### Copy .env Example File

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/gdrive_text_embedding/README.md

Copy the example environment file to start configuring your credentials and folder IDs.

```bash
cp .env.exmaple .env
$EDITOR .env
```

--------------------------------

### Run Frontend Application

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/image_search/README.md

Navigates to the frontend directory, installs its Node.js dependencies, and starts the development server.

```bash
cd frontend
```

```bash
npm install
```

```bash
npm run dev
```

--------------------------------

### FastAPI Application Setup and Event Handlers

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Sets up a FastAPI application, initializes CocoIndex, establishes a database connection pool, and starts a live updater for the document index flow on startup. Shuts down the live updater and closes the connection pool on shutdown.

```python
app = FastAPI(title="Semantic Search API")


@app.on_event("startup")
def startup_event():
    load_dotenv()
    cocoindex.init()

    # Initialize connection pool
    app.state.pool = ConnectionPool(os.getenv("COCOINDEX_DATABASE_URL"))

    # Start live updater to keep index fresh
    app.state.live_updater = cocoindex.FlowLiveUpdater(document_index_flow)
    app.state.live_updater.start()


@app.on_event("shutdown")
def shutdown_event():
    app.state.live_updater.abort()
    app.state.live_updater.wait()
    app.state.pool.close()
```

--------------------------------

### Install psycopg Library

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/getting_started/quickstart.md

Install the psycopg library with binary and pool support, which is required for database interactions and querying.

```bash
pip install psycopg[binary,pool]
```

--------------------------------

### Bash: Running the CocoIndex Quickstart Script

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/getting_started/quickstart.md

This command executes the Python quickstart script to interactively search the index. It will prompt the user for search queries.

```bash
python quickstart.py
```

--------------------------------

### Example .env File Configuration

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/amazon_s3_embedding/README.md

This is an example of the .env file content, showing required database and Amazon S3 configuration variables. Ensure you replace placeholder values with your actual credentials.

```dotenv
# Database Configuration
DATABASE_URL=postgresql://localhost:5432/cocoindex

# Amazon S3 Configuration
AMAZON_S3_BUCKET_NAME=your-bucket-name
AMAZON_S3-SQS_QUEUE_URL=https://sqs.us-west-2.amazonaws.com/123456789/S3ChangeNotifications
```

--------------------------------

### Install Project Dependencies

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/README.md

Installs project dependencies using Yarn. This is the first step before running any other commands.

```bash
$ yarn
```

--------------------------------

### Start Neo4j Docker Compose

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/ops/storages.md

Starts a Neo4j Enterprise instance using a provided Docker Compose configuration. Note that this uses an evaluation license with a 30-day trial period.

```bash
docker compose -f <(curl -L https://raw.githubusercontent.com/cocoindex-io/cocoindex/refs/heads/main/dev/neo4j.yaml) up -d
```

--------------------------------

### Kuzu API Server Setup

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/ops/storages.md

Instructions for setting up and running a local Kuzu API server using Docker.

```APIDOC
## Kuzu API Server Setup

CocoIndex supports interacting with Kuzu through its API server. You can set up a local Kuzu API server using Docker with the following command:

### Command
```bash
KUZU_DB_DIR=$HOME/.kuzudb
KUZU_PORT=8123
docker run -d --name kuzu -p ${KUZU_PORT}:8000 -v ${KUZU_DB_DIR}:/database kuzudb/api-server:latest
```

### Description
This command starts a Kuzu API server in detached mode (`-d`), names the container `kuzu`, maps the host port `8123` to the container port `8000`, and mounts a local directory `$HOME/.kuzudb` to `/database` inside the container for persistent storage.
```

--------------------------------

### Install Rust

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/about/contributing.md

Install Rust using the official script. This is a prerequisite for building the CocoIndex project.

```bash
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
```

--------------------------------

### Start Local Development Server

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/README.md

Starts a local development server for the Docusaurus website. Changes are reflected live without server restarts.

```bash
$ yarn start
```

--------------------------------

### Copy and Edit .env Example

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/amazon_s3_embedding/README.md

Copy the example environment file and edit it to include your Amazon S3 bucket name and optionally a prefix. This file configures database and S3 connection details.

```bash
cp .env.example .env
$EDITOR .env
```

--------------------------------

### Build and Run Docker Containers

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/fastapi_server_docker/README.md

Build the Docker images and start the services defined in the docker-compose.yml file.

```bash
docker compose up --build
```

--------------------------------

### Start CocoInsight Server

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/docs_to_knowledge_graph/README.md

Run CocoInsight to troubleshoot index generation and understand data lineage. This command starts a local server for CocoInsight.

```bash
cocoindex server -ci main.py
```

--------------------------------

### Start HTTP Server for CocoInsight

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Start the CocoIndex HTTP server, enabling integration with CocoInsight. Specify the address and enable CORS for CocoIndex.

```bash
cocoindex server main.py --address 0.0.0.0:8080 --cors-cocoindex
```

--------------------------------

### Install Maturin

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/about/contributing.md

Install the maturin build tool, which is used for building and packaging Rust-based Python extensions.

```bash
pip install maturin
```

--------------------------------

### Setup Cocoindex with Main Script

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding/Text_Embedding.ipynb

This command initializes the Cocoindex environment using the specified Python script 'main.py'. The 'yes yes |' part is used to automatically confirm any prompts during the setup process.

```bash
!yes yes | cocoindex setup main.py
```

--------------------------------

### Setup CocoIndex Index

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/getting_started/quickstart.md

Use this command to set up the necessary tables in the database for the CocoIndex flow. Enter 'yes' when prompted to confirm creation.

```bash
cocoindex setup quickstart.py
```

--------------------------------

### Start Postgres Database with Docker Compose

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/getting_started/installation.md

Start a PostgreSQL database with the pgvector extension for CocoIndex using Docker Compose. This command downloads the configuration and starts the service in detached mode.

```bash
docker compose -f <(curl -L https://raw.githubusercontent.com/cocoindex-io/cocoindex/refs/heads/main/dev/postgres.yaml) up -d
```

--------------------------------

### Setup CocoIndex Main Configuration

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/docs_to_knowledge_graph/README.md

Set up the main configuration file for CocoIndex. This command prepares the indexing process.

```bash
cocoindex setup main.py
```

--------------------------------

### Run Qdrant Docker Container

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding_qdrant/README.md

This command starts a Qdrant instance in a Docker container, exposing the necessary ports for operation.

```bash
docker run -d -p 6334:6334 -p 6333:6333 qdrant/qdrant
```

--------------------------------

### Start Server with Live Updates and CORS

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Launch the CocoIndex server with live update capabilities enabled and configure CORS for a local development environment.

```bash
cocoindex server main.py --live-update --cors-local 3000
```

--------------------------------

### Run Qdrant Vector Storage

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/image_search/README.md

Starts a Qdrant instance using Docker. Ensure Postgres is also running and the COCOINDEX_DATABASE_URL environment variable is set.

```bash
docker run -d -p 6334:6334 -p 6333:6333 qdrant/qdrant
```

```bash
export COCOINDEX_DATABASE_URL="postgres://cocoindex:cocoindex@localhost/cocoindex"
```

--------------------------------

### Drop Specific Flow Setup

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Remove the backend setup (tables and metadata) for a particular flow. Use this to clean up resources for a specific flow.

```bash
cocoindex drop main.py:TextEmbedding
```

--------------------------------

### Run Kuzu API Server with Docker

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/ops/storages.md

Use this command to start a Kuzu API server locally using Docker. Ensure KUZU_DB_DIR and KUZU_PORT environment variables are set.

```bash
KUZU_DB_DIR=$HOME/.kuzudb
KUZU_PORT=8123
docker run -d --name kuzu -p ${KUZU_PORT}:8000 -v ${KUZU_DB_DIR}:/database kuzudb/api-server:latest
```

--------------------------------

### Build CocoIndex Index

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/getting_started/quickstart.md

Execute this command to build the index after setup. It processes data and provides statistics on the changes.

```bash
cocoindex update quickstart.py
```

--------------------------------

### Example Usage of Semantic Search

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Demonstrates how to initialize CocoIndex, set up a connection pool, and perform a semantic search using the `semantic_search` function. It then prints the results.

```python
if __name__ == "__main__":
    from dotenv import load_dotenv
    load_dotenv()

    cocoindex.init()
    pool = ConnectionPool(os.getenv("COCOINDEX_DATABASE_URL"))

    # Search for documents
    results = semantic_search(pool, "How does incremental processing work?")
    for result in results:
        print(f"[{result['score']:.3f}] {result['filename']}: {result['text'][:100]}...")
```

--------------------------------

### Enable Cache for Standalone Function

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/core/custom_function.mdx

This example demonstrates how to enable caching for a standalone custom function using the `cache=True` and `behavior_version=1` arguments.

```python
@cocoindex.op.function(cache=True, behavior_version=1)
def compute_something(arg1: str, arg2: int | None = None) -> str:
    ...
```

--------------------------------

### Export to Qdrant Vector Database

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Set up data export to the Qdrant vector database. This example demonstrates collecting data with a generated UUID as the primary key and exporting embeddings.

```python
# Qdrant vector database export
@cocoindex.flow_def(name="QdrantExport")
def qdrant_flow(flow_builder: cocoindex.FlowBuilder, data_scope: cocoindex.DataScope):
    data_scope["docs"] = flow_builder.add_source(cocoindex.sources.LocalFile(path="./docs"))
    collector = data_scope.add_collector()

    with data_scope["docs"].row() as doc:
        doc["embedding"] = doc["content"].transform(
            cocoindex.functions.SentenceTransformerEmbed(model="sentence-transformers/all-MiniLM-L6-v2")
        )
        collector.collect(
            id=cocoindex.GeneratedField.UUID,
            filename=doc["filename"],
            embedding=doc["embedding"]
        )

    collector.export(
        "doc_embeddings",
        cocoindex.storages.Qdrant(
            collection_name="my_collection",
            connection=cocoindex.ref_auth_entry("QdrantConnection")  # Optional: use auth registry
        ),
        primary_key_fields=["id"],
    )
```

--------------------------------

### Initialize CocoIndex with Environment Variables

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/core/settings.mdx

Load settings from environment variables. Ensure that `python-dotenv` is installed if you need to load variables from a `.env` file.

```python
from dotenv import load_dotenv
import cocoindex

load_dotenv()
cocoindex.init()
```

--------------------------------

### Run FastAPI Application Locally

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/fastapi_server_docker/README.md

Start the FastAPI application using uvicorn for local development. Use --reload for automatic restarts on code changes.

```bash
uvicorn main:fastapi_app --reload --host 0.0.0.0 --port 8000
```

--------------------------------

### Enable Cache for Function with Executor

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/core/custom_function.mdx

This example shows how to enable caching for a function defined by a spec and an executor, using `cache=True` and `behavior_version=1` on the executor class.

```python
class ComputeSomething(cocoindex.op.FunctionSpec):
    ...

@cocoindex.op.executor_class(cache=True, behavior_version=1)
class ComputeSomethingExecutor:
    spec: ComputeSomething

    ...
```

--------------------------------

### Install CocoIndex Dependencies

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/docs_to_knowledge_graph/README.md

Install the necessary Python dependencies for CocoIndex. Ensure you have Python installed.

```bash
pip install -e .
```

--------------------------------

### Install CocoIndex Package

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/getting_started/installation.md

Install the CocoIndex Python package using pip. Ensure you have Python and pip installed.

```bash
pip install -U cocoindex
```

--------------------------------

### Python: Main Script Logic for CocoIndex Quickstart

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/getting_started/quickstart.md

This script initializes CocoIndex and a database connection pool, then enters a loop to process user search queries. It requires the 'search' function to be defined elsewhere and handles user input for queries.

```python
if __name__ == "__main__":
    # Initialize CocoIndex library states
    cocoindex.init()

    # Initialize the database connection pool.
    pool = ConnectionPool(os.getenv("COCOINDEX_DATABASE_URL"))
    # Run queries in a loop to demonstrate the query capabilities.
    while True:
        try:
            query = input("Enter search query (or Enter to quit): ")
            if query == '':
                break
            # Run the query function with the database connection pool and the query.
            results = search(pool, query)
            print("\nSearch results:")
            for result in results:
                print(f"[{result['score']:.3f}] {result['filename']}")
                print(f"    {result['text']}")
                print("---")
            print()
        except KeyboardInterrupt:
            break
```

--------------------------------

### Install Cocoindex and Dependencies

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding/Text_Embedding.ipynb

Install Cocoindex and other required packages using pip. This command ensures all necessary libraries are available for the project.

```python
%pip install cocoindex python-dotenv psycopg[binary,pool]
```

--------------------------------

### JSON Primitive Value Examples

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding/markdown_files/rfc8259.md

These examples demonstrate valid JSON texts containing only primitive values: a string, a number, and a boolean.

```json
"Hello world!"
```

```json
42
```

```json
true
```

--------------------------------

### Download Demo Markdown Files

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding/Text_Embedding.ipynb

Download sample markdown files for demonstration purposes. These files will be used as the data source for embedding.

```bash
!mkdir -p markdown_files && \
wget -P markdown_files https://raw.githubusercontent.com/cocoindex-io/cocoindex/refs/heads/main/examples/text_embedding/markdown_files/1706.03762v7.md && \
wget -P markdown_files https://raw.githubusercontent.com/cocoindex-io/cocoindex/refs/heads/main/examples/text_embedding/markdown_files/1810.04805v2.md && \
wget -P markdown_files https://raw.githubusercontent.com/cocoindex-io/cocoindex/refs/heads/main/examples/text_embedding/markdown_files/rfc8259.md
```

--------------------------------

### Install CocoIndex via Pip

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Install the CocoIndex package using pip. This is the first step before using any CocoIndex CLI commands or Python APIs.

```bash
pip install cocoindex
```

--------------------------------

### Update Rust

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/about/contributing.md

Update an existing Rust installation to the latest version. Ensure your Rust toolchain is current.

```bash
rustup update
```

--------------------------------

### Drop All Persisted Flows

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Remove the backend setup for all persisted flows in the system. This is the most extensive cleanup command.

```bash
cocoindex drop --all
```

--------------------------------

### Getting App Namespace

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/core/flow_def.mdx

Explains how to retrieve and utilize the application namespace for organizing backends and naming resources.

```APIDOC
## GET /api/app_namespace

### Description
Retrieves the current application namespace, which can be used to organize flows and name backends across different environments.

### Method
GET

### Endpoint
`/api/app_namespace`

### Parameters
#### Path Parameters
None

#### Query Parameters
- **trailing_delimiter** (string) - Optional - A string to append to the app namespace if it's not empty.

### Request Example
```python
collection_name=cocoindex.get_app_namespace(trailing_delimiter='__') + "doc_embeddings"
```

### Response
#### Success Response (200)
- **app_namespace** (string) - The current application namespace.

#### Response Example
```json
{
  "app_namespace": "Staging"
}
```
```

--------------------------------

### Run the Python Search Application

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding/Text_Embedding.ipynb

This command executes the Python script 'main.py'. The script will then prompt the user for search queries.

```bash
!python main.py
```

--------------------------------

### Show All Subcommands with CLI

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/docs/docs/core/cli.mdx

Run `cocoindex --help` to display a list of all available subcommands for the CocoIndex CLI.

```sh
cocoindex --help
```

--------------------------------

### Run CocoInsight Server with Live Updates

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/amazon_s3_embedding/README.md

Run the CocoInsight server with the `-L` flag to enable continuous index updates. This ensures the server reflects the latest changes from the source as they occur.

```bash
cocoindex server -ci -L main.py
```

--------------------------------

### Run CocoIndex Backend

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/image_search/README.md

Sets up the CocoIndex configuration and runs the FastAPI backend server. The server will reload automatically on code changes.

```bash
cocoindex setup main.py
```

```bash
uvicorn main:app --reload --host 0.0.0.0 --port 8000
```

--------------------------------

### JSON Object Example

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding/markdown_files/rfc8259.md

This is a standard JSON object representing image metadata. It includes nested objects and an array of numbers.

```json
{
  "Image": {
    "Width": 800,
    "Height": 600,
    "Title": "View from 15th Floor",
    "Thumbnail": {
      "Url": "http://www.example.com/image/481989943",
      "Height": 125,
      "Width": 100
    },
    "Animated" : false,
    "IDs": [116, 943, 234, 38793]
  }
}
```

--------------------------------

### Define Search Function and Main Execution Loop

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding/Text_Embedding.ipynb

This Python script initializes a database connection pool and enters a loop to accept user search queries. It prints formatted search results including score, filename, and text snippet. Handles KeyboardInterrupt for graceful exit.

```python
%%writefile -a main.py

def _main():
    # Initialize the database connection pool.
    pool = ConnectionPool(os.getenv("COCOINDEX_DATABASE_URL"))
    # Run queries in a loop to demonstrate the query capabilities.
    while True:
        try:
            query = input("Enter search query (or Enter to quit): ")
            if query == '':
                break
            # Run the query function with the database connection pool and the query.
            results = search(pool, query)
            print("\nSearch results:")
            for result in results:
                print(f"[{result['score']:.3f}] {result['filename']}")
                print(f"    {result['text']}")
                print("---")
            print()
        except KeyboardInterrupt:
            break

if __name__ == "__main__":
    load_dotenv(override=True)
    cocoindex.init()
    _main()

```

--------------------------------

### JSON Object Example

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/text_embedding_qdrant/markdown_files/rfc8259.md

This is a standard JSON object representing image metadata. It includes nested objects and an array of numbers.

```json
{
  "Image": {
    "Width": 800,
    "Height": 600,
    "Title": "View from 15th Floor",
    "Thumbnail": {
      "Url": "http://www.example.com/image/481989943",
      "Height": 125,
      "Width": 100
    },
    "Animated": false,
    "IDs": [116, 943, 234, 38793]
  }
}
```

--------------------------------

### Connect to Postgres Shell

Source: https://github.com/cocoindex-dev/cocoindex/blob/main/examples/manuals_llm_extraction/README.md

Connect to the local PostgreSQL database using the psql command-line tool. Ensure your database credentials and host are correct.

```bash
psql postgres://cocoindex:cocoindex@localhost/cocoindex
```

--------------------------------

### Handle Asynchronous Updates

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

This example shows how to use CocoIndex with asynchronous operations, utilizing an async context manager for live updaters.

```python
async def run_async_updates():
    cocoindex.init()

    async with cocoindex.FlowLiveUpdater(live_index_flow) as updater:
        await updater.wait_async()
```

--------------------------------

### Initialize CocoIndex with Settings

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Initializes the CocoIndex library. Settings can be loaded from environment variables or provided directly. Ensure COCOINDEX_DATABASE_URL is set if loading from environment variables.

```python
import cocoindex
from cocoindex import Settings, DatabaseConnectionSpec

# Option 1: Load settings from environment variables
# Required: COCOINDEX_DATABASE_URL
# Optional: COCOINDEX_DATABASE_USER, COCOINDEX_DATABASE_PASSWORD, COCOINDEX_APP_NAMESPACE
cocoindex.init()

# Option 2: Provide settings directly
settings = Settings(
    database=DatabaseConnectionSpec(
        url="postgresql://localhost:5432/cocoindex",
        user="postgres",
        password="secret"
    ),
    app_namespace="my_app"  # Optional: prefix for flow names
)
cocoindex.init(settings)
```

--------------------------------

### Drop All Flows in a Module

Source: https://context7.com/cocoindex-dev/cocoindex/llms.txt

Remove the backend setup for all flows defined within a specified Python module. This is a more comprehensive cleanup operation.

```bash
cocoindex drop main.py
```