feat: initial commit

This commit is contained in:
Admin
2026-02-26 12:56:25 +05:00
commit d68ea71239
15 changed files with 1893 additions and 0 deletions

View File

@@ -0,0 +1,101 @@
// Package browser defines the BrowserClient interface and helper types for
// communicating with a Browserless instance.
package browser
import "context"
// Strategy selects which Browserless API endpoint / protocol to use.
type Strategy string
const (
// StrategyContent uses the POST /content endpoint, which returns the final
// rendered HTML of the page. Fastest; suitable for most JS-rendered sites.
StrategyContent Strategy = "content"
// StrategyScrape uses the POST /scrape endpoint, which accepts a list of
// CSS selectors and returns structured JSON. Good when you know exactly
// which elements you need.
StrategyScrape Strategy = "scrape"
// StrategyCDP uses the WebSocket /devtools/browser endpoint (Chrome
// DevTools Protocol). Most powerful; required for complex interactions
// (clicking, scrolling, waiting for network idle, etc.).
StrategyCDP Strategy = "cdp"
)
// ContentRequest is the body sent to POST /content.
type ContentRequest struct {
URL string `json:"url"`
WaitFor string `json:"waitForSelector,omitempty"`
WaitForTimeout int `json:"waitForTimeout,omitempty"` // ms
RejectResources bool `json:"rejectResources,omitempty"`
}
// ScrapeElement is one element descriptor inside a ScrapeRequest.
type ScrapeElement struct {
Selector string `json:"selector"`
Timeout int `json:"timeout,omitempty"` // ms
}
// ScrapeRequest is the body sent to POST /scrape.
type ScrapeRequest struct {
URL string `json:"url"`
Elements []ScrapeElement `json:"elements"`
WaitFor string `json:"waitForSelector,omitempty"`
}
// ScrapeResult is one entry in the response from POST /scrape.
type ScrapeResult struct {
Selector string `json:"selector"`
Results []ScrapeElement `json:"results"`
}
// ScrapeAttribute holds a single attribute value from a scraped element.
type ScrapeAttribute struct {
Name string `json:"name"`
Value string `json:"value"`
}
// ScrapedElement is one item inside ScrapeResult.Results.
type ScrapedElement struct {
Text string `json:"text"`
Attributes []ScrapeAttribute `json:"attributes"`
}
// ScrapeResponse is the top-level response from POST /scrape.
type ScrapeResponse struct {
Data []ScrapeResult `json:"data"`
}
// BrowserClient is an abstraction over the three Browserless API strategies.
// Callers choose the strategy best suited to the target site; the interface
// signature is identical regardless of strategy.
type BrowserClient interface {
// Strategy returns the strategy this client uses.
Strategy() Strategy
// GetContent fetches the fully-rendered HTML of url using the /content
// endpoint. Only meaningful when Strategy() == StrategyContent.
GetContent(ctx context.Context, req ContentRequest) (string, error)
// ScrapePage calls the /scrape endpoint and returns structured data.
// Only meaningful when Strategy() == StrategyScrape.
ScrapePage(ctx context.Context, req ScrapeRequest) (ScrapeResponse, error)
// CDPSession opens a CDP WebSocket session and calls fn with the raw
// WebSocket connection. Only meaningful when Strategy() == StrategyCDP.
// The session is closed when fn returns.
CDPSession(ctx context.Context, pageURL string, fn CDPSessionFunc) error
}
// CDPSessionFunc is the callback invoked inside a CDP session.
// conn is a live *websocket.Conn connected to a Browserless CDP endpoint.
type CDPSessionFunc func(ctx context.Context, conn CDPConn) error
// CDPConn is the minimal interface the orchestrator needs over a CDP WebSocket.
type CDPConn interface {
// Send sends a raw CDP command (JSON-encoded) and returns the response.
Send(ctx context.Context, method string, params map[string]any) (map[string]any, error)
// Close closes the underlying connection.
Close() error
}