Merge branch 'master' of github.com:debanjum/khoj into features/simplify-configuration-steps

Khoj.desktop

@@ -1,7 +1,7 @@
 [Desktop Entry]
 Type=Application
 Name=Khoj
-Comment=A natural language search engine for your personal notes, transactions and images.
+Comment=An AI personal assistant for your Digital Brain
 Path=/opt
 Exec=/opt/Khoj
 Icon=Khoj

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "khoj-assistant"
-description = "A natural language search engine for your personal notes, transactions and images"
+description = "An AI personal assistant for your Digital Brain"
 readme = "README.md"
 license = "GPL-3.0-or-later"
 requires-python = ">=3.8"

src/interface/obsidian/README.md

@@ -1,6 +1,6 @@
 <img src="/src/khoj/interface/web/assets/icons/khoj-logo-sideways.svg" width="200" alt="Khoj Logo">Obsidian
 
-> Natural language search for your Obsidian notes using [Khoj](https://github.com/khoj-ai/khoj)
+> An AI personal assistant for your Digital Brain in Obsidian
 
 ## Table of Contents
 

src/interface/obsidian/src/search_modal.ts

@@ -161,7 +161,7 @@ export class KhojSearchModal extends SuggestModal<SearchResult> {
         // Open vault file at heading of chosen search result
         if (file_match) {
             let resultHeading = file_match.extension !== 'pdf' ? result.entry.split('\n', 1)[0] : '';
-            let linkToEntry = `${file_match.path}${resultHeading}`
+            let linkToEntry = resultHeading.startsWith('#') ? `${file_match.path}${resultHeading}` : file_match.path;
             this.app.workspace.openLinkText(linkToEntry, '');
             console.log(`Link: ${linkToEntry}, File: ${file_match.path}, Heading: ${resultHeading}`);
         }

src/interface/obsidian/src/utils.ts

@@ -72,7 +72,10 @@ export async function configureKhojBackend(vault: Vault, setting: KhojSetting, n
                 }
             }
             // Else if khoj is not configured to index markdown files in configured obsidian vault
-            else if (data["content-type"]["markdown"]["input-filter"].length != 1 ||
+            else if (
+                data["content-type"]["markdown"]["input-files"] != null ||
+                data["content-type"]["markdown"]["input-filter"] == null ||
+                data["content-type"]["markdown"]["input-filter"].length != 1 ||
                 data["content-type"]["markdown"]["input-filter"][0] !== mdInVault) {
                     // Update markdown config in khoj content-type config
                     // Set markdown config to only index markdown files in configured obsidian vault
@@ -86,19 +89,30 @@ export async function configureKhojBackend(vault: Vault, setting: KhojSetting, n
             }
 
             if (khoj_already_configured && !data["content-type"]["pdf"]) {
-                // Add pdf config to khoj content-type config
+                const hasPdfFiles = app.vault.getFiles().some(file => file.extension === 'pdf');
-                // Set pdf config to index pdf files in configured obsidian vault
+
+                if (hasPdfFiles) {
                     data["content-type"]["pdf"] = {
                         "input-filter": [pdfInVault],
                         "input-files": null,
                         "embeddings-file": `${khojDefaultPdfIndexDirectory}/${indexName}.pt`,
                         "compressed-jsonl": `${khojDefaultPdfIndexDirectory}/${indexName}.jsonl.gz`,
                     }
+                } else {
+                    data["content-type"]["pdf"] = null;
+                }
             }
             // Else if khoj is not configured to index pdf files in configured obsidian vault
             else if (khoj_already_configured &&
-                (data["content-type"]["pdf"]["input-filter"].length != 1 ||
+                (
+                    data["content-type"]["pdf"]["input-files"] != null ||
+                    data["content-type"]["pdf"]["input-filter"] == null ||
+                    data["content-type"]["pdf"]["input-filter"].length != 1 ||
                     data["content-type"]["pdf"]["input-filter"][0] !== pdfInVault)) {
+
+                let hasPdfFiles = app.vault.getFiles().some(file => file.extension === 'pdf');
+
+                if (hasPdfFiles) {
                     // Update pdf config in khoj content-type config
                     // Set pdf config to only index pdf files in configured obsidian vault
                     let khojPdfIndexDirectory = getIndexDirectoryFromBackendConfig(data["content-type"]["pdf"]["embeddings-file"]);
@@ -108,6 +122,9 @@ export async function configureKhojBackend(vault: Vault, setting: KhojSetting, n
                         "embeddings-file": `${khojPdfIndexDirectory}/${indexName}.pt`,
                         "compressed-jsonl": `${khojPdfIndexDirectory}/${indexName}.jsonl.gz`,
                     }
+                } else {
+                    data["content-type"]["pdf"] = null;
+                }
             }
 
             // If OpenAI API key not set in Khoj plugin settings

src/khoj/configure.py

@@ -93,6 +93,7 @@ def configure_search(model: SearchModels, config: FullConfig, regenerate: bool,
         logger.warning("🚨 No Content or Search type is configured.")
         return
 
+    try:
         # Initialize Org Notes Search
         if (t == state.SearchType.Org or t == None) and config.content_type.org and config.search_type.asymmetric:
             logger.info("🦄 Setting up search for orgmode notes")
@@ -118,7 +119,11 @@ def configure_search(model: SearchModels, config: FullConfig, regenerate: bool,
             )
 
         # Initialize Markdown Search
-    if (t == state.SearchType.Markdown or t == None) and config.content_type.markdown and config.search_type.asymmetric:
+        if (
+            (t == state.SearchType.Markdown or t == None)
+            and config.content_type.markdown
+            and config.search_type.asymmetric
+        ):
             logger.info("💎 Setting up search for markdown notes")
             # Extract Entries, Generate Markdown Embeddings
             model.markdown_search = text_search.setup(
@@ -184,6 +189,9 @@ def configure_search(model: SearchModels, config: FullConfig, regenerate: bool,
                     regenerate=regenerate,
                     filters=[DateFilter(), WordFilter(), FileFilter()],
                 )
+    except Exception as e:
+        logger.error("🚨 Failed to setup search")
+        raise e
 
     # Invalidate Query Cache
     state.query_cache = LRU()

src/khoj/routers/api.py

@@ -384,7 +384,12 @@ def update(
 ):
     try:
         state.search_index_lock.acquire()
+        try:
             state.model = configure_search(state.model, state.config, regenerate=force or False, t=t)
+        except Exception as e:
+            logger.error(e)
+            raise HTTPException(status_code=500, detail=str(e))
+        finally:
             state.search_index_lock.release()
     except ValueError as e:
         logger.error(e)

src/khoj/utils/cli.py

@@ -10,9 +10,7 @@ from khoj.utils.yaml import parse_config_from_file
 
 def cli(args=None):
     # Setup Argument Parser for the Commandline Interface
-    parser = argparse.ArgumentParser(
+    parser = argparse.ArgumentParser(description="Start Khoj; An AI personal assistant for your Digital Brain")
-        description="Start Khoj; A Natural Language Search Engine for your personal Notes, Transactions and Photos"
-    )
     parser.add_argument(
         "--config-file", "-c", default="~/.khoj/khoj.yml", type=pathlib.Path, help="YAML file to configure Khoj"
     )