refactor: split modules and harden scraping workflows

internal/service/subscription_rss.go

@@ -0,0 +1,79 @@
+package service
+
+import (
+	"context"
+	"encoding/xml"
+	"fmt"
+	"io"
+	"net/http"
+	"regexp"
+	"strings"
+)
+
+// rssFeed is the minimal RSS subset we need to decode.
+type rssFeed struct {
+	XMLName xml.Name `xml:"rss"`
+	Channel struct {
+		Items []rssItem `xml:"item"`
+	} `xml:"channel"`
+}
+
+type rssItem struct {
+	Title       string `xml:"title"`
+	Link        string `xml:"link"`
+	GUID        string `xml:"guid"`
+	Description string `xml:"description"`
+	Enclosure   struct {
+		URL string `xml:"url,attr"`
+	} `xml:"enclosure"`
+}
+
+func (s *SubscriptionService) fetch(ctx context.Context, feedURL string) (*rssFeed, error) {
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, feedURL, nil)
+	if err != nil {
+		return nil, err
+	}
+	req.Header.Set("User-Agent", "MediaStationGo/0.1")
+	resp, err := http.DefaultClient.Do(req)
+	if err != nil {
+		return nil, err
+	}
+	defer resp.Body.Close()
+	if resp.StatusCode >= 400 {
+		return nil, fmt.Errorf("rss %s: %d", feedURL, resp.StatusCode)
+	}
+	body, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return nil, err
+	}
+	var f rssFeed
+	if err := xml.Unmarshal(body, &f); err != nil {
+		return nil, err
+	}
+	return &f, nil
+}
+
+func compileFilter(pat string) *regexp.Regexp {
+	pat = strings.TrimSpace(pat)
+	if pat == "" {
+		return nil
+	}
+	if r, err := regexp.Compile("(?i)" + pat); err == nil {
+		return r
+	}
+	return nil
+}
+
+func splitNonEmpty(s string) []string {
+	if s == "" {
+		return nil
+	}
+	out := make([]string, 0)
+	for _, p := range strings.Split(s, "\n") {
+		p = strings.TrimSpace(p)
+		if p != "" {
+			out = append(out, p)
+		}
+	}
+	return out
+}