Merge pull request #292 from weaviate/feature-custom-deployment

Add Custom Deployment

CHANGELOG.md

@@ -2,6 +2,17 @@
 
 All notable changes to this project will be documented in this file.
 
+## [2.1.0] Importastic
+
+## Added
+
+- Added new deployment type: Custom
+- Added new port configuration
+
+# Fixed
+
+- Catch Exception when trying to access the OpenAI API Embedding endpoint to retrieve model names
+
 ## [2.0.0] Importastic
 
 ## Added

README.md

@@ -42,7 +42,7 @@ Verba is a fully-customizable personal assistant utilizing [Retrieval Augmented
 
 ### Watch our newest Verba video here:
 
-[![VIDEO LINK](https://github.com/weaviate/Verba/blob/main/img/thumbnail.png)](https://www.youtube.com/watch?v=swKKRdLBhas&t)
+[![VIDEO LINK](https://github.com/weaviate/Verba/blob/main/img/thumbnail.png)](https://www.youtube.com/watch?v=2VCy-YjRRhA&t=40s&ab_channel=Weaviate%E2%80%A2VectorDatabase)
 
 ## Feature Lists
 
@@ -63,15 +63,15 @@ Verba is a fully-customizable personal assistant utilizing [Retrieval Augmented
 | VoyageAI             | ✅          | Embedding Models by VoyageAI             |
 | OpenAI               | ✅          | Embedding Models by OpenAI               |
 
-| 📁 Data Support                                         | Implemented | Description                                    |
-| ------------------------------------------------------- | ----------- | -----------------------------------------------|
-| [UnstructuredIO](https://docs.unstructured.io/welcome)  | ✅          | Import Data through Unstructured               |
-| [Firecrawl](https://www.firecrawl.dev/)                 | ✅          | Scrape and Crawl URL through Firecrawl         |
-| PDF Ingestion                                           | ✅          | Import PDF into Verba                          |
-| GitHub & GitLab                                         | ✅          | Import Files from Github and GitLab            |
-| CSV/XLSX Ingestion                                      | ✅          | Import Table Data into Verba                   |
-| .DOCX                                                   | ✅          | Import .docx files                             |
-| Multi-Modal (using [AssemblyAI](https://assemblyai.com))| ✅          | Import and Transcribe Audio through AssemblyAI |
+| 📁 Data Support                                          | Implemented | Description                                    |
+| -------------------------------------------------------- | ----------- | ---------------------------------------------- |
+| [UnstructuredIO](https://docs.unstructured.io/welcome)   | ✅          | Import Data through Unstructured               |
+| [Firecrawl](https://www.firecrawl.dev/)                  | ✅          | Scrape and Crawl URL through Firecrawl         |
+| PDF Ingestion                                            | ✅          | Import PDF into Verba                          |
+| GitHub & GitLab                                          | ✅          | Import Files from Github and GitLab            |
+| CSV/XLSX Ingestion                                       | ✅          | Import Table Data into Verba                   |
+| .DOCX                                                    | ✅          | Import .docx files                             |
+| Multi-Modal (using [AssemblyAI](https://assemblyai.com)) | ✅          | Import and Transcribe Audio through AssemblyAI |
 
 | ✨ RAG Features         | Implemented | Description                                                               |
 | ----------------------- | ----------- | ------------------------------------------------------------------------- |
@@ -213,8 +213,6 @@ Verba supports importing documents through Unstructured IO (e.g plain text, .pdf
 
 Verba supports importing documents through AssemblyAI (audio files or audio from video files). To use them you need the `ASSEMBLYAI_API_KEY` environment variable. You can get it from [AssemblyAI](https://assemblyai.com)
 
-
-
 ## OpenAI
 
 Verba supports OpenAI Models such as Ada, GPT3, and GPT4. To use them, you need to specify the `OPENAI_API_KEY` environment variable. You can get it from [OpenAI](https://openai.com/)
@@ -359,6 +357,10 @@ RUN pip install -e '.'
 
 ## Verba Walkthrough
 
+### Select your Deployment
+
+The first screen you'll see is the deployment screen. Here you can select between `Local`, `Docker`, `Weaviate Cloud`, or `Custom` deployment. The `Local` deployment is using Weaviate Embedded under the hood, which initializes a Weaviate instance behind the scenes. The `Docker` deployment is using a separate Weaviate instance that is running inside the same Docker network. The `Weaviate Cloud` deployment is using a Weaviate instance that is hosted on Weaviate Cloud Services (WCS). The `Custom` deployment allows you to specify your own Weaviate instance URL, PORT, and API key.
+
 ### Import Your Data
 
 First thing you need to do is to add your data. You can do this by clicking on `Import Data` and selecting either `Add Files`, `Add Directory`, or `Add URL` tab. Here you can add all your files that you want to ingest.

frontend/app/api.ts

@@ -78,7 +78,8 @@ export const fetchHealth = (): Promise<HealthPayload | null> =>
 export const connectToVerba = async (
   deployment: string,
   url: string,
-  apiKey: string
+  apiKey: string,
+  port: string
 ): Promise<ConnectPayload | null> => {
   const host = await detectHost();
   const response = await fetch(`${host}/api/connect`, {
@@ -92,6 +93,7 @@ export const connectToVerba = async (
         url: url,
         key: apiKey,
       },
+      port: port,
     }),
   });
   const data = await response.json();

frontend/app/components/Login/LoginView.tsx

@@ -19,6 +19,7 @@ import { GrConnect } from "react-icons/gr";
 import { CgWebsite } from "react-icons/cg";
 import { FaBackspace } from "react-icons/fa";
 import { HiMiniSparkles } from "react-icons/hi2";
+import { TbDatabaseEdit } from "react-icons/tb";
 
 import { connectToVerba } from "@/app/api";
 
@@ -161,11 +162,12 @@ const LoginView: React.FC<LoginViewProps> = ({
   const [errorText, setErrorText] = useState("");
 
   const [selectedDeployment, setSelectedDeployment] = useState<
-    "Weaviate" | "Docker" | "Local"
+    "Weaviate" | "Docker" | "Local" | "Custom"
   >("Local");
 
   const [weaviateURL, setWeaviateURL] = useState(credentials.url);
   const [weaviateAPIKey, setWeaviateAPIKey] = useState(credentials.key);
+  const [port, setPort] = useState("8080");
 
   useEffect(() => {
     const timer = setTimeout(() => {
@@ -175,18 +177,26 @@ const LoginView: React.FC<LoginViewProps> = ({
     return () => clearTimeout(timer);
   }, []);
 
-  const connect = async (deployment: "Local" | "Weaviate" | "Docker") => {
+  const connect = async (
+    deployment: "Local" | "Weaviate" | "Docker" | "Custom"
+  ) => {
     setErrorText("");
     setIsConnecting(true);
     const response = await connectToVerba(
       deployment,
       weaviateURL,
-      weaviateAPIKey
+      weaviateAPIKey,
+      port
     );
     if (response) {
-      if (response.error) {
+      if (!("error" in response)) {
         setIsLoggedIn(false);
-        setErrorText(response.error);
+        setErrorText(JSON.stringify(response));
+      } else if (response.connected == false) {
+        setIsLoggedIn(false);
+        setErrorText(
+          response.error == "" ? "Couldn't connect to Weaviate" : response.error
+        );
       } else {
         setIsLoggedIn(true);
         setCredentials({
@@ -287,6 +297,16 @@ const LoginView: React.FC<LoginViewProps> = ({
                       }}
                       loading={isConnecting && selectedDeployment == "Docker"}
                     />
+                    <VerbaButton
+                      title="Custom"
+                      Icon={TbDatabaseEdit}
+                      disabled={isConnecting}
+                      onClick={() => {
+                        setSelectedDeployment("Custom");
+                        setSelectStage(false);
+                      }}
+                      loading={isConnecting && selectedDeployment == "Custom"}
+                    />
                     <VerbaButton
                       title="Local"
                       Icon={FaLaptopCode}
@@ -336,18 +356,35 @@ const LoginView: React.FC<LoginViewProps> = ({
                         connect(selectedDeployment);
                       }}
                     >
-                      <label className="input flex items-center gap-2 border-none shadow-md bg-bg-verba">
-                        <FaDatabase className="text-text-alt-verba" />
-                        <input
-                          type="text"
-                          name="username"
-                          value={weaviateURL}
-                          onChange={(e) => setWeaviateURL(e.target.value)}
-                          placeholder="Weaviate URL"
-                          className="grow bg-button-verba text-text-alt-verba hover:text-text-verba w-full"
-                          autoComplete="username"
-                        />
-                      </label>
+                      <div className="flex gap-2 items-center justify-between">
+                        <label className="input flex items-center gap-2 border-none shadow-md w-full bg-bg-verba">
+                          <FaDatabase className="text-text-alt-verba" />
+                          <input
+                            type="text"
+                            name="username"
+                            value={weaviateURL}
+                            onChange={(e) => setWeaviateURL(e.target.value)}
+                            placeholder="Weaviate URL"
+                            className="grow bg-button-verba text-text-alt-verba hover:text-text-verba w-full"
+                            autoComplete="username"
+                          />
+                        </label>
+                        {selectedDeployment == "Custom" && (
+                          <label className="input flex items-center gap-2 border-none shadow-md bg-bg-verba">
+                            <p className="text-text-alt-verba text-xs">Port</p>
+                            <input
+                              type="text"
+                              name="Port"
+                              value={port}
+                              onChange={(e) => setPort(e.target.value)}
+                              placeholder="Port"
+                              className="grow bg-button-verba text-text-alt-verba hover:text-text-verba w-full"
+                              autoComplete="port"
+                            />
+                          </label>
+                        )}
+                      </div>
+
                       <label className="input flex items-center gap-2 border-none shadow-md bg-bg-verba mt-4">
                         <FaKey className="text-text-alt-verba" />
                         <input
@@ -371,18 +408,20 @@ const LoginView: React.FC<LoginViewProps> = ({
                               selected_color="bg-primary-verba"
                               loading={isConnecting}
                             />
-                            <VerbaButton
-                              Icon={CgWebsite}
-                              title="Register"
-                              type="button"
-                              disabled={isConnecting}
-                              onClick={() =>
-                                window.open(
-                                  "https://console.weaviate.cloud",
-                                  "_blank"
-                                )
-                              }
-                            />
+                            {selectedDeployment == "Weaviate" && (
+                              <VerbaButton
+                                Icon={CgWebsite}
+                                title="Register"
+                                type="button"
+                                disabled={isConnecting}
+                                onClick={() =>
+                                  window.open(
+                                    "https://console.weaviate.cloud",
+                                    "_blank"
+                                  )
+                                }
+                              />
+                            )}
                             <VerbaButton
                               Icon={FaBackspace}
                               title="Back"
@@ -402,8 +441,10 @@ const LoginView: React.FC<LoginViewProps> = ({
               </div>
             )}
             {errorText && (
-              <div className="bg-warning-verba p-4 rounded w-full">
-                <p className="flex w-full whitespace-pre-wrap">{errorText}</p>
+              <div className="bg-warning-verba p-4 rounded w-full h-full overflow-auto">
+                <p className="flex w-full h-full whitespace-pre-wrap">
+                  {errorText}
+                </p>
               </div>
             )}
           </div>

frontend/app/types.ts

@@ -1,5 +1,5 @@
 export type Credentials = {
-  deployment: "Weaviate" | "Docker" | "Local";
+  deployment: "Weaviate" | "Docker" | "Local" | "Custom";
   url: string;
   key: string;
 };

frontend/package.json

@@ -1,6 +1,6 @@
 {
   "name": "verba",
-  "version": "1.0.4",
+  "version": "2.1.0",
   "private": true,
   "scripts": {
     "dev": "next dev",

goldenverba/components/embedding/OpenAIEmbedder.py

@@ -104,20 +104,28 @@ async def vectorize(self, config: dict, content: List[str]) -> List[List[float]]
     @staticmethod
     def get_models(token: str, url: str) -> List[str]:
         """Fetch available embedding models from OpenAI API."""
-        if token is None:
+        try:
+            if token is None:
+                return [
+                    "text-embedding-ada-002",
+                    "text-embedding-3-small",
+                    "text-embedding-3-large",
+                ]
+
+            import requests  # Import here to avoid dependency if not needed
+
+            headers = {"Authorization": f"Bearer {token}"}
+            response = requests.get(f"{url}/models", headers=headers)
+            response.raise_for_status()
+            return [
+                model["id"]
+                for model in response.json()["data"]
+                if "embedding" in model["id"]
+            ]
+        except Exception as e:
+            msg.info(f"Failed to fetch OpenAI embedding models: {str(e)}")
             return [
                 "text-embedding-ada-002",
                 "text-embedding-3-small",
                 "text-embedding-3-large",
             ]
-
-        import requests  # Import here to avoid dependency if not needed
-
-        headers = {"Authorization": f"Bearer {token}"}
-        response = requests.get(f"{url}/models", headers=headers)
-        response.raise_for_status()
-        return [
-            model["id"]
-            for model in response.json()["data"]
-            if "embedding" in model["id"]
-        ]

goldenverba/components/managers.py

@@ -175,32 +175,32 @@ async def connect_to_docker(self, w_url):
             ),
         )
 
-    async def connect_to_custom(self, host, w_key):
+    async def connect_to_custom(self, host, w_key, port):
         # Extract the port from the host
-        parsed_url = urlparse(host)
-        port = parsed_url.port
-        if port is None:
-            raise Exception("No port specified in the host URL")
-        _host = parsed_url.hostname  # Use only the hostname part
         msg.info(f"Connecting to Weaviate Custom")
 
+        if host is None or host == "":
+            raise Exception("No Host URL provided")
+
         if w_key is None or w_key == "":
             return weaviate.use_async_with_local(
-                host=_host,
-                port=port,
+                host=host,
+                port=int(port),
+                skip_init_checks=True,
+                additional_config=AdditionalConfig(
+                    timeout=Timeout(init=60, query=300, insert=300)
+                ),
+            )
+        else:
+            return weaviate.use_async_with_local(
+                host=host,
+                port=int(port),
+                skip_init_checks=True,
+                auth_credentials=AuthApiKey(w_key),
                 additional_config=AdditionalConfig(
                     timeout=Timeout(init=60, query=300, insert=300)
                 ),
             )
-
-        return weaviate.use_async_with_custom(
-            http_host=_host,
-            http_port=port,
-            auth_credentials=AuthApiKey(w_key),
-            additional_config=AdditionalConfig(
-                timeout=Timeout(init=60, query=300, insert=300)
-            ),
-        )
 
     async def connect_to_embedded(self):
         msg.info(f"Connecting to Weaviate Embedded")
@@ -211,7 +211,7 @@ async def connect_to_embedded(self):
         )
 
     async def connect(
-        self, deployment: str, weaviateURL: str, weaviateAPIKey: str
+        self, deployment: str, weaviateURL: str, weaviateAPIKey: str, port: str = "8080"
     ) -> WeaviateAsyncClient:
         try:
 
@@ -226,7 +226,7 @@ async def connect(
             elif deployment == "Local":
                 client = await self.connect_to_embedded()
             elif deployment == "Custom":
-                client = await self.connect_to_custom(weaviateURL, weaviateAPIKey)
+                client = await self.connect_to_custom(weaviateURL, weaviateAPIKey, port)
             else:
                 raise Exception(f"Invalid deployment type: {deployment}")