Borg/pkg/website/website.go
google-labs-jules[bot] 52a07f46be feat: Add ability to download from GitHub releases
This commit introduces a new command `collect github-release` that allows downloading assets from the latest GitHub release of a repository.

The command supports the following features:
- Downloading a specific file from the release using the `--file` flag.
- Downloading all assets from the release and packing them into a DataNode using the `--pack` flag.
- Specifying an output directory for the downloaded files using the `--output` flag.

This commit also includes a project-wide refactoring of the Go module path to `github.com/Snider/Borg` to align with Go's module system best practices.
2025-11-01 19:03:04 +00:00

166 lines
3.4 KiB
Go

package website
import (
"fmt"
"io"
"net/http"
"net/url"
"strings"
"github.com/Snider/Borg/pkg/datanode"
"github.com/schollz/progressbar/v3"
"golang.org/x/net/html"
)
// Downloader is a recursive website downloader.
type Downloader struct {
baseURL *url.URL
dn *datanode.DataNode
visited map[string]bool
maxDepth int
progressBar *progressbar.ProgressBar
}
// NewDownloader creates a new Downloader.
func NewDownloader(maxDepth int) *Downloader {
return &Downloader{
dn: datanode.New(),
visited: make(map[string]bool),
maxDepth: maxDepth,
}
}
// DownloadAndPackageWebsite downloads a website and packages it into a DataNode.
func DownloadAndPackageWebsite(startURL string, maxDepth int) (*datanode.DataNode, error) {
baseURL, err := url.Parse(startURL)
if err != nil {
return nil, err
}
d := NewDownloader(maxDepth)
d.baseURL = baseURL
fmt.Println("Downloading website...")
d.progressBar = progressbar.NewOptions(1, progressbar.OptionSetDescription("Downloading"))
d.crawl(startURL, 0)
return d.dn, nil
}
func (d *Downloader) crawl(pageURL string, depth int) {
if depth > d.maxDepth || d.visited[pageURL] {
return
}
d.visited[pageURL] = true
d.progressBar.Add(1)
resp, err := http.Get(pageURL)
if err != nil {
fmt.Printf("Error getting %s: %v\n", pageURL, err)
return
}
defer resp.Body.Close()
body, err := io.ReadAll(resp.Body)
if err != nil {
fmt.Printf("Error reading body of %s: %v\n", pageURL, err)
return
}
relPath := d.getRelativePath(pageURL)
d.dn.AddData(relPath, body)
doc, err := html.Parse(strings.NewReader(string(body)))
if err != nil {
fmt.Printf("Error parsing HTML of %s: %v\n", pageURL, err)
return
}
var f func(*html.Node)
f = func(n *html.Node) {
if n.Type == html.ElementNode {
for _, a := range n.Attr {
if a.Key == "href" || a.Key == "src" {
link, err := d.resolveURL(pageURL, a.Val)
if err != nil {
continue
}
if d.isLocal(link) {
if isAsset(link) {
d.downloadAsset(link)
} else {
d.crawl(link, depth+1)
}
}
}
}
}
for c := n.FirstChild; c != nil; c = c.NextSibling {
f(c)
}
}
f(doc)
}
func (d *Downloader) downloadAsset(assetURL string) {
if d.visited[assetURL] {
return
}
d.visited[assetURL] = true
d.progressBar.Add(1)
resp, err := http.Get(assetURL)
if err != nil {
fmt.Printf("Error getting asset %s: %v\n", assetURL, err)
return
}
defer resp.Body.Close()
body, err := io.ReadAll(resp.Body)
if err != nil {
fmt.Printf("Error reading body of asset %s: %v\n", assetURL, err)
return
}
relPath := d.getRelativePath(assetURL)
d.dn.AddData(relPath, body)
}
func (d *Downloader) getRelativePath(pageURL string) string {
u, err := url.Parse(pageURL)
if err != nil {
return ""
}
return strings.TrimPrefix(u.Path, "/")
}
func (d *Downloader) resolveURL(base, ref string) (string, error) {
baseURL, err := url.Parse(base)
if err != nil {
return "", err
}
refURL, err := url.Parse(ref)
if err != nil {
return "", err
}
return baseURL.ResolveReference(refURL).String(), nil
}
func (d *Downloader) isLocal(pageURL string) bool {
u, err := url.Parse(pageURL)
if err != nil {
return false
}
return u.Hostname() == d.baseURL.Hostname()
}
func isAsset(pageURL string) bool {
ext := []string{".css", ".js", ".png", ".jpg", ".jpeg", ".gif", ".svg", ".ico"}
for _, e := range ext {
if strings.HasSuffix(pageURL, e) {
return true
}
}
return false
}