From 43b3a99e29b34cc768b77e2adb5d6106afd1e01f Mon Sep 17 00:00:00 2001
From: Andreas <andreasv@fsmpi.rwth-aachen.de>
Date: Wed, 5 Oct 2016 01:46:18 +0200
Subject: [PATCH] finished sorter, now correctly sorting most of our old date
---
sorter.py | 54 +++++++++++++++++++++++++++++-------------------------
1 file changed, 29 insertions(+), 25 deletions(-)
diff --git a/sorter.py b/sorter.py
index 0638dcd..7830402 100644
--- a/sorter.py
+++ b/sorter.py
@@ -14,7 +14,7 @@ def sort_log():
FROM sortlog
JOIN lectures ON lectures.id = sortlog.lecture_id
JOIN courses ON courses.id = lectures.course_id
- ORDER BY sortlog.`when`
+ ORDER BY sortlog.`when` DESC
'''))
@@ -47,27 +47,25 @@ def sort_now():
filepath = coursepath + '/' + f
# filenames: <handle>-<sorter>-<format>.mp4
# sorter musst be found with fuzzy matching. musst be one or more of the following: (inside the loop)
- splitfilename = f.replace('_','-').split('-')
+ splitfilename = f.replace('_','-').replace(' ','-').split('-')
if not os.path.splitext(f)[1] == '.mp4':
continue
data = {'keywords': []}
# parse the file name and save all data in 'data'
for s in splitfilename:
+ s = s.replace('.mp4','')
#-<YYMMDD> (date)
#-<HHMM> (time)
#-<keyword>
- # Looking for keywords in: id,title,speaker,comment, comma seperated list in internal starting with "tags:" (in this order). first match counts
- if len(s) == 6:
- try:
+ # Looking for keywords in: id,title,speaker,comment, comma seperated list in internal
+ try:
+ if len(s) == 6:
data['date'] = datetime.strptime(s,'%y%m%d').date()
- except ValueError:
- pass
- elif len(s) == 4:
- try:
+ elif len(s) == 4:
data['time'] = datetime.strptime(s,'%H%M').time()
- except ValueError:
- pass
- else:
+ else:
+ data['keywords'].append(s)
+ except ValueError:
data['keywords'].append(s)
# try to match the file on a single lecture
matches = []
@@ -75,16 +73,19 @@ def sort_now():
# first try date and time (if one of them is set)
if 'date' in data:
for l in lectures:
- if (l['time'].date() == data['date']) or (('time' in data) and l['time'] == datetime.combine(data['date'],data['time'])) :
+ if ((l['time'].date() == data['date']) and not ('time' in data)) or (('time' in data) and l['time'] == datetime.combine(data['date'],data['time'])) :
matches.append(l)
# if we can't match based on date and time, we have to match keywords
if ((len(matches) != 1) and (len(data['keywords']) > 0)):
found = False
for field in ['id','title','speaker','comment','internal']:
#only test lectures with the correct date/time
+ if len(matches) == 0:
+ matches.extend(lectures)
for l in matches:
for k in data['keywords']:
- if (k == l[field]) or (k in str(l[field])):
+ # first test for exact match, else make it asci and try substring test
+ if (k == l[field]) or (str(k).lower() in str(l[field]).lower().replace('ä','ae').replace('ü','ue').replace('ö','oe').replace('ß','ss') ):
found = True
matches = [l]
if found:
@@ -98,20 +99,23 @@ def sort_now():
# now match the format
for i in formats:
#we match the last part of the file name without the extension
- formatstring = splitfilename[-1].split('.',1)[0]
- if splitfilename[-1].split('.',1)[0] in i['keywords'].split(','):
+ formatstring = splitfilename[-1].split('.',1)[0].lower()
+ if formatstring in i['keywords'].replace(',',' ').split(' '):
data['format'] = i['id']
break
# if we found the format, insert the video
- if 'format' in data:
- modify('BEGIN')
- video_id = modify('INSERT INTO videos_data (lecture_id,visible,path,video_format,title,comment,internal,file_modified,time_created,time_updated,created_by,hash,file_size) VALUES (?,0,?,?,"","","",?,?,?,?,"",?)',matches[0]['id'],c['handle']+'/'+f,data['format'],datetime.now(),datetime.now(),datetime.now(),session['user']['givenName'],os.stat(coursepath+'/'+f).st_size)
- query('INSERT INTO sortlog (lecture_id,video_id,path,`when`) VALUES (?,?,?,?)',matches[0]['id'],video_id,c['handle']+'/'+f,datetime.now())
- modify('COMMIT')
- else:
- print('unknown format',formatstring,[data,f])
- else:
- print('failed',[data,f])
+ if not 'format' in data:
+ data['format'] = 0
+ modify('BEGIN')
+ video_id = modify('INSERT INTO videos_data (lecture_id,visible,path,video_format,title,comment,internal,file_modified,time_created,time_updated,created_by,hash,file_size) VALUES (?,0,?,?,"","","",?,?,?,?,"",?)',matches[0]['id'],c['handle']+'/'+f,data['format'],datetime.now(),datetime.now(),datetime.now(),session['user']['givenName'],os.stat(coursepath+'/'+f).st_size)
+ query('INSERT INTO sortlog (lecture_id,video_id,path,`when`) VALUES (?,?,?,?)',matches[0]['id'],video_id,c['handle']+'/'+f,datetime.now())
+ modify('COMMIT')
+# for debuging only
+# else:
+# d = []
+# for m in matches:
+# d.append(m['id'])
+# print('failed',{"data":data,"path":f,"results":d})
if 'ref' in request.values:
--
GitLab