### Install Gemini Data Extractor using Pip

Source: https://github.com/nicodds/gemini-data-extractor/blob/master/README.md

Installs the Gemini Data Extractor tool from the Python Package Index (PyPI). This is the primary method for setting up the tool on your system.

```bash
pip install gemini-data-extractor
```

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

### Authentication using API Key

Source: https://github.com/nicodds/gemini-data-extractor/blob/master/README.md

Illustrates how to authenticate with the Gemini API using a provided API key, either directly via a command-line argument or by setting an environment variable.

```bash
gemini-data-extractor --path path/to/your/file.pdf --prompt "Your prompt here" --api-key your-api-key
```

```bash
export GEMINI_API_KEY=your-api-key
gemini-data-extractor --path path/to/your/file.pdf --prompt "Your prompt here"
```

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

### Authentication using Google Cloud SDK

Source: https://github.com/nicodds/gemini-data-extractor/blob/master/README.md

Demonstrates authentication for the Gemini API using the Google Cloud SDK's application default login. Requires project ID and location, which can be set via arguments or environment variables.

```bash
gcloud auth application-default login
```

```bash
gemini-data-extractor --path path/to/your/file.pdf --prompt "Your prompt here" --project-id your-project-id --location your-location
```

```bash
export GCP_PROJECT_ID=your-project-id
export GCP_LOCATION=your-location
gemini-data-extractor --path path/to/your/file.pdf --prompt "Your prompt here"
```

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

### Environment Variables for Configuration

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Shows how to configure authentication and settings for the Gemini Data Extractor using environment variables. These variables serve as fallbacks when command-line arguments are not provided, with CLI arguments taking precedence.

```Bash
# Google AI Studio API Key
export GEMINI_API_KEY=your-api-key

# Google Cloud Platform / Vertex AI
export GCP_PROJECT_ID=my-gcp-project
export GCP_LOCATION=us-central1

# Environment variables are used as fallback when CLI args are not provided
# CLI arguments take precedence over environment variables
```

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

### Basic Usage: Extract Data from PDF

Source: https://github.com/nicodds/gemini-data-extractor/blob/master/README.md

Demonstrates the basic command-line usage for extracting data from a PDF file. Requires the path to the PDF and a text prompt.

```bash
gemini-data-extractor --path path/to/your/file.pdf --prompt-text "Your prompt here"
```

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

### Environment Variables Configuration

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Details on how to configure authentication and settings using environment variables for the Gemini Data Extractor.

```APIDOC
## Environment Variables

### Description
Configure authentication and settings using environment variables.

### Environment Variables
- **GEMINI_API_KEY** (string) - Your Google AI Studio API Key.
- **GCP_PROJECT_ID** (string) - Your Google Cloud Platform Project ID (for Vertex AI).
- **GCP_LOCATION** (string) - The location for Google Cloud Vertex AI services (e.g., `us-central1`).

### Usage
Environment variables are used as a fallback when CLI arguments are not provided. CLI arguments take precedence over environment variables.

### Example
```bash
# Google AI Studio API Key
export GEMINI_API_KEY=your-api-key

# Google Cloud Platform / Vertex AI
export GCP_PROJECT_ID=my-gcp-project
export GCP_LOCATION=us-central1
```
```

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

### Configure Vertex AI for Extraction

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Authenticate using Google Cloud SDK to utilize Vertex AI endpoints for document extraction.

```bash
gcloud auth application-default login
export GCP_PROJECT_ID=my-gcp-project
export GCP_LOCATION=us-central1
gemini-data-extractor --path document.pdf --prompt-text "Extract all text content as JSON"
```

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

### Usage with Prompt File: Extract Data from Image

Source: https://github.com/nicodds/gemini-data-extractor/blob/master/README.md

Shows how to use a separate text file for the prompt when extracting data from an image file. This is useful for longer or more complex prompts.

```bash
gemini-data-extractor --path path/to/your/file.png --prompt-file path/to/your/prompt.txt
```

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

### Initialize ApiCaller Class

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Initialize the Python ApiCaller class using either an API key or Google Cloud project credentials.

```python
from gemini_data_extractor.api_caller import ApiCaller

# Initialize with API key
caller = ApiCaller(api_key="your-gemini-api-key")

# Initialize with Google Cloud credentials
caller = ApiCaller(project_id="my-gcp-project", location="us-central1")
```

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

### Extract Data via CLI

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Use the command-line interface to process files using either inline prompt text or external prompt files.

```bash
export GEMINI_API_KEY=your-api-key
gemini-data-extractor --path document.pdf --prompt-text "Extract the following fields as JSON: name, email, phone number"
gemini-data-extractor --path scanned_id.pdf --prompt-file extraction_prompt.txt
```

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

### Python API: Extract Data from Documents

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Demonstrates how to use the ApiCaller class to extract data from different file types like PDFs and images. The tool automatically detects MIME types and supports various document formats. It requires an API key for authentication and accepts either prompt text or a file path for processing.

```Python
from gemini_data_extractor.api_caller import ApiCaller

caller = ApiCaller(api_key="your-gemini-api-key")

# Process a PDF document
pdf_result = caller.extract(
    prompt_text="List all section headings as a JSON array",
    prompt_file=None,
    path="report.pdf"
)

# Process a PNG image
image_result = caller.extract(
    prompt_text="Extract all visible text from this receipt image as JSON",
    prompt_file=None,
    path="receipt.png"
)

# Process a JPEG photo
photo_result = caller.extract(
    prompt_text="Identify objects in this image and return as JSON array",
    prompt_file=None,
    path="photo.jpg"
)

# The MIME type is automatically detected from file extension
# Supported: application/pdf, image/png, image/jpeg, etc.
```

```Python
try:
    result = caller.extract(
        prompt_text="some prompt",
        prompt_file="some_file.txt",
        path="document.pdf"
    )
except ValueError as e:
    print(e)  # "Either prompt_text or prompt_file must be provided"
```

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

### Python API: Processing Different File Types

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Demonstrates how to use the ApiCaller to extract data from PDF, PNG, and JPEG files. The tool automatically detects MIME types.

```APIDOC
## Python API: Processing Different File Types

### Description
The tool automatically detects MIME types and supports various document formats including PDFs and images.

### Method
`ApiCaller.extract()`

### Parameters
#### Request Body (Implicit via arguments)
- **prompt_text** (string) - Required - The text prompt to guide the extraction.
- **prompt_file** (string) - Required - Path to a file containing the prompt. Either `prompt_text` or `prompt_file` must be provided, not both.
- **path** (string) - Required - Path to the document file (PDF, PNG, JPEG, etc.) to be processed.

### Request Example
```python
from gemini_data_extractor.api_caller import ApiCaller

caller = ApiCaller(api_key="your-gemini-api-key")

# Process a PDF document
pdf_result = caller.extract(
    prompt_text="List all section headings as a JSON array",
    prompt_file=None,
    path="report.pdf"
)

# Process a PNG image
image_result = caller.extract(
    prompt_text="Extract all visible text from this receipt image as JSON",
    prompt_file=None,
    path="receipt.png"
)

# Process a JPEG photo
photo_result = caller.extract(
    prompt_text="Identify objects in this image and return as JSON array",
    prompt_file=None,
    path="photo.jpg"
)
```

### Notes
- The MIME type is automatically detected from the file extension.
- Supported formats include: `application/pdf`, `image/png`, `image/jpeg`, etc.
```

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

### Extract Data via Python API

Source: https://context7.com/nicodds/gemini-data-extractor/llms.txt

Process documents programmatically using the extract method with either inline strings or prompt files.

```python
from gemini_data_extractor.api_caller import ApiCaller

caller = ApiCaller(api_key="your-gemini-api-key")

# Extract using inline prompt
result = caller.extract(prompt_text="Extract invoice details as JSON", path="invoice.pdf")

# Extract using prompt file
result = caller.extract(prompt_file="prompts/identity_extraction.txt", path="scanned_document.pdf")
```

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

### Define Data Extraction JSON Schema

Source: https://github.com/nicodds/gemini-data-extractor/blob/master/examples/prompt_001.txt

This snippet represents the expected JSON output format for the data extraction process. It defines the required fields and their respective data types for person identification.

```json
{
  "name": "Mark",
  "surname": "Twain",
  "sex": "M",
  "birth_date": "1835-11-30"
}
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.