50 lines
1.1 KiB
Go
50 lines
1.1 KiB
Go
package netutil
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"io"
|
|
"net/http"
|
|
|
|
"github.com/PuerkitoBio/goquery"
|
|
)
|
|
|
|
func FetchHTMLDocument(
|
|
ctx context.Context,
|
|
httpClient *http.Client,
|
|
url string,
|
|
prepareRequest func(*http.Request),
|
|
buildStatusError func(*http.Response, []byte) error,
|
|
) (*goquery.Document, *http.Response, error) {
|
|
client := httpClient
|
|
if client == nil {
|
|
client = http.DefaultClient
|
|
}
|
|
|
|
request, err := http.NewRequestWithContext(ctx, http.MethodGet, url, nil)
|
|
if err != nil {
|
|
return nil, nil, fmt.Errorf("failed to create request: %w", err)
|
|
}
|
|
if prepareRequest != nil {
|
|
prepareRequest(request)
|
|
}
|
|
|
|
response, err := client.Do(request)
|
|
if err != nil {
|
|
return nil, nil, fmt.Errorf("request failed: %w", err)
|
|
}
|
|
defer func() { _ = response.Body.Close() }()
|
|
|
|
if response.StatusCode != http.StatusOK {
|
|
body, _ := io.ReadAll(io.LimitReader(response.Body, Bytes512))
|
|
return nil, response, buildStatusError(response, body)
|
|
}
|
|
|
|
document, err := goquery.NewDocumentFromReader(response.Body)
|
|
if err != nil {
|
|
return nil, response, fmt.Errorf("failed to parse html: %w", err)
|
|
}
|
|
|
|
return document, response, nil
|
|
}
|