Merge pull request #3 from pgEdge/BR-378

BR-378 - Added note about installing Vectorizer with PEP packages

Author: dpage

docs/index.md

@@ -1,11 +1,16 @@
 # pgEdge Vectorizer
 
-pgEdge Vectorizer is a PostgreSQL extension that automatically chunks text content and generates vector embeddings using background workers. Vectorizer provides a seamless integration between your PostgreSQL database and embedding providers like OpenAI, making it easy to build AI-powered search and retrieval applications.
+pgEdge Vectorizer is a PostgreSQL extension that automatically chunks text
+content and generates vector embeddings using background workers. Vectorizer
+provides a seamless integration between your PostgreSQL database and embedding
+providers like OpenAI, making it easy to build AI-powered search and retrieval
+applications.
 
 pgEdge Vectorizer:
 
 - intelligently splits text into optimal-sized chunks.
-- handles embedding generation asynchronously using background workers without blocking.
+- handles embedding generation asynchronously using background workers
+  without blocking.
 - enables easy switching between OpenAI, Voyage AI, and Ollama.
 - processes embeddings efficiently in batches for better API usage.
 - automatically retries failed operations with exponential backoff.
@@ -14,14 +19,20 @@ pgEdge Vectorizer:
 
 ## pgEdge Vectorizer Architecture
 
-pgEdge Vectorizer uses a trigger-based architecture with background workers to process text asynchronously. The following steps describe the processing flow from data insertion to embedding storage:
+pgEdge Vectorizer uses a trigger-based architecture with background workers to
+process text asynchronously. The following steps describe the processing flow
+from data insertion to embedding storage:
 
 1. A trigger detects INSERT or UPDATE operations on the configured table.
-2. The chunking module splits the text into chunks using the configured strategy.
-3. The system inserts chunk records and queue items into the processing queue.
-4. Background workers pick up queue items using SKIP LOCKED for concurrent processing.
+2. The chunking module splits the text into chunks using the configured
+   strategy.
+3. The system inserts chunk records and queue items into the processing
+   queue.
+4. Background workers pick up queue items using SKIP LOCKED for concurrent
+   processing.
 5. The configured provider generates embeddings via its API.
-6. The storage layer updates the chunk table with the generated embeddings.
+6. The storage layer updates the chunk table with the generated
+   embeddings.
 
 
 ## Component Diagram

docs/installation.md

@@ -1,5 +1,11 @@
 # Installing pgEdge Vectorizer
 
+pgEdge Vectorizer automatically chunks text content and generates vector
+embeddings using background workers. You can install pgEdge Vectorizer with
+[pgEdge Enterprise Postgres](https://docs.pgedge.com/enterprise/) packages
+or build Vectorizer from source code from the
+[pgEdge repository](https://github.com/pgEdge/pgedge-vectorizer).
+
 Before installing pgEdge Vectorizer, you need to install:
 
 * a Postgres server, version 14 or above
@@ -8,7 +14,8 @@ Before installing pgEdge Vectorizer, you need to install:
 
 Then, to build Vectorizer:
 
-Clone the [pgedge-vectorizer](https://github.com/pgEdge/pgedge-vectorizer) repository, and move into the repository root:
+Clone the [pgedge-vectorizer](https://github.com/pgEdge/pgedge-vectorizer) 
+repository, and move into the repository root:
 
 ```bash
 git clone https://github.com/pgEdge/pgedge-vectorizer.git
@@ -29,14 +36,15 @@ echo "your-api-key" > ~/.pgedge-vectorizer-llm-api-key
 chmod 600 ~/.pgedge-vectorizer-llm-api-key
 ```
 
-Then, modify the `postgresql.conf` file, adding the Vectorizer extension and API key file details:
+Then, modify the `postgresql.conf` file, adding the Vectorizer extension and
+API key file details:
 
 ```ini
 shared_preload_libraries = 'pgedge_vectorizer'
 pgedge_vectorizer.provider = 'openai'
 pgedge_vectorizer.api_key_file = '~/.pgedge-vectorizer-llm-api-key'
 pgedge_vectorizer.model = 'text-embedding-3-small'
-pgedge_vectorizer.databases = 'mydb'  # Comma-separated list of databases to monitor
+pgedge_vectorizer.databases = 'mydb'  # Comma-separated list of monitored databases
 ```
 
 Restart PostgreSQL; then use your Postgres client to create the `vector` and `pgedge-vectorizer` extensions: