Struct MultiModalAdapter 

pub struct MultiModalAdapter { /* private fields */ }

Multi-modal content extraction adapter

Implements ScrapingService by routing content to the appropriate parser based on the declared content_type parameter.

An optional AIProvider (vision-capable) can be injected for image analysis.

Implementations

impl MultiModalAdapter

pub fn new( config: MultiModalConfig, vision_provider: Option<Arc<dyn AIProvider>>, ) -> Self

Create a new multi-modal adapter

Arguments

• config - Extraction configuration
• vision_provider - Optional vision-capable AI provider (e.g. Claude, GPT-4o)

Example

use stygian_graph::adapters::multimodal::{MultiModalAdapter, MultiModalConfig};

let adapter = MultiModalAdapter::new(MultiModalConfig::default(), None);

Trait Implementations

impl ScrapingService for MultiModalAdapter

fn execute<'life0, 'async_trait>( &'life0 self, input: ServiceInput, ) -> Pin<Box<dyn Future<Output = Result<ServiceOutput>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Extract structured content from multi-modal input

Reads content_type from params (or falls back to "unknown"). The actual content must be in params["content"] or input.url.

Example

use stygian_graph::adapters::multimodal::{MultiModalAdapter, MultiModalConfig};
use stygian_graph::ports::{ScrapingService, ServiceInput};
use serde_json::json;

let adapter = MultiModalAdapter::new(MultiModalConfig::default(), None);
let input = ServiceInput {
    url: "name,age\nalice,30".to_string(),
    params: json!({ "content_type": "text/csv" }),
};
// let output = adapter.execute(input).await.unwrap();

fn name(&self) -> &'static str

Service name for identification in logs and metrics