Do not crawl existing entry URLs

This commit is contained in:
Frédéric Guillot 2018-01-19 18:43:27 -08:00
parent 09785df07f
commit 3b62f904d6
3 changed files with 23 additions and 8 deletions

View File

@ -70,7 +70,7 @@ func (h *Handler) CreateFeed(userID, categoryID int64, url string, crawler bool)
return nil, err
}
feedProcessor := processor.NewFeedProcessor(subscription)
feedProcessor := processor.NewFeedProcessor(userID, h.store, subscription)
feedProcessor.WithCrawler(crawler)
feedProcessor.Process()
@ -162,7 +162,7 @@ func (h *Handler) RefreshFeed(userID, feedID int64) error {
return err
}
feedProcessor := processor.NewFeedProcessor(subscription)
feedProcessor := processor.NewFeedProcessor(userID, h.store, subscription)
feedProcessor.WithScraperRules(originalFeed.ScraperRules)
feedProcessor.WithRewriteRules(originalFeed.RewriteRules)
feedProcessor.WithCrawler(originalFeed.Crawler)

View File

@ -10,10 +10,13 @@ import (
"github.com/miniflux/miniflux/reader/rewrite"
"github.com/miniflux/miniflux/reader/sanitizer"
"github.com/miniflux/miniflux/reader/scraper"
"github.com/miniflux/miniflux/storage"
)
// FeedProcessor handles the processing of feed contents.
type FeedProcessor struct {
userID int64
store *storage.Storage
feed *model.Feed
scraperRules string
rewriteRules string
@ -39,11 +42,15 @@ func (f *FeedProcessor) WithRewriteRules(rules string) {
func (f *FeedProcessor) Process() {
for _, entry := range f.feed.Entries {
if f.crawler {
content, err := scraper.Fetch(entry.URL, f.scraperRules)
if err != nil {
logger.Error("[FeedProcessor] %v", err)
if f.store.EntryURLExists(f.userID, entry.URL) {
logger.Debug(`[FeedProcessor] Do not crawl existing entry URL: "%s"`, entry.URL)
} else {
entry.Content = content
content, err := scraper.Fetch(entry.URL, f.scraperRules)
if err != nil {
logger.Error("[FeedProcessor] %v", err)
} else {
entry.Content = content
}
}
}
@ -53,6 +60,6 @@ func (f *FeedProcessor) Process() {
}
// NewFeedProcessor returns a new FeedProcessor.
func NewFeedProcessor(feed *model.Feed) *FeedProcessor {
return &FeedProcessor{feed: feed, crawler: false}
func NewFeedProcessor(userID int64, store *storage.Storage, feed *model.Feed) *FeedProcessor {
return &FeedProcessor{userID: userID, store: store, feed: feed, crawler: false}
}

View File

@ -226,3 +226,11 @@ func (s *Storage) MarkAllAsRead(userID int64) error {
return nil
}
// EntryURLExists returns true if an entry with this URL already exists.
func (s *Storage) EntryURLExists(userID int64, entryURL string) bool {
var result int
query := `SELECT count(*) as c FROM entries WHERE user_id=$1 AND url=$2`
s.db.QueryRow(query, userID, entryURL).Scan(&result)
return result >= 1
}