Moved sorting algorithm out of sort_now

sorter.py

@@ -52,44 +52,20 @@ def schedule_thumbnail(lectureid, filePath=None):
 	data = '{"lectureid": "'+str(lectureid)+'", "path": "'+path+'"}'
 	query('INSERT INTO jobs (type, data, time_created) VALUES ("thumbnail", ?, ?)', data, datetime.now());
 
-@app.route('/internal/sort/now')
-@mod_required
-@sched_func(600)
-def sort_now():
-	courses = query('SELECT * FROM courses')
-	formats = query('SELECT * FROM formats ORDER BY prio')
-	for course in courses:
-		modify('BEGIN')
-		for mountpoint in config['VIDEOMOUNT']:
-			existingvideos = query('SELECT videos.path FROM videos JOIN lectures ON (videos.lecture_id = lectures.id) WHERE lectures.course_id = ?',course['id'])
-			knownerrors = query('SELECT sorterrorlog.path FROM sorterrorlog WHERE sorterrorlog.course_id = ?',course['id'])
-			ignorefiles = existingvideos + knownerrors
-			lectures = query('SELECT * from lectures where course_id = ?',course['id'])
-			coursepath = mountpoint['mountpoint']+course['handle']
-			try:
-				files = os.listdir(coursepath)
-			except FileNotFoundError:
-				files = []
-			for filename in files:
-				try:
-					# if the video is in the table "videos" already (with the correct course), skip it
-					ignore = False
-					for file_to_ignore in ignorefiles:
-						# path is something like
-						# vpnonline/08ws-swt/08ws-swt-081118.mp4
-						if os.path.basename(filename) == os.path.basename(file_to_ignore['path']):
-							ignore = True
-							break
-					if ignore:
-						continue
-					filepath = coursepath + '/' + filename
+def sort_file(filename, course=None, lectures=None):
 	# filenames: <handle>-<sorter>-<format>.mp4
 	# "sorter" musst be found with fuzzy matching. "sorter" musst be one or more of the following types: (inside the loop)
 	
 	# '_' and ' ' are handled like '-'
 	splitfilename = filename.replace('_','-').replace(' ','-').split('-')
-					if not	os.path.splitext(filename)[1] == '.mp4':
-						continue
+	if not course:
+		handle = '-'.join(splitfilename[:2])
+		courses = query('SELECT * FROM courses WHERE handle = ?', handle)
+		if not courses:
+			return []
+		course = courses[0]
+	if not lectures:
+		lectures = query('SELECT * from lectures where course_id = ?', course['id'])
 	# we save all extraced data in a dict
 	data = {'keywords': []}
 	# parse the file name and save all data in 'data'
@@ -143,20 +119,56 @@ def sort_now():
 			if found:
 				break
 	# now we should have found exactly one match
+	return matches
+
+@app.route('/internal/sort/now')
+@mod_required
+@sched_func(600)
+def sort_now():
+	courses = query('SELECT * FROM courses')
+	formats = query('SELECT * FROM formats ORDER BY prio')
+	for course in courses:
+		modify('BEGIN')
+		for mountpoint in config['VIDEOMOUNT']:
+			existingvideos = query('SELECT videos.path FROM videos JOIN lectures ON (videos.lecture_id = lectures.id) WHERE lectures.course_id = ?',course['id'])
+			knownerrors = query('SELECT sorterrorlog.path FROM sorterrorlog WHERE sorterrorlog.course_id = ?',course['id'])
+			ignorefiles = existingvideos + knownerrors
+			lectures = query('SELECT * from lectures where course_id = ?',course['id'])
+			coursepath = mountpoint['mountpoint']+course['handle']
+			try:
+				files = os.listdir(coursepath)
+			except FileNotFoundError:
+				files = []
+			for filename in files:
+				try:
+					# if the video is in the table "videos" already (with the correct course), skip it
+					ignore = False
+					for file_to_ignore in ignorefiles:
+						# path is something like
+						# vpnonline/08ws-swt/08ws-swt-081118.mp4
+						if os.path.basename(filename) == os.path.basename(file_to_ignore['path']):
+							ignore = True
+							break
+					if ignore:
+						continue
+					filepath = coursepath + '/' + filename
+					if not os.path.splitext(filename)[1] == '.mp4':
+						continue
+					matches = sort_file(filename, course=course, lectures=lectures)
 					dbfilepath = mountpoint['prefix']+course['handle']+'/'+filename
 					if len(matches) == 1:
 						# now match the format
+						splitfilename = filename.replace('_','-').replace(' ','-').split('-')
+						# default format is "unknown", with id 0
+						fmt = 0
 						for videoformat in formats:
 							#we match the last part of the file name without the extension
 							formatstring = splitfilename[-1].split('.',1)[0].lower()
 							if formatstring in videoformat['keywords'].replace(',',' ').split(' '):
-								data['format'] = videoformat['id']
+								fmt = videoformat['id']
 								break
-						# default format is "unknown", with id 0
-						if not 'format' in data:
-							data['format'] = 0
 						# insert the video into videos_data and log
-						insert_video( matches[0]['id'], dbfilepath, filepath, data['format'])
+						insert_video( matches[0]['id'], dbfilepath, filepath, fmt)
 					else:
 						# if we couldn't match the video on exactly one lecture, log an error
 						matches_id = []