Add min and max duration filter

README.md

@@ -23,7 +23,7 @@ any device in podcast client.
 - Supports feeds configuration: video/audio, high/low quality, max video height, etc.
 - mp3 encoding
 - Update scheduler supports cron expressions
-- Episodes filtering (match by title).
+- Episodes filtering (match by title, duration).
 - Feeds customizations (custom artwork, category, language, etc).
 - OPML export.
 - Supports episodes cleanup (keep last X episodes).

cmd/podsync/config_test.go

@@ -37,7 +37,7 @@ timeout = 15
   update_period = "5h"
   format = "audio"
   quality = "low"
-  filters = { title = "regex for title here" }
+  filters = { title = "regex for title here", min_duration = 0, max_duration = 86400}
   playlist_sort = "desc"
   clean = { keep_last = 10 }
   [feeds.XYZ.custom]
@@ -78,6 +78,8 @@ timeout = 15
 	assert.EqualValues(t, "audio", feed.Format)
 	assert.EqualValues(t, "low", feed.Quality)
 	assert.EqualValues(t, "regex for title here", feed.Filters.Title)
+	assert.EqualValues(t, 0, feed.Filters.MinDuration)
+	assert.EqualValues(t, 86400, feed.Filters.MaxDuration)
 	assert.EqualValues(t, 10, feed.Clean.KeepLast)
 	assert.EqualValues(t, model.SortingDesc, feed.PlaylistSort)
 

config.toml.example

@@ -70,7 +70,8 @@ vimeo = [ # Multiple keys will be rotated.
 
   # Optional Golang regexp format.
   # If set, then only download matching episodes.
-  filters = { title = "regex for title here", not_title = "regex for negative title match", description = "...", not_description = "..." }
+  # Duration filters are in seconds.
+  filters = { title = "regex for title here", not_title = "regex for negative title match", description = "...", not_description = "...", min_duration = 0, max_duration = 86400 }
 
   # Optional extra arguments passed to youtube-dl when downloading videos from this feed.
   # This example would embed available English closed captions in the videos.

pkg/feed/config.go

@@ -28,7 +28,7 @@ type Config struct {
 	MaxHeight int `toml:"max_height"`
 	// Format to use for this feed
 	Format model.Format `toml:"format"`
-	// Only download episodes that match this regexp (defaults to matching anything)
+	// Only download episodes that match the filters (defaults to matching anything)
 	Filters Filters `toml:"filters"`
 	// Clean is a cleanup policy to use for this feed
 	Clean Cleanup `toml:"clean"`
@@ -49,6 +49,8 @@ type Filters struct {
 	NotTitle       string `toml:"not_title"`
 	Description    string `toml:"description"`
 	NotDescription string `toml:"not_description"`
+	MinDuration    int64  `toml:"min_duration"`
+	MaxDuration    int64  `toml:"max_duration"`
 	// More filters to be added here
 }
 

services/update/matcher.go

@@ -8,14 +8,27 @@ import (
 	log "github.com/sirupsen/logrus"
 )
 
+func matchDurationFilter(duration_limit int64, episode_duration int64, operator string, logger log.FieldLogger) bool {
+	if duration_limit != 0 {
+		if operator == "min" && episode_duration < duration_limit {
+			logger.Info("skipping due to duration filter")
+			return false
+		} else if operator == "max" && episode_duration > duration_limit {
+			logger.Info("skipping due to duration filter")
+			return false
+		}
+	}
+	return true
+}
+
 func matchRegexpFilter(pattern, str string, negative bool, logger log.FieldLogger) bool {
 	if pattern != "" {
 		matched, err := regexp.MatchString(pattern, str)
 		if err != nil {
 			logger.Warnf("pattern %q is not a valid")
 		} else {
 			if matched == negative {
-				logger.Infof("skipping due to mismatch")
+				logger.Infof("skipping due to regexp mismatch")
 				return false
 			}
 		}
@@ -40,6 +53,14 @@ func matchFilters(episode *model.Episode, filters *feed.Filters) bool {
 	if !matchRegexpFilter(filters.NotDescription, episode.Description, true, logger.WithField("filter", "not_description")) {
 		return false
 	}
+
+	if !matchDurationFilter(filters.MinDuration, episode.Duration, "min", logger.WithField("filter", "min_duration")) {
+		return false
+	}
+
+	if !matchDurationFilter(filters.MaxDuration, episode.Duration, "max", logger.WithField("filter", "max_duration")) {
+		return false
+	}
 
 	return true
 }