webweav.ing/rss/rss.go

90 lines
2.1 KiB
Go
Raw Normal View History

2024-01-19 03:33:37 +00:00
package rss
import (
"fmt"
"io"
"net/http"
"strings"
"time"
"golang.org/x/net/html"
)
func fetchPage(url string) (string, error) {
resp, err := http.Get(url)
if err != nil {
return "", fmt.Errorf("Error sending Get request: %w", err)
}
defer resp.Body.Close()
body, err := io.ReadAll(resp.Body)
return string(body), nil
}
func parseTime(timestr string) (*time.Time, error) {
var formats = []string {
time.ANSIC,
time.UnixDate,
time.RubyDate,
time.RFC822,
time.RFC822Z,
time.RFC850,
time.RFC1123,
time.RFC1123Z,
time.RFC3339,
time.RFC3339Nano,
time.DateTime,
time.DateOnly,
}
for _, f := range formats {
pagetime, err := time.Parse(f, timestr)
if err == nil {
return &pagetime, nil
}
}
return nil, fmt.Errorf("Error parsing time: invalid format")
}
// parseArticle returns an error if it could not parse the HTML or if it could not parse a time
// if a time could not be parsed, the parsed html article will still be returned
func parseArticle(content string) (string, *time.Time, error) {
doc, err := html.Parse(strings.NewReader(content))
if err != nil {
return "", nil, fmt.Errorf("Error parsing HTML: %w", err)
}
var f func(*html.Node, string)
var element *html.Node
var pagetime *time.Time
f = func(n *html.Node, tag string) {
if n.Type == html.ElementNode && n.Data == tag {
element = n
return
}
for c := n.FirstChild; c != nil; c = c.NextSibling {
f(c, tag)
}
}
f(doc, "article")
var builder strings.Builder
html.Render(&builder, element)
f(element, "time")
for _, d := range element.Attr {
if d.Key == "datetime" {
pagetime, err = parseTime(d.Val)
}
}
return builder.String(), pagetime, nil
}
func GenerateRss(siteUrl, siteTitle string, pageUrls []string) string {
// get page
// parse article
// parse date
// create item element
// collect item elements into feed
return ""
}