Initial rss with partially tested items

rss/item.go

@@ -0,0 +1,68 @@
+package rss
+
+import (
+	"html"
+	"regexp"
+	"strings"
+	"time"
+
+	"github.com/mmcdole/gofeed"
+)
+
+type Item struct {
+	Name    string
+	Link    string
+	Content string
+	TS      time.Time
+}
+
+func fromGofeedItem(gfitem *gofeed.Item, filter string) *Item {
+	item := &Item{
+		Name:    gfitem.Title,
+		Link:    gfitem.Link,
+		Content: "",
+		TS:      *gofeedItemTS(gfitem),
+	}
+	if filter == "" {
+		item.Content = gfitem.Content
+		return item
+	}
+	r := regexp.MustCompile(filter)
+	matches := r.FindAllString(gfitem.Content, -1)
+	content := strings.Join(matches, "\n<br>\n")
+	content = cleanImgTags(content, item.Link)
+	item.Content = content
+	return item
+}
+
+func cleanImgTags(s, url string) string {
+	domain := regexp.MustCompile("(https?://)?(www\\.)?[a-zA-Z0-9]+\\.+[a-z]{2}[a-z]?").FindString(url)
+	reg := regexp.MustCompile("<img.+?/(img)?>")
+	s = html.UnescapeString(s)
+	matches := reg.FindAllString(s, -1)
+	if len(matches) > 0 {
+		// get img src="..." and build
+		regImgSrc := regexp.MustCompile("src=\"[^\"]+\"")
+		for j := range matches {
+			theseMatches := regImgSrc.FindAllString(matches[j], -1)
+			for k := range theseMatches {
+				if strings.HasPrefix(theseMatches[k], "src=\"/") {
+					theseMatches[k] = "src=\"" + domain + theseMatches[k][5:]
+				}
+				theseMatches[k] = "<img " + theseMatches[k] + " />"
+			}
+			s = strings.Replace(s, matches[j], strings.Join(theseMatches, "<br>"), 1)
+		}
+	}
+	return s
+}
+
+func gofeedItemTS(gfitem *gofeed.Item) *time.Time {
+	var t time.Time
+	if gfitem.UpdatedParsed != nil {
+		t = *gfitem.UpdatedParsed
+	} else if gfitem.PublishedParsed != nil {
+		t = *gfitem.PublishedParsed
+	}
+	return &t
+}