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

2024-11-23 23:48:56 +01:00 · 2023-07-02 16:21:54 -07:00 · 2023-07-02 16:21:54 -07:00 · a8b83da872
commit a8b83da872
parent 1a1b044d12 77a45f4215
8 changed files with 151 additions and 123 deletions
--- a/Khoj.desktop
+++ b/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
--- a/pyproject.toml
+++ b/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"
--- a/src/interface/obsidian/README.md
+++ b/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

--- a/src/interface/obsidian/src/search_modal.ts
+++ b/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: ${resultHeading}`);
        }
--- a/src/interface/obsidian/src/utils.ts
+++ b/src/interface/obsidian/src/utils.ts
@ -72,41 +72,58 @@ export async function configureKhojBackend(vault: Vault, setting: KhojSetting, n
                }
            }
            // Else if khoj is not configured to index markdown files in configured obsidian vault
-            else if (data["content-type"]["markdown"]["input-filter"].length != 1 ||
+            else if (
+                data["content-type"]["markdown"]["input-files"] != null ||
+                data["content-type"]["markdown"]["input-filter"] == null ||
+                data["content-type"]["markdown"]["input-filter"].length != 1 ||
                data["content-type"]["markdown"]["input-filter"][0] !== mdInVault) {
-                // Update markdown config in khoj content-type config
-                // Set markdown config to only index markdown files in configured obsidian vault
-                let khojMdIndexDirectory = getIndexDirectoryFromBackendConfig(data["content-type"]["markdown"]["embeddings-file"]);
-                data["content-type"]["markdown"] = {
-                    "input-filter": [mdInVault],
-                    "input-files": null,
-                    "embeddings-file": `${khojMdIndexDirectory}/${indexName}.pt`,
-                    "compressed-jsonl": `${khojMdIndexDirectory}/${indexName}.jsonl.gz`,
-                }
+                    // Update markdown config in khoj content-type config
+                    // Set markdown config to only index markdown files in configured obsidian vault
+                    let khojMdIndexDirectory = getIndexDirectoryFromBackendConfig(data["content-type"]["markdown"]["embeddings-file"]);
+                    data["content-type"]["markdown"] = {
+                        "input-filter": [mdInVault],
+                        "input-files": null,
+                        "embeddings-file": `${khojMdIndexDirectory}/${indexName}.pt`,
+                        "compressed-jsonl": `${khojMdIndexDirectory}/${indexName}.jsonl.gz`,
+                    }
            }

            if (khoj_already_configured && !data["content-type"]["pdf"]) {
-                // Add pdf config to khoj content-type config
-                // Set pdf config to index pdf files in configured obsidian vault
-                data["content-type"]["pdf"] = {
-                    "input-filter": [pdfInVault],
-                    "input-files": null,
-                    "embeddings-file": `${khojDefaultPdfIndexDirectory}/${indexName}.pt`,
-                    "compressed-jsonl": `${khojDefaultPdfIndexDirectory}/${indexName}.jsonl.gz`,
+                const hasPdfFiles = app.vault.getFiles().some(file => file.extension === 'pdf');
+
+                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-filter"][0] !== pdfInVault)) {
-                // 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"]);
-                data["content-type"]["pdf"] = {
-                    "input-filter": [pdfInVault],
-                    "input-files": null,
-                    "embeddings-file": `${khojPdfIndexDirectory}/${indexName}.pt`,
-                    "compressed-jsonl": `${khojPdfIndexDirectory}/${indexName}.jsonl.gz`,
+                (
+                    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"]);
+                    data["content-type"]["pdf"] = {
+                        "input-filter": [pdfInVault],
+                        "input-files": null,
+                        "embeddings-file": `${khojPdfIndexDirectory}/${indexName}.pt`,
+                        "compressed-jsonl": `${khojPdfIndexDirectory}/${indexName}.jsonl.gz`,
+                    }
+                } else {
+                    data["content-type"]["pdf"] = null;
                }
            }

--- a/src/khoj/configure.py
+++ b/src/khoj/configure.py
@ -93,98 +93,106 @@ def configure_search(model: SearchModels, config: FullConfig, regenerate: bool,
        logger.warning("🚨 No Content or Search type is configured.")
        return

-    # 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")
-        # Extract Entries, Generate Notes Embeddings
-        model.org_search = text_search.setup(
-            OrgToJsonl,
-            config.content_type.org,
-            search_config=config.search_type.asymmetric,
-            regenerate=regenerate,
-            filters=[DateFilter(), WordFilter(), FileFilter()],
-        )
-
-    # Initialize Org Music Search
-    if (t == state.SearchType.Music or t == None) and config.content_type.music and config.search_type.asymmetric:
-        logger.info("🎺 Setting up search for org-music")
-        # Extract Entries, Generate Music Embeddings
-        model.music_search = text_search.setup(
-            OrgToJsonl,
-            config.content_type.music,
-            search_config=config.search_type.asymmetric,
-            regenerate=regenerate,
-            filters=[DateFilter(), WordFilter()],
-        )
-
-    # Initialize Markdown Search
-    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(
-            MarkdownToJsonl,
-            config.content_type.markdown,
-            search_config=config.search_type.asymmetric,
-            regenerate=regenerate,
-            filters=[DateFilter(), WordFilter(), FileFilter()],
-        )
-
-    # Initialize Ledger Search
-    if (t == state.SearchType.Ledger or t == None) and config.content_type.ledger and config.search_type.symmetric:
-        logger.info("💸 Setting up search for ledger")
-        # Extract Entries, Generate Ledger Embeddings
-        model.ledger_search = text_search.setup(
-            BeancountToJsonl,
-            config.content_type.ledger,
-            search_config=config.search_type.symmetric,
-            regenerate=regenerate,
-            filters=[DateFilter(), WordFilter(), FileFilter()],
-        )
-
-    # Initialize PDF Search
-    if (t == state.SearchType.Pdf or t == None) and config.content_type.pdf and config.search_type.asymmetric:
-        logger.info("🖨️ Setting up search for pdf")
-        # Extract Entries, Generate PDF Embeddings
-        model.pdf_search = text_search.setup(
-            PdfToJsonl,
-            config.content_type.pdf,
-            search_config=config.search_type.asymmetric,
-            regenerate=regenerate,
-            filters=[DateFilter(), WordFilter(), FileFilter()],
-        )
-
-    # Initialize Image Search
-    if (t == state.SearchType.Image or t == None) and config.content_type.image and config.search_type.image:
-        logger.info("🌄 Setting up search for images")
-        # Extract Entries, Generate Image Embeddings
-        model.image_search = image_search.setup(
-            config.content_type.image, search_config=config.search_type.image, regenerate=regenerate
-        )
-
-    if (t == state.SearchType.Github or t == None) and config.content_type.github and config.search_type.asymmetric:
-        logger.info("🐙 Setting up search for github")
-        # Extract Entries, Generate Github Embeddings
-        model.github_search = text_search.setup(
-            GithubToJsonl,
-            config.content_type.github,
-            search_config=config.search_type.asymmetric,
-            regenerate=regenerate,
-            filters=[DateFilter(), WordFilter(), FileFilter()],
-        )
-
-    # Initialize External Plugin Search
-    if (t == None or t in state.SearchType) and config.content_type.plugins:
-        logger.info("🔌 Setting up search for plugins")
-        model.plugin_search = {}
-        for plugin_type, plugin_config in config.content_type.plugins.items():
-            model.plugin_search[plugin_type] = text_search.setup(
-                JsonlToJsonl,
-                plugin_config,
+    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")
+            # Extract Entries, Generate Notes Embeddings
+            model.org_search = text_search.setup(
+                OrgToJsonl,
+                config.content_type.org,
                search_config=config.search_type.asymmetric,
                regenerate=regenerate,
                filters=[DateFilter(), WordFilter(), FileFilter()],
            )

+        # Initialize Org Music Search
+        if (t == state.SearchType.Music or t == None) and config.content_type.music and config.search_type.asymmetric:
+            logger.info("🎺 Setting up search for org-music")
+            # Extract Entries, Generate Music Embeddings
+            model.music_search = text_search.setup(
+                OrgToJsonl,
+                config.content_type.music,
+                search_config=config.search_type.asymmetric,
+                regenerate=regenerate,
+                filters=[DateFilter(), WordFilter()],
+            )
+
+        # Initialize Markdown Search
+        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(
+                MarkdownToJsonl,
+                config.content_type.markdown,
+                search_config=config.search_type.asymmetric,
+                regenerate=regenerate,
+                filters=[DateFilter(), WordFilter(), FileFilter()],
+            )
+
+        # Initialize Ledger Search
+        if (t == state.SearchType.Ledger or t == None) and config.content_type.ledger and config.search_type.symmetric:
+            logger.info("💸 Setting up search for ledger")
+            # Extract Entries, Generate Ledger Embeddings
+            model.ledger_search = text_search.setup(
+                BeancountToJsonl,
+                config.content_type.ledger,
+                search_config=config.search_type.symmetric,
+                regenerate=regenerate,
+                filters=[DateFilter(), WordFilter(), FileFilter()],
+            )
+
+        # Initialize PDF Search
+        if (t == state.SearchType.Pdf or t == None) and config.content_type.pdf and config.search_type.asymmetric:
+            logger.info("🖨️ Setting up search for pdf")
+            # Extract Entries, Generate PDF Embeddings
+            model.pdf_search = text_search.setup(
+                PdfToJsonl,
+                config.content_type.pdf,
+                search_config=config.search_type.asymmetric,
+                regenerate=regenerate,
+                filters=[DateFilter(), WordFilter(), FileFilter()],
+            )
+
+        # Initialize Image Search
+        if (t == state.SearchType.Image or t == None) and config.content_type.image and config.search_type.image:
+            logger.info("🌄 Setting up search for images")
+            # Extract Entries, Generate Image Embeddings
+            model.image_search = image_search.setup(
+                config.content_type.image, search_config=config.search_type.image, regenerate=regenerate
+            )
+
+        if (t == state.SearchType.Github or t == None) and config.content_type.github and config.search_type.asymmetric:
+            logger.info("🐙 Setting up search for github")
+            # Extract Entries, Generate Github Embeddings
+            model.github_search = text_search.setup(
+                GithubToJsonl,
+                config.content_type.github,
+                search_config=config.search_type.asymmetric,
+                regenerate=regenerate,
+                filters=[DateFilter(), WordFilter(), FileFilter()],
+            )
+
+        # Initialize External Plugin Search
+        if (t == None or t in state.SearchType) and config.content_type.plugins:
+            logger.info("🔌 Setting up search for plugins")
+            model.plugin_search = {}
+            for plugin_type, plugin_config in config.content_type.plugins.items():
+                model.plugin_search[plugin_type] = text_search.setup(
+                    JsonlToJsonl,
+                    plugin_config,
+                    search_config=config.search_type.asymmetric,
+                    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()

--- a/src/khoj/routers/api.py
+++ b/src/khoj/routers/api.py
@ -384,8 +384,13 @@ def update(
 ):
    try:
        state.search_index_lock.acquire()
-        state.model = configure_search(state.model, state.config, regenerate=force or False, t=t)
-        state.search_index_lock.release()
+        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)
        raise HTTPException(status_code=500, detail=str(e))
--- a/src/khoj/utils/cli.py
+++ b/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(
-        description="Start Khoj; A Natural Language Search Engine for your personal Notes, Transactions and Photos"
-    )
+    parser = argparse.ArgumentParser(description="Start Khoj; An AI personal assistant for your Digital Brain")
    parser.add_argument(
        "--config-file", "-c", default="~/.khoj/khoj.yml", type=pathlib.Path, help="YAML file to configure Khoj"
    )