embedding-ada-002";
|
||||
|
||||
// Limit of how many strings we can process in a single pass to stay with resource or network limits
|
||||
this.maxConcurrentChunks = 500;
|
||||
|
@ -30,7 +31,7 @@ class OpenAiEmbedder {
|
|||
new Promise((resolve) => {
|
||||
this.openai
|
||||
.createEmbedding({
|
||||
model: "text-embedding-ada-002",
|
||||
model: this.model,
|
||||
input: chunk,
|
||||
})
|
||||
.then((res) => {
|
||||
|
|
|
@ -269,6 +269,7 @@ function handleStreamResponses(response, stream, responseProps) {
|
|||
for (const choice of event.choices) {
|
||||
const delta = choice.delta?.content;
|
||||
if (!delta) continue;
|
||||
fullText += delta;
|
||||
writeResponseChunk(response, {
|
||||
uuid,
|
||||
sources: [],
|
||||
|
|
|
@ -22,6 +22,9 @@ function getVectorDbClass() {
|
|||
case "zilliz":
|
||||
const { Zilliz } = require("../vectorDbProviders/zilliz");
|
||||
return Zilliz;
|
||||
case "astra":
|
||||
const { AstraDB } = require("../vectorDbProviders/astra");
|
||||
return AstraDB;
|
||||
default:
|
||||
throw new Error("ENV: No VECTOR_DB value found in environment!");
|
||||
}
|
||||
|
|
|
@ -204,6 +204,17 @@ const KEY_MAPPING = {
|
|||
checks: [isNotEmpty],
|
||||
},
|
||||
|
||||
// Astra DB Options
|
||||
|
||||
AstraDBApplicationToken: {
|
||||
envKey: "ASTRA_DB_APPLICATION_TOKEN",
|
||||
checks: [isNotEmpty],
|
||||
},
|
||||
AstraDBEndpoint: {
|
||||
envKey: "ASTRA_DB_ENDPOINT",
|
||||
checks: [isNotEmpty],
|
||||
},
|
||||
|
||||
// Together Ai Options
|
||||
TogetherAiApiKey: {
|
||||
envKey: "TOGETHER_AI_API_KEY",
|
||||
|
@ -322,6 +333,7 @@ function supportedVectorDB(input = "") {
|
|||
"qdrant",
|
||||
"milvus",
|
||||
"zilliz",
|
||||
"astra",
|
||||
];
|
||||
return supported.includes(input)
|
||||
? null
|
||||
|
|
|
@ -0,0 +1,22 @@
|
|||
# How to setup Astra Vector Database for AnythingLLM
|
||||
|
||||
[Official Astra DB Docs](https://docs.datastax.com/en/astra/astra-db-vector/get-started/quickstart.html) for reference.
|
||||
|
||||
### How to get started
|
||||
|
||||
**Requirements**
|
||||
|
||||
- Astra Vector Database with active status.
|
||||
|
||||
**Instructions**
|
||||
|
||||
- [Create an Astra account or sign in to an existing Astra account](astra.datastax.com)
|
||||
- Create an Astra Serverless(Vector) Database.
|
||||
- Make sure DB is in active state.
|
||||
- Get `API ENDPOINT`and `Application Token` from Overview screen
|
||||
|
||||
```
|
||||
VECTOR_DB="astra"
|
||||
ASTRA_DB_ENDPOINT=Astra DB API endpoint
|
||||
ASTRA_DB_APPLICATION_TOKEN=AstraCS:..
|
||||
```
|
|
@ -0,0 +1,380 @@
|
|||
const { AstraDB: AstraClient } = require("@datastax/astra-db-ts");
|
||||
const { RecursiveCharacterTextSplitter } = require("langchain/text_splitter");
|
||||
const { storeVectorResult, cachedVectorInformation } = require("../../files");
|
||||
const { v4: uuidv4 } = require("uuid");
|
||||
const {
|
||||
toChunks,
|
||||
getLLMProvider,
|
||||
getEmbeddingEngineSelection,
|
||||
} = require("../../helpers");
|
||||
|
||||
const AstraDB = {
|
||||
name: "AstraDB",
|
||||
connect: async function () {
|
||||
if (process.env.VECTOR_DB !== "astra")
|
||||
throw new Error("AstraDB::Invalid ENV settings");
|
||||
|
||||
const client = new AstraClient(
|
||||
process?.env?.ASTRA_DB_APPLICATION_TOKEN,
|
||||
process?.env?.ASTRA_DB_ENDPOINT
|
||||
);
|
||||
return { client };
|
||||
},
|
||||
heartbeat: async function () {
|
||||
return { heartbeat: Number(new Date()) };
|
||||
},
|
||||
// Astra interface will return a valid collection object even if the collection
|
||||
// does not actually exist. So we run a simple check which will always throw
|
||||
// when the table truly does not exist. Faster than iterating all collections.
|
||||
isRealCollection: async function (astraCollection = null) {
|
||||
if (!astraCollection) return false;
|
||||
return await astraCollection
|
||||
.countDocuments()
|
||||
.then(() => true)
|
||||
.catch(() => false);
|
||||
},
|
||||
totalVectors: async function () {
|
||||
const { client } = await this.connect();
|
||||
const collectionNames = await this.allNamespaces(client);
|
||||
var totalVectors = 0;
|
||||
for (const name of collectionNames) {
|
||||
const collection = await client.collection(name).catch(() => null);
|
||||
const count = await collection.countDocuments().catch(() => 0);
|
||||
totalVectors += count ? count : 0;
|
||||
}
|
||||
return totalVectors;
|
||||
},
|
||||
namespaceCount: async function (_namespace = null) {
|
||||
const { client } = await this.connect();
|
||||
const namespace = await this.namespace(client, _namespace);
|
||||
return namespace?.vectorCount || 0;
|
||||
},
|
||||
namespace: async function (client, namespace = null) {
|
||||
if (!namespace) throw new Error("No namespace value provided.");
|
||||
const collection = await client.collection(namespace).catch(() => null);
|
||||
if (!(await this.isRealCollection(collection))) return null;
|
||||
|
||||
const count = await collection.countDocuments().catch((e) => {
|
||||
console.error("Astra::namespaceExists", e.message);
|
||||
return null;
|
||||
});
|
||||
|
||||
return {
|
||||
name: namespace,
|
||||
...collection,
|
||||
vectorCount: typeof count === "number" ? count : 0,
|
||||
};
|
||||
},
|
||||
hasNamespace: async function (namespace = null) {
|
||||
if (!namespace) return false;
|
||||
const { client } = await this.connect();
|
||||
return await this.namespaceExists(client, namespace);
|
||||
},
|
||||
namespaceExists: async function (client, namespace = null) {
|
||||
if (!namespace) throw new Error("No namespace value provided.");
|
||||
const collection = await client.collection(namespace);
|
||||
return await this.isRealCollection(collection);
|
||||
},
|
||||
deleteVectorsInNamespace: async function (client, namespace = null) {
|
||||
await client.dropCollection(namespace);
|
||||
return true;
|
||||
},
|
||||
// AstraDB requires a dimension aspect for collection creation
|
||||
// we pass this in from the first chunk to infer the dimensions like other
|
||||
// providers do.
|
||||
getOrCreateCollection: async function (client, namespace, dimensions = null) {
|
||||
const isExists = await this.namespaceExists(client, namespace);
|
||||
if (!isExists) {
|
||||
if (!dimensions)
|
||||
throw new Error(
|
||||
`AstraDB:getOrCreateCollection Unable to infer vector dimension from input. Open an issue on Github for support.`
|
||||
);
|
||||
|
||||
await client.createCollection(namespace, {
|
||||
vector: {
|
||||
dimension: dimensions,
|
||||
metric: "cosine",
|
||||
},
|
||||
});
|
||||
}
|
||||
return await client.collection(namespace);
|
||||
},
|
||||
addDocumentToNamespace: async function (
|
||||
namespace,
|
||||
documentData = {},
|
||||
fullFilePath = null
|
||||
) {
|
||||
const { DocumentVectors } = require("../../../models/vectors");
|
||||
try {
|
||||
let vectorDimension = null;
|
||||
const { pageContent, docId, ...metadata } = documentData;
|
||||
if (!pageContent || pageContent.length == 0) return false;
|
||||
|
||||
console.log("Adding new vectorized document into namespace", namespace);
|
||||
const cacheResult = await cachedVectorInformation(fullFilePath);
|
||||
if (cacheResult.exists) {
|
||||
const { client } = await this.connect();
|
||||
const { chunks } = cacheResult;
|
||||
const documentVectors = [];
|
||||
vectorDimension = chunks[0][0].values.length || null;
|
||||
|
||||
const collection = await this.getOrCreateCollection(
|
||||
client,
|
||||
namespace,
|
||||
vectorDimension
|
||||
);
|
||||
if (!(await this.isRealCollection(collection)))
|
||||
throw new Error("Failed to create new AstraDB collection!", {
|
||||
namespace,
|
||||
});
|
||||
|
||||
for (const chunk of chunks) {
|
||||
// Before sending to Astra and saving the records to our db
|
||||
// we need to assign the id of each chunk that is stored in the cached file.
|
||||
const newChunks = chunk.map((chunk) => {
|
||||
const _id = uuidv4();
|
||||
documentVectors.push({ docId, vectorId: _id });
|
||||
return {
|
||||
_id: _id,
|
||||
$vector: chunk.values,
|
||||
metadata: chunk.metadata || {},
|
||||
};
|
||||
});
|
||||
|
||||
await collection.insertMany(newChunks);
|
||||
}
|
||||
await DocumentVectors.bulkInsert(documentVectors);
|
||||
return { vectorized: true, error: null };
|
||||
}
|
||||
|
||||
const textSplitter = new RecursiveCharacterTextSplitter({
|
||||
chunkSize:
|
||||
getEmbeddingEngineSelection()?.embeddingMaxChunkLength || 1_000,
|
||||
chunkOverlap: 20,
|
||||
});
|
||||
const textChunks = await textSplitter.splitText(pageContent);
|
||||
|
||||
console.log("Chunks created from document:", textChunks.length);
|
||||
const LLMConnector = getLLMProvider();
|
||||
const documentVectors = [];
|
||||
const vectors = [];
|
||||
const vectorValues = await LLMConnector.embedChunks(textChunks);
|
||||
|
||||
if (!!vectorValues && vectorValues.length > 0) {
|
||||
for (const [i, vector] of vectorValues.entries()) {
|
||||
if (!vectorDimension) vectorDimension = vector.length;
|
||||
const vectorRecord = {
|
||||
_id: uuidv4(),
|
||||
$vector: vector,
|
||||
metadata: { ...metadata, text: textChunks[i] },
|
||||
};
|
||||
|
||||
vectors.push(vectorRecord);
|
||||
documentVectors.push({ docId, vectorId: vectorRecord._id });
|
||||
}
|
||||
} else {
|
||||
throw new Error(
|
||||
"Could not embed document chunks! This document will not be recorded."
|
||||
);
|
||||
}
|
||||
const { client } = await this.connect();
|
||||
const collection = await this.getOrCreateCollection(
|
||||
client,
|
||||
namespace,
|
||||
vectorDimension
|
||||
);
|
||||
if (!(await this.isRealCollection(collection)))
|
||||
throw new Error("Failed to create new AstraDB collection!", {
|
||||
namespace,
|
||||
});
|
||||
|
||||
if (vectors.length > 0) {
|
||||
const chunks = [];
|
||||
|
||||
console.log("Inserting vectorized chunks into Astra DB.");
|
||||
|
||||
// AstraDB has maximum upsert size of 20 records per-request so we have to use a lower chunk size here
|
||||
// in order to do the queries - this takes a lot more time than other providers but there
|
||||
// is no way around it. This will save the vector-cache with the same layout, so we don't
|
||||
// have to chunk again for cached files.
|
||||
for (const chunk of toChunks(vectors, 20)) {
|
||||
chunks.push(
|
||||
chunk.map((c) => {
|
||||
return { id: c._id, values: c.$vector, metadata: c.metadata };
|
||||
})
|
||||
);
|
||||
await collection.insertMany(chunk);
|
||||
}
|
||||
await storeVectorResult(chunks, fullFilePath);
|
||||
}
|
||||
|
||||
await DocumentVectors.bulkInsert(documentVectors);
|
||||
return { vectorized: true, error: null };
|
||||
} catch (e) {
|
||||
console.error("addDocumentToNamespace", e.message);
|
||||
return { vectorized: false, error: e.message };
|
||||
}
|
||||
},
|
||||
deleteDocumentFromNamespace: async function (namespace, docId) {
|
||||
const { DocumentVectors } = require("../../../models/vectors");
|
||||
const { client } = await this.connect();
|
||||
if (!(await this.namespaceExists(client, namespace)))
|
||||
throw new Error(
|
||||
"Invalid namespace - has it been collected and populated yet?"
|
||||
);
|
||||
const collection = await client.collection(namespace);
|
||||
|
||||
const knownDocuments = await DocumentVectors.where({ docId });
|
||||
if (knownDocuments.length === 0) return;
|
||||
|
||||
const vectorIds = knownDocuments.map((doc) => doc.vectorId);
|
||||
for (const id of vectorIds) {
|
||||
await collection.deleteMany({
|
||||
_id: id,
|
||||
});
|
||||
}
|
||||
|
||||
const indexes = knownDocuments.map((doc) => doc.id);
|
||||
await DocumentVectors.deleteIds(indexes);
|
||||
return true;
|
||||
},
|
||||
performSimilaritySearch: async function ({
|
||||
namespace = null,
|
||||
input = "",
|
||||
LLMConnector = null,
|
||||
similarityThreshold = 0.25,
|
||||
topN = 4,
|
||||
}) {
|
||||
if (!namespace || !input || !LLMConnector)
|
||||
throw new Error("Invalid request to performSimilaritySearch.");
|
||||
|
||||
const { client } = await this.connect();
|
||||
if (!(await this.namespaceExists(client, namespace))) {
|
||||
return {
|
||||
contextTexts: [],
|
||||
sources: [],
|
||||
message:
|
||||
"Invalid query - no namespace found for workspace in vector db!",
|
||||
};
|
||||
}
|
||||
|
||||
const queryVector = await LLMConnector.embedTextInput(input);
|
||||
const { contextTexts, sourceDocuments } = await this.similarityResponse(
|
||||
client,
|
||||
namespace,
|
||||
queryVector,
|
||||
similarityThreshold,
|
||||
topN
|
||||
);
|
||||
|
||||
const sources = sourceDocuments.map((metadata, i) => {
|
||||
return { ...metadata, text: contextTexts[i] };
|
||||
});
|
||||
return {
|
||||
contextTexts,
|
||||
sources: this.curateSources(sources),
|
||||
message: false,
|
||||
};
|
||||
},
|
||||
similarityResponse: async function (
|
||||
client,
|
||||
namespace,
|
||||
queryVector,
|
||||
similarityThreshold = 0.25,
|
||||
topN = 4
|
||||
) {
|
||||
const result = {
|
||||
contextTexts: [],
|
||||
sourceDocuments: [],
|
||||
scores: [],
|
||||
};
|
||||
|
||||
const collection = await client.collection(namespace);
|
||||
const responses = await collection
|
||||
.find(
|
||||
{},
|
||||
{
|
||||
sort: { $vector: queryVector },
|
||||
limit: topN,
|
||||
includeSimilarity: true,
|
||||
}
|
||||
)
|
||||
.toArray();
|
||||
|
||||
responses.forEach((response) => {
|
||||
if (response.$similarity < similarityThreshold) return;
|
||||
result.contextTexts.push(response.metadata.text);
|
||||
result.sourceDocuments.push(response);
|
||||
result.scores.push(response.$similarity);
|
||||
});
|
||||
return result;
|
||||
},
|
||||
allNamespaces: async function (client) {
|
||||
try {
|
||||
let header = new Headers();
|
||||
header.append("Token", client?.httpClient?.applicationToken);
|
||||
header.append("Content-Type", "application/json");
|
||||
|
||||
let raw = JSON.stringify({
|
||||
findCollections: {},
|
||||
});
|
||||
|
||||
let requestOptions = {
|
||||
method: "POST",
|
||||
headers: header,
|
||||
body: raw,
|
||||
redirect: "follow",
|
||||
};
|
||||
|
||||
const call = await fetch(client?.httpClient?.baseUrl, requestOptions);
|
||||
const resp = await call?.text();
|
||||
const collections = resp ? JSON.parse(resp)?.status?.collections : [];
|
||||
return collections;
|
||||
} catch (e) {
|
||||
console.error("Astra::AllNamespace", e);
|
||||
return [];
|
||||
}
|
||||
},
|
||||
"namespace-stats": async function (reqBody = {}) {
|
||||
const { namespace = null } = reqBody;
|
||||
if (!namespace) throw new Error("namespace required");
|
||||
const { client } = await this.connect();
|
||||
if (!(await this.namespaceExists(client, namespace)))
|
||||
throw new Error("Namespace by that name does not exist.");
|
||||
const stats = await this.namespace(client, namespace);
|
||||
return stats
|
||||
? stats
|
||||
: { message: "No stats were able to be fetched from DB for namespace" };
|
||||
},
|
||||
"delete-namespace": async function (reqBody = {}) {
|
||||
const { namespace = null } = reqBody;
|
||||
const { client } = await this.connect();
|
||||
if (!(await this.namespaceExists(client, namespace)))
|
||||
throw new Error("Namespace by that name does not exist.");
|
||||
|
||||
const details = await this.namespace(client, namespace);
|
||||
await this.deleteVectorsInNamespace(client, namespace);
|
||||
return {
|
||||
message: `Namespace ${namespace} was deleted along with ${
|
||||
details?.vectorCount || "all"
|
||||
} vectors.`,
|
||||
};
|
||||
},
|
||||
curateSources: function (sources = []) {
|
||||
const documents = [];
|
||||
for (const source of sources) {
|
||||
if (Object.keys(source).length > 0) {
|
||||
const metadata = source.hasOwnProperty("metadata")
|
||||
? source.metadata
|
||||
: source;
|
||||
documents.push({
|
||||
...metadata,
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
return documents;
|
||||
},
|
||||
};
|
||||
|
||||
module.exports.AstraDB = AstraDB;
|
|
@ -207,9 +207,9 @@ const LanceDb = {
|
|||
|
||||
vectors.push(vectorRecord);
|
||||
submissions.push({
|
||||
...vectorRecord.metadata,
|
||||
id: vectorRecord.id,
|
||||
vector: vectorRecord.values,
|
||||
...vectorRecord.metadata,
|
||||
});
|
||||
documentVectors.push({ docId, vectorId: vectorRecord.id });
|
||||
}
|
||||
|
|
|
@ -174,6 +174,15 @@
|
|||
enabled "2.0.x"
|
||||
kuler "^2.0.0"
|
||||
|
||||
"@datastax/astra-db-ts@^0.1.3":
|
||||
version "0.1.3"
|
||||
resolved "https://registry.yarnpkg.com/@datastax/astra-db-ts/-/astra-db-ts-0.1.3.tgz#fcc25cda8d146c06278860054f09d687ff031568"
|
||||
integrity sha512-7lnpym0HhUtfJVd8+vu6vYdDQpFyYof7TVLFVD2fgoIjUwj3EksFXmqDqicLAlLferZDllqSVthX9pXQ5Rdapw==
|
||||
dependencies:
|
||||
axios "^1.4.0"
|
||||
bson "^6.2.0"
|
||||
winston "^3.7.2"
|
||||
|
||||
"@eslint-community/eslint-utils@^4.2.0":
|
||||
version "4.4.0"
|
||||
resolved "https://registry.yarnpkg.com/@eslint-community/eslint-utils/-/eslint-utils-4.4.0.tgz#a23514e8fb9af1269d5f7788aa556798d61c6b59"
|
||||
|
@ -1353,6 +1362,11 @@ braces@~3.0.2:
|
|||
dependencies:
|
||||
fill-range "^7.0.1"
|
||||
|
||||
bson@^6.2.0:
|
||||
version "6.2.0"
|
||||
resolved "https://registry.yarnpkg.com/bson/-/bson-6.2.0.tgz#4b6acafc266ba18eeee111373c2699304a9ba0a3"
|
||||
integrity sha512-ID1cI+7bazPDyL9wYy9GaQ8gEEohWvcUl/Yf0dIdutJxnmInEEyCsb4awy/OiBfall7zBA179Pahi3vCdFze3Q==
|
||||
|
||||
btoa-lite@^1.0.0:
|
||||
version "1.0.0"
|
||||
resolved "https://registry.yarnpkg.com/btoa-lite/-/btoa-lite-1.0.0.tgz#337766da15801210fdd956c22e9c6891ab9d0337"
|
||||
|
@ -5636,7 +5650,7 @@ winston-transport@^4.5.0:
|
|||
readable-stream "^3.6.0"
|
||||
triple-beam "^1.3.0"
|
||||
|
||||
winston@^3.9.0:
|
||||
winston@^3.7.2, winston@^3.9.0:
|
||||
version "3.11.0"
|
||||
resolved "https://registry.yarnpkg.com/winston/-/winston-3.11.0.tgz#2d50b0a695a2758bb1c95279f0a88e858163ed91"
|
||||
integrity sha512-L3yR6/MzZAOl0DsysUXHVjOwv8mKZ71TrA/41EIduGpOOV5LQVodqN+QdQ6BS6PJ/RdIshZhq84P/fStEZkk7g==
|
||||
|
|
Loading…
Reference in New Issue