From 39b36b6857e48ad9d352f374b957f604d11caa20 Mon Sep 17 00:00:00 2001
From: Alex <a@tushynski.me>
Date: Tue, 13 May 2025 14:03:05 +0100
Subject: [PATCH] Feat: Add MD gen script, enable Qdrant lazy loading

---
 application/requirements.txt      |  1 -
 application/vectorstore/qdrant.py |  5 ++--
 md-gen.py                         | 47 +++++++++++++++++++++++++++++++
 3 files changed, 50 insertions(+), 3 deletions(-)
 create mode 100644 md-gen.py

diff --git a/application/requirements.txt b/application/requirements.txt
index 02476089..d68554a3 100644
--- a/application/requirements.txt
+++ b/application/requirements.txt
@@ -71,7 +71,6 @@ python-dateutil==2.9.0.post0
 python-dotenv==1.0.1
 python-jose==3.4.0
 python-pptx==1.0.2
-qdrant-client==1.13.2
 redis==5.2.1
 referencing==0.30.2
 regex==2024.11.6
diff --git a/application/vectorstore/qdrant.py b/application/vectorstore/qdrant.py
index 3f94505f..61a9d63d 100644
--- a/application/vectorstore/qdrant.py
+++ b/application/vectorstore/qdrant.py
@@ -1,11 +1,12 @@
-from langchain_community.vectorstores.qdrant import Qdrant
 from application.vectorstore.base import BaseVectorStore
 from application.core.settings import settings
-from qdrant_client import models
 
 
 class QdrantStore(BaseVectorStore):
     def __init__(self, source_id: str = "", embeddings_key: str = "embeddings"):
+        from qdrant_client import models
+        from langchain_community.vectorstores.qdrant import Qdrant
+
         self._filter = models.Filter(
             must=[
                 models.FieldCondition(
diff --git a/md-gen.py b/md-gen.py
new file mode 100644
index 00000000..93754475
--- /dev/null
+++ b/md-gen.py
@@ -0,0 +1,47 @@
+import os
+
+def create_markdown_from_directory(directory=".", output_file="combined.md"):
+    """
+    Recursively traverses the given directory, reads all files (ignoring files/folders in ignore_list),
+    and creates a single markdown file containing the contents of each file, prefixed with the
+    relative path of the file.
+
+    Args:
+        directory (str): The directory to traverse. Defaults to the current directory.
+        output_file (str): The name of the output markdown file. Defaults to 'combined.md'.
+    """
+    ignore_list = [
+        "node_modules", "__pycache__", ".git", ".DS_Store", "inputs", "indexes",
+        "model", "models", ".venv", "temp", ".pytest_cache", ".ruff_cache",
+        "extensions", "dir_tree.py", "map.txt", "signal-desktop-keyring.gpg",
+        ".husky", ".next", "docs", "index.pkl", "index.faiss", "assets", "fonts", "public",
+        "yarn.lock", "package-lock.json",
+        ]
+
+    with open(output_file, "w", encoding="utf-8") as outfile:
+        for root, dirs, files in os.walk(directory):
+            # Filter out directories in ignore_list so they won't be traversed
+            dirs[:] = [d for d in dirs if d not in ignore_list]
+            
+            for filename in files:
+                if filename in ignore_list:
+                    continue
+                filepath = os.path.join(root, filename)
+                
+                try:
+                    with open(filepath, "r", encoding="utf-8") as infile:
+                        content = infile.read()
+                    
+                    # Get a relative path to better indicate file location
+                    rel_path = os.path.relpath(filepath, directory)
+                    outfile.write(f"## File: {rel_path}\n\n")
+                    outfile.write(content)
+                    outfile.write("\n\n---\n\n")  # Separator between files
+
+                except Exception as e:
+                    print(f"Error processing file {filepath}: {e}")
+
+    print(f"Successfully created {output_file}")
+
+if __name__ == "__main__":
+    create_markdown_from_directory()