From 243c410775724e10237a19dd81cea9132739c736 Mon Sep 17 00:00:00 2001
From: kolaczyn <kolaczyn@protonmail.com>
Date: Thu, 25 Jun 2020 01:50:13 +0200
Subject: [PATCH] finished cleaning up

---
 clean.sh      |  5 +++++
 gen-script.py | 39 ++++++++++++---------------------------
 run.sh        |  3 +--
 3 files changed, 18 insertions(+), 29 deletions(-)
 create mode 100755 clean.sh

diff --git a/clean.sh b/clean.sh
new file mode 100755
index 0000000..c9ad2af
--- /dev/null
+++ b/clean.sh
@@ -0,0 +1,5 @@
+#!/bin/bash
+
+rm list.txt rip.sh rm output.mp4
+rm -r ../out
+mkdir ../out
diff --git a/gen-script.py b/gen-script.py
index 73b103d..3911eb2 100755
--- a/gen-script.py
+++ b/gen-script.py
@@ -1,13 +1,9 @@
 import re
-import sys
 from pathlib import Path
 import os
-import time
-from datetime import datetime, timedelta
-# jargon explanation: rip - 10-20 second part of a video
-# # TODO:
-# remove not used imports
-# searched string is hardcoded. fix that
+
+search = '(pesky|bird)' # the searched phrase
+
 # make it so the script rips ~10 second before the subtitle and 10 seconds after
 # implement overlaping chunks recognition
 # use regex instead of hardcoding
@@ -17,32 +13,21 @@ file_names = []
 
 for file in os.listdir(path):
     file_names.append(file)
-file_names.sort() # not necessary, but it makes the generated script look better
+file_names.sort()
 
-data =[]
+data = []
 
 for f in file_names:
     file = open(path + f)
     for i, line in enumerate(file):
-        if re.search('(pesky|bird)', line):
-            data.append({'fname': f, 'beg': prev[0:8], 'end': prev[17:25], 'desc': line}) # get file name, time stamps and desciption
-        prev = line
+        if re.search(search, line):
+            data.append({'fname':f, 'beg':prev[0:8], 'end':prev[17:25], 'desc':line}) # get file name, time stamps and desciption
+        prev = line # a line just before text is a timestamp
 
-# datestring1 = data[-1][1]
-# datestring2 = data[-1][2]
-#
-# #!!!! t1 = time.strptime(datestring1, "%M:%S")
-# #!!!! t2 = time.strptime('00:12', '%M:%S')
-# #!!!! start = time.mktime(t1)-time.mktime(t2)
-# #!!!! print
-#
-# TODO I am using enumerato to make it so different rips from the same video
-# have a different name and are in order. the order won't work if there is more
-# than 9 rips from the same video
 file_rip = open('rip.sh', 'w')
 file_list = open('list.txt', 'w')
 for (i, d) in enumerate(data):
-      name = d['fname'][:2] # get a name without an extension, eg. 02.srt -> 02
-      outname = '../out/' + name + '-' + str(i) + ".mp4" # name of a rip, eg ../out/02-4.mp4
-      file_list.write("file '" + outname + "''\n") # TODO use {} or %s instead of +
-      file_rip .write('ffmpeg -ss ' + d['beg'] + ' -i ../original/' + name + '.mp4 -to ' + '00:00:12' + ' -c copy ' +outname+'\n')
+      name = d['fname'][:2] # get a name without an extension
+      outname ="../out/{}-{:0>2}.mp4".format(name, i)
+      file_list.write("file '{}'\n".format(outname))
+      file_rip .write('ffmpeg -ss {} -i ../original/{}.mp4 -to 00:00:12 -c copy {}\n'.format(d['beg'], name, outname))
diff --git a/run.sh b/run.sh
index 7fe374e..1c80eec 100755
--- a/run.sh
+++ b/run.sh
@@ -4,6 +4,5 @@
 # is an URL of a youtube playlist/ video you want to edit
 
 python gen-script.py
-bash rip.bash
+bash rip.sh
 ffmpeg -f concat -safe 0 -i list.txt -c copy output.mp4 # splice all the clips
-rm list.txt rip.sh # cleanup