Add Offline chat to Obsidian (#359)

* Add support for configuring/using offline chat from within Obsidian
* Fix type checking for search type
* If Github is not configured, /update call should fail
* Fix regenerate tests same as the update ones
* Update help text for offline chat in obsidian
* Update relevant description for Khoj settings in Obsidian
* Simplify configuration logic and use smarter defaults

src/interface/obsidian/src/main.ts

@@ -38,9 +38,9 @@ export default class Khoj extends Plugin {
             id: 'chat',
             name: 'Chat',
             checkCallback: (checking) => {
-                if (!checking && this.settings.connectedToBackend && !!this.settings.openaiApiKey)
+                if (!checking && this.settings.connectedToBackend && (!!this.settings.openaiApiKey || this.settings.enableOfflineChat))
                     new KhojChatModal(this.app, this.settings).open();
-                return !!this.settings.openaiApiKey;
+                return !!this.settings.openaiApiKey || this.settings.enableOfflineChat;
             }
         });
 

src/interface/obsidian/src/settings.ts

@@ -2,6 +2,7 @@ import { App, Notice, PluginSettingTab, request, Setting } from 'obsidian';
 import Khoj from 'src/main';
 
 export interface KhojSetting {
+    enableOfflineChat: boolean;
     openaiApiKey: string;
     resultsCount: number;
     khojUrl: string;
@@ -10,6 +11,7 @@ export interface KhojSetting {
 }
 
 export const DEFAULT_SETTINGS: KhojSetting = {
+    enableOfflineChat: false,
     resultsCount: 6,
     khojUrl: 'http://127.0.0.1:42110',
     connectedToBackend: false,
@@ -35,7 +37,7 @@ export class KhojSettingTab extends PluginSettingTab {
         // Add khoj settings configurable from the plugin settings tab
         new Setting(containerEl)
             .setName('Khoj URL')
-            .setDesc('The URL of the Khoj backend')
+            .setDesc('The URL of the Khoj backend.')
             .addText(text => text
                 .setValue(`${this.plugin.settings.khojUrl}`)
                 .onChange(async (value) => {
@@ -45,16 +47,25 @@ export class KhojSettingTab extends PluginSettingTab {
                 }));
         new Setting(containerEl)
             .setName('OpenAI API Key')
-            .setDesc('Your OpenAI API Key for Khoj Chat')
+            .setDesc('Use OpenAI for Khoj Chat with your API key.')
             .addText(text => text
                 .setValue(`${this.plugin.settings.openaiApiKey}`)
                 .onChange(async (value) => {
                     this.plugin.settings.openaiApiKey = value.trim();
                     await this.plugin.saveSettings();
                 }));
+        new Setting(containerEl)
+            .setName('Enable Offline Chat')
+            .setDesc('Chat privately without an internet connection. Enabling this will use offline chat even if OpenAI is configured.')
+            .addToggle(toggle => toggle
+                .setValue(this.plugin.settings.enableOfflineChat)
+                .onChange(async (value) => {
+                    this.plugin.settings.enableOfflineChat = value;
+                    await this.plugin.saveSettings();
+                }));
         new Setting(containerEl)
             .setName('Results Count')
-            .setDesc('The number of results to show in search and use for chat')
+            .setDesc('The number of results to show in search and use for chat.')
             .addSlider(slider => slider
                 .setLimits(1, 10, 1)
                 .setValue(this.plugin.settings.resultsCount)
@@ -65,7 +76,7 @@ export class KhojSettingTab extends PluginSettingTab {
                 }));
         new Setting(containerEl)
             .setName('Auto Configure')
-            .setDesc('Automatically configure the Khoj backend')
+            .setDesc('Automatically configure the Khoj backend.')
             .addToggle(toggle => toggle
                 .setValue(this.plugin.settings.autoConfigure)
                 .onChange(async (value) => {
@@ -75,7 +86,7 @@ export class KhojSettingTab extends PluginSettingTab {
         let indexVaultSetting = new Setting(containerEl);
         indexVaultSetting
             .setName('Index Vault')
-            .setDesc('Manually force Khoj to re-index your Obsidian Vault')
+            .setDesc('Manually force Khoj to re-index your Obsidian Vault.')
             .addButton(button => button
                 .setButtonText('Update')
                 .setCta()

src/interface/obsidian/src/utils.ts

@@ -9,6 +9,19 @@ export function getVaultAbsolutePath(vault: Vault): string {
     return '';
 }
 
+type OpenAIType = null | {
+    "chat-model": string;
+    "api-key": string;
+};
+
+interface ProcessorData {
+    conversation: {
+      "conversation-logfile": string;
+      openai: OpenAIType;
+      "enable-offline-chat": boolean;
+    };
+}
+
 export async function configureKhojBackend(vault: Vault, setting: KhojSetting, notify: boolean = true) {
     let vaultPath = getVaultAbsolutePath(vault);
     let mdInVault = `${vaultPath}/**/*.md`;
@@ -132,48 +145,36 @@ export async function configureKhojBackend(vault: Vault, setting: KhojSetting, n
                 }
             }
 
-            // If OpenAI API key not set in Khoj plugin settings
+            let conversationLogFile = data?.["processor"]?.["conversation"]?.["conversation-logfile"] ?? `${khojDefaultChatDirectory}/conversation.json`;
-            if (!setting.openaiApiKey) {
+
-                // Disable khoj processors, as not required
+            let processorData: ProcessorData = {
-                delete data["processor"];
+                "conversation": {
+                    "conversation-logfile": conversationLogFile,
+                    "openai": null,
+                    "enable-offline-chat": setting.enableOfflineChat,
+                }
             }
-            // Else if khoj backend not configured yet
+
-            else if (!khoj_already_configured || !data["processor"]) {
+            // If the Open AI API Key was configured in the plugin settings
-                data["processor"] = {
+            if (!!setting.openaiApiKey) {
+
+                let openAIChatModel = data?.["processor"]?.["conversation"]?.["openai"]?.["chat-model"] ?? khojDefaultChatModelName;
+
+                processorData = {
                     "conversation": {
-                        "conversation-logfile": `${khojDefaultChatDirectory}/conversation.json`,
+                        "conversation-logfile": conversationLogFile,
                         "openai": {
-                            "chat-model": khojDefaultChatModelName,
+                            "chat-model": openAIChatModel,
                             "api-key": setting.openaiApiKey,
-                        }
+                        },
-                    },
+                        "enable-offline-chat": setting.enableOfflineChat,
-                }
-            }
-            // Else if khoj config has no conversation processor config
-            else if (!data["processor"]["conversation"] || !data["processor"]["conversation"]["openai"]) {
-                data["processor"] = {
-                    "conversation": {
-                        "conversation-logfile": `${khojDefaultChatDirectory}/conversation.json`,
-                        "openai": {
-                            "chat-model": khojDefaultChatModelName,
-                            "api-key": setting.openaiApiKey,
-                        }
-                    },
-                }
-            }
-            // Else if khoj is not configured with OpenAI API key from khoj plugin settings
-            else if (data["processor"]["conversation"]["openai"]["api-key"] !== setting.openaiApiKey) {
-                data["processor"] = {
-                    "conversation": {
-                        "conversation-logfile": data["processor"]["conversation"]["conversation-logfile"],
-                        "openai": {
-                            "chat-model": data["processor"]["conversation"]["openai"]["chat-model"],
-                            "api-key": setting.openaiApiKey,
-                        }
                     },
                 }
             }
 
+            // Set khoj processor config to conversation processor config
+            data["processor"] = processorData;
+
             // Save updated config and refresh index on khoj backend
             updateKhojBackend(setting.khojUrl, data);
             if (!khoj_already_configured)

src/khoj/configure.py

@@ -169,7 +169,7 @@ def configure_content(
 
     try:
         # Initialize Org Notes Search
-        if (t == state.SearchType.Org or t == None) and content_config.org and search_models.text_search:
+        if (t == None or t.value == state.SearchType.Org.value) and content_config.org and search_models.text_search:
             logger.info("🦄 Setting up search for orgmode notes")
             # Extract Entries, Generate Notes Embeddings
             content_index.org = text_search.setup(
@@ -181,7 +181,11 @@ def configure_content(
             )
 
         # Initialize Markdown Search
-        if (t == state.SearchType.Markdown or t == None) and content_config.markdown and search_models.text_search:
+        if (
+            (t == None or t.value == state.SearchType.Markdown.value)
+            and content_config.markdown
+            and search_models.text_search
+        ):
             logger.info("💎 Setting up search for markdown notes")
             # Extract Entries, Generate Markdown Embeddings
             content_index.markdown = text_search.setup(
@@ -193,7 +197,7 @@ def configure_content(
             )
 
         # Initialize PDF Search
-        if (t == state.SearchType.Pdf or t == None) and content_config.pdf and search_models.text_search:
+        if (t == None or t.value == state.SearchType.Pdf.value) and content_config.pdf and search_models.text_search:
             logger.info("🖨️ Setting up search for pdf")
             # Extract Entries, Generate PDF Embeddings
             content_index.pdf = text_search.setup(
@@ -205,14 +209,22 @@ def configure_content(
             )
 
         # Initialize Image Search
-        if (t == state.SearchType.Image or t == None) and content_config.image and search_models.image_search:
+        if (
+            (t == None or t.value == state.SearchType.Image.value)
+            and content_config.image
+            and search_models.image_search
+        ):
             logger.info("🌄 Setting up search for images")
             # Extract Entries, Generate Image Embeddings
             content_index.image = image_search.setup(
                 content_config.image, search_models.image_search.image_encoder, regenerate=regenerate
             )
 
-        if (t == state.SearchType.Github or t == None) and content_config.github and search_models.text_search:
+        if (
+            (t == None or t.value == state.SearchType.Github.value)
+            and content_config.github
+            and search_models.text_search
+        ):
             logger.info("🐙 Setting up search for github")
             # Extract Entries, Generate Github Embeddings
             content_index.github = text_search.setup(
@@ -223,6 +235,21 @@ def configure_content(
                 filters=[DateFilter(), WordFilter(), FileFilter()],
             )
 
+        # Initialize Notion Search
+        if (
+            (t == None or t.value in state.SearchType.Notion.value)
+            and content_config.notion
+            and search_models.text_search
+        ):
+            logger.info("🔌 Setting up search for notion")
+            content_index.notion = text_search.setup(
+                NotionToJsonl,
+                content_config.notion,
+                search_models.text_search.bi_encoder,
+                regenerate=regenerate,
+                filters=[DateFilter(), WordFilter(), FileFilter()],
+            )
+
         # Initialize External Plugin Search
         if (t == None or t in state.SearchType) and content_config.plugins and search_models.text_search:
             logger.info("🔌 Setting up search for plugins")
@@ -236,17 +263,6 @@ def configure_content(
                     filters=[DateFilter(), WordFilter(), FileFilter()],
                 )
 
-        # Initialize Notion Search
-        if (t == None or t in state.SearchType) and content_config.notion and search_models.text_search:
-            logger.info("🔌 Setting up search for notion")
-            content_index.notion = text_search.setup(
-                NotionToJsonl,
-                content_config.notion,
-                search_models.text_search.bi_encoder,
-                regenerate=regenerate,
-                filters=[DateFilter(), WordFilter(), FileFilter()],
-            )
-
     except Exception as e:
         logger.error(f"🚨 Failed to setup search: {e}", exc_info=True)
         raise e

src/khoj/processor/github/github_to_jsonl.py

@@ -38,6 +38,9 @@ class GithubToJsonl(TextToJsonl):
             return
 
     def process(self, previous_entries=[]):
+        if self.config.pat_token is None or self.config.pat_token == "":
+            logger.error(f"Github PAT token is not set. Skipping github content")
+            raise ValueError("Github PAT token is not set. Skipping github content")
         current_entries = []
         for repo in self.config.repos:
             current_entries += self.process_repo(repo)

tests/test_client.py

@@ -53,13 +53,26 @@ def test_update_with_invalid_content_type(client):
 
 # ----------------------------------------------------------------------------------------------------
 def test_update_with_valid_content_type(client):
-    for content_type in ["all", "org", "markdown", "image", "pdf", "github", "notion", "plugin1"]:
+    for content_type in ["all", "org", "markdown", "image", "pdf", "notion", "plugin1"]:
         # Act
         response = client.get(f"/api/update?t={content_type}")
         # Assert
         assert response.status_code == 200, f"Returned status: {response.status_code} for content type: {content_type}"
 
 
+# ----------------------------------------------------------------------------------------------------
+def test_update_with_github_fails_without_pat(client):
+    # Act
+    response = client.get(f"/api/update?t=github")
+
+    # Assert
+    assert response.status_code == 500, f"Returned status: {response.status_code} for content type: github"
+    assert (
+        response.json()["detail"]
+        == "🚨 Failed to update server via API: Github PAT token is not set. Skipping github content"
+    )
+
+
 # ----------------------------------------------------------------------------------------------------
 def test_regenerate_with_invalid_content_type(client):
     # Act
@@ -71,13 +84,26 @@ def test_regenerate_with_invalid_content_type(client):
 
 # ----------------------------------------------------------------------------------------------------
 def test_regenerate_with_valid_content_type(client):
-    for content_type in ["all", "org", "markdown", "image", "pdf", "github", "notion", "plugin1"]:
+    for content_type in ["all", "org", "markdown", "image", "pdf", "notion", "plugin1"]:
         # Act
         response = client.get(f"/api/update?force=true&t={content_type}")
         # Assert
         assert response.status_code == 200, f"Returned status: {response.status_code} for content type: {content_type}"
 
 
+# ----------------------------------------------------------------------------------------------------
+def test_regenerate_with_github_fails_without_pat(client):
+    # Act
+    response = client.get(f"/api/update?force=true&t=github")
+
+    # Assert
+    assert response.status_code == 500, f"Returned status: {response.status_code} for content type: github"
+    assert (
+        response.json()["detail"]
+        == "🚨 Failed to update server via API: Github PAT token is not set. Skipping github content"
+    )
+
+
 # ----------------------------------------------------------------------------------------------------
 def test_get_configured_types_via_api(client):
     # Act