commit 59a6d1db9e0439c5ae396fc999807827c65e92a1
parent 469cfc80b647f86b2acfa17ccf5804d113f6bc17
Author: gearsix <gearsix@tuta.io>
Date: Wed, 5 May 2021 17:27:21 +0100
improved timestamp regex; entire line is searched for timestamp
Diffstat:
2 files changed, 12 insertions(+), 8 deletions(-)
diff --git a/TODO b/TODO
@@ -13,7 +13,7 @@
[ ] user install
- installs ffmpeg/youtube-dl if not found ?
- just use corresponding python libaries
-[ ] detect timestamp anywhere in line (not just at start)
+[*] detect timestamp anywhere in line (not just at start)
[x] detect tracks from missing timestamps
- Could be done using: _ffmpeg -i audio.mp3 -af silencedetect=d=0.5 -f null -_
but finding the right parameters for the silencedetect filter (see _man ffmpeg-filters)
diff --git a/sya.py b/sya.py
@@ -5,7 +5,7 @@ import subprocess
import re
import os
-Timestamp = re.compile(':?\d\d')
+Timestamp = re.compile('(:?\d{1,2}){3}')
class TracklistItem:
def __init__(self, timestamp, title):
@@ -64,9 +64,15 @@ def load_tracklist(path):
def parse_tracks(tracklist):
tracks = []
for lcount, line in enumerate(tracklist):
- sline = line.split(' ', maxsplit=1)
- if Timestamp.match(sline[0]):
- tracks.append(TracklistItem(sline[0], sline[1].strip()))
+ sline = line.split(' ')
+ timestamp = sline[0]
+ for l in sline: # check line in case timestamp is in another element
+ if Timestamp.match(l):
+ timestamp = l
+ sline.remove(l)
+ title = ' '.join(sline).strip()
+ if Timestamp.match(timestamp):
+ tracks.append(TracklistItem(timestamp, title))
else:
print('missing timestamp: ', line)
return tracks
@@ -82,10 +88,8 @@ def split_tracks(ffmpeg, audio_fpath, tracks, outpath):
cmd = ['ffmpeg', '-v', 'quiet', '-stats', '-i', 'audio.mp3',
'-f', 'null', '-']
ret = subprocess.run(cmd, stderr=subprocess.PIPE)
- # do some nasty string manip. to extract length printed to stderr
length = str(ret.stderr).split('\\r')
- print(len(length))
- print(length[len(length)-1])
+ # some nasty string manip. to extract length (printed to stderr)
length = length[len(length)-1].split(' ')[1].split('=')[1][:-3]
for i, t in enumerate(tracks):
