Delay time

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2015-2018 Isaac Bernat
+Copyright (c) 2015-2025 Isaac Bernat
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the

README.md

@@ -2,7 +2,7 @@
 1. [Get the subtitles](https://github.com/isaacbernat/netflix-to-srt#get-the-subtitles) (`.xml` dfxp or `.vtt` files from Netflix, YouTube... streaming media services).
     - [From Netflix](https://github.com/isaacbernat/netflix-to-srt#from-netflix)
     - [From YouTube](https://github.com/isaacbernat/netflix-to-srt#from-youtube)
-2. [Convert them](https://github.com/isaacbernat/netflix-to-srt#convert-them-into-srt) into `.srt` files.
+2. [Convert them](https://github.com/isaacbernat/netflix-to-srt#convert-them-into-srt) into `.srt` files (and/or shift timestamps).
 3. [Star this repo ⭐](https://github.com/isaacbernat/netflix-to-srt#star-this-repo)
 
 ## Get the subtitles
@@ -69,6 +69,7 @@ You need FireFox and AdblockPlus Add-On. *not tested on other browsers*
   - Copy your subtitle files in the same directory as `to_srt.py`
     - Or use `-i INPUT_PATH` and `-o OUTPUT_PATH` for custom file locations
   - All `.xml` and `.vtt` files in the input directory will generate a converted `.srt` file on the output one
+- *Optional:* Use `-d DELAY_MS` parameter when running the script to delay all the timestamps by the given number of milliseconds. Negative values shift timestamps backwards. Example: `python to_srt.py -i samples/delays -o samples/delays -d -1500` will take all the eligible files in `samples/delays` and shift the subtitles to be 1.5 seconds earlier than the original version
 - Enjoy! (And **star the repo ⭐** if you liked it ;D)
 
 ## Star this repo
@@ -84,7 +85,9 @@ If you like this project, please **star the repository ⭐**. It's free and it h
 - Thanks for your contribution!
 
 ## Why this repository?
-VLC player could not reproduce that kind of xml subtitles and I could not find any tool that could easily transform the xml files to a suitable format (e.g. SubRip (`.srt`)) in Linux or Mac, so I wrote this script and decided to share. I got a request for WebVTT (`.vtt`) and did the same.
+[VideoLAN's VLC media player](https://www.videolan.org/vlc/) could not reproduce that kind of xml subtitles and I could not find any tool that could easily transform the xml files to a suitable format (e.g. SubRip (`.srt`)) in Linux or Mac, so I wrote this script and decided to share. I got a request for WebVTT (`.vtt`) and did the same.
+
+Similarly, adjusting timestamps in 50ms increments was inconvenient using VLC's hotkeys (G, H and/or J) for large mismatches (e.g. 60 seconds because openings or summaries), so I added the `-d DELAY_MS` parameter so I could "advance" all the subtitles lines easily.
 
 ## TODOs
 - More robust file parsing than just some quick and dirty regexes.

samples/delay/sample_vtt.vtt

@@ -0,0 +1,33 @@
+WEBVTT
+
+NOTE Netflix
+NOTE Profile: webvtt-lssdh-ios8
+NOTE Date: 2019/11/27 21:10:46
+
+NOTE SegmentIndex
+NOTE Segment=591.758 7643@494 61
+NOTE Segment=596.513 12163@8137 99
+NOTE Segment=595.011 9765@20300 77
+NOTE Segment=597.430 8648@30065 68
+NOTE Segment=390.473 4112@38713 34
+NOTE /SegmentIndex
+
+
+
+
+1
+00:01:15.450 --> 00:01:17.577 position:50.00%,middle align:middle size:80.00% line:84.67%
+Que veux-tu ?
+
+2
+00:01:17.660 --> 00:01:21.247 position:50.00%,middle align:middle size:80.00% line:79.33%
+Je veux savoir si une femme
+montera sur le trône de Kattegat.
+
+3
+00:01:22.165 --> 00:01:24.250 position:50.00%,middle align:middle size:80.00% line:84.67%
+Tu veux dire, après le décès de Ragnar ?
+
+4
+00:01:59.503 --> 00:02:00.629 position:50.00%,middle align:middle size:80.00% line:84.67%
+Oui.

samples/delay/sample_vtt_minus_900.srt

@@ -0,0 +1,17 @@
+1
+00:01:14,550 --> 00:01:16,677
+Que veux-tu ?
+
+2
+00:01:16,760 --> 00:01:20,347
+Je veux savoir si une femme
+montera sur le trône de Kattegat.
+
+3
+00:01:21,265 --> 00:01:23,350
+Tu veux dire, après le décès de Ragnar ?
+
+4
+00:01:58,603 --> 00:01:59,729
+Oui.
+

samples/delay/sample_vtt_plus_900.srt

@@ -0,0 +1,17 @@
+1
+00:01:16,350 --> 00:01:18,477
+Que veux-tu ?
+
+2
+00:01:18,560 --> 00:01:22,147
+Je veux savoir si une femme
+montera sur le trône de Kattegat.
+
+3
+00:01:23,065 --> 00:01:25,150
+Tu veux dire, après le décès de Ragnar ?
+
+4
+00:02:00,403 --> 00:02:01,529
+Oui.
+

samples/delay/sample_xml.xml

@@ -0,0 +1,38 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<tt xmlns:tt="http://www.w3.org/ns/ttml" xmlns:ttm="http://www.w3.org/ns/ttml#metadata" xmlns:ttp="http://www.w3.org/ns/ttml#parameter" xmlns:tts="http://www.w3.org/ns/ttml#styling" ttp:cellResolution="40 19" ttp:pixelAspectRatio="1 1" ttp:tickRate="10000000" ttp:timeBase="media" tts:extent="640px 480px" xmlns="http://www.w3.org/ns/ttml">
+<head>
+<ttp:profile use="http://netflix.com/ttml/profile/dfxp-ls-sdh"/>
+<styling>
+<style tts:color="white" tts:fontFamily="monospaceSansSerif" tts:fontSize="100%" xml:id="style0"/>
+</styling>
+<layout>
+<region xml:id="region_00">
+<style tts:textAlign="left"/>
+<style tts:displayAlign="center"/>
+<style tts:backgroundColor="transparent"/>
+</region>
+<region xml:id="region_01">
+<style tts:textAlign="left"/>
+<style tts:displayAlign="center"/>
+<style tts:backgroundColor="transparent"/>
+</region>
+<region xml:id="region_02">
+<style tts:textAlign="left"/>
+<style tts:displayAlign="center"/>
+<style tts:backgroundColor="transparent"/>
+</region>
+</layout>
+</head>
+<body style="style0">
+<div xml:space="preserve">
+<p begin="241910001t" end="280280000t" region="region_00" tts:extent="65.00% 5.33%" tts:origin="17.50% 79.29%" xml:id="subtitle0">Ricky! Ricky! Wake the f***</p>
+<p begin="241910001t" end="280280000t" region="region_01" tts:extent="42.50% 5.33%" tts:origin="20.00% 84.62%" xml:id="subtitle1">up! Ricky, get up!</p>
+<p begin="281115000t" end="301970001t" region="region_00" tts:extent="52.50% 5.33%" tts:origin="22.50% 79.29%" xml:id="subtitle2">Bubbles, f*** off! I'm</p>
+<p begin="281115000t" end="301970001t" region="region_01" tts:extent="45.00% 5.33%" tts:origin="20.00% 84.62%" xml:id="subtitle3">sleeping in today. </p>
+<p begin="302805001t" end="330330000t" region="region_00" tts:extent="75.00% 5.33%" tts:origin="10.00% 79.29%" xml:id="subtitle4">No, you're not. You told me you</p>
+<p begin="302805001t" end="330330000t" region="region_01" tts:extent="60.00% 5.33%" tts:origin="17.50% 84.62%" xml:id="subtitle5">were gonna help me put up</p>
+<p begin="331165000t" end="359117508t" region="region_00" tts:extent="67.50% 5.33%" tts:origin="15.00% 79.29%" xml:id="subtitle6">flyers for the grand opening</p>
+<p begin="591165000t" end="608117508t" region="region_01" tts:extent="62.50% 5.33%" tts:origin="17.50% 84.62%" xml:id="subtitle7">of the Shed and Breakfast.</p>
+</div>
+</body>
+</tt>

samples/delay/sample_xml_minus_900.srt

@@ -0,0 +1,22 @@
+1
+00:00:23,291 --> 00:00:27,128
+Ricky! Ricky! Wake the f***
+up! Ricky, get up!
+
+2
+00:00:27,211 --> 00:00:29,297
+Bubbles, f*** off! I'm
+sleeping in today. 
+
+3
+00:00:29,380 --> 00:00:32,133
+No, you're not. You told me you
+were gonna help me put up
+
+4
+00:00:32,216 --> 00:00:35,011
+flyers for the grand opening
+
+5
+00:00:58,216 --> 00:00:59,911
+of the Shed and Breakfast.

samples/delay/sample_xml_plus_900.srt

@@ -0,0 +1,22 @@
+1
+00:00:25,091 --> 00:00:28,928
+Ricky! Ricky! Wake the f***
+up! Ricky, get up!
+
+2
+00:00:29,011 --> 00:00:31,097
+Bubbles, f*** off! I'm
+sleeping in today. 
+
+3
+00:00:31,180 --> 00:00:33,933
+No, you're not. You told me you
+were gonna help me put up
+
+4
+00:00:34,016 --> 00:00:36,811
+flyers for the grand opening
+
+5
+00:01:00,016 --> 00:01:01,711
+of the Shed and Breakfast.

to_srt.py

@@ -69,11 +69,36 @@ def xml_cleanup_spans_end(span_end_re, text, has_cursive):
     return text
 
 
-def to_srt(text, extension):
+def to_srt(text, extension, delay_ms):
     if extension.lower() == ".xml":
-        return xml_to_srt(text)
-    if extension.lower() == ".vtt":
-        return vtt_to_srt(text)
+        text = xml_to_srt(text)
+    elif extension.lower() == ".vtt":
+        text = vtt_to_srt(text)
+    return shift_srt_timestamp(text, delay_ms)
+
+
+def shift_srt_timestamp(text, delay_ms=0):
+    if not delay_ms:
+        return text
+
+    def shift_time(time_str, shift):
+        h, m, s_ms = time_str.split(":")
+        s, ms = s_ms.split(",")
+        total_ms = int(h)*3600000 + int(m)*60000 + int(s)*1000 + int(ms)
+        new_ms = total_ms + shift
+
+        h = new_ms // 3600000; new_ms %= 3600000
+        m = new_ms // 60000; new_ms %= 60000
+        s = new_ms // 1000; ms = new_ms % 1000
+        return f"{h:02}:{m:02}:{s:02},{ms:03}"
+
+    def replace_timestamp(match):
+        start = shift_time(match[1], delay_ms)
+        end = shift_time(match[2], delay_ms)
+        return f"{start} --> {end}" if start and end else match[0]
+
+    timestamp_regex = r"(\d{2}:\d{2}:\d{2},\d{3}) --> (\d{2}:\d{2}:\d{2},\d{3})"
+    return re.sub(timestamp_regex, replace_timestamp, text)
 
 
 def convert_vtt_time(line):
@@ -205,14 +230,16 @@ def main():
                         help=help_text.format("input", directory))
     parser.add_argument("-o", "--output", type=str, default=directory,
                         help=help_text.format("output", directory))
+    parser.add_argument("-d", "--delay", type=int, default=0,
+                        help="delay all subtitles by the given number of milliseconds")
     a = parser.parse_args()
     filenames = [fn for fn in os.listdir(a.input)
                  if fn[-4:].lower() in SUPPORTED_EXTENSIONS]
     for fn in filenames:
         with codecs.open(u"{}/{}".format(a.input, fn), 'rb', "utf-8") as f:
             text = f.read()
         with codecs.open(u"{}/{}.srt".format(a.output, fn[:fn.rfind('.')]), 'wb', "utf-8") as f:
-            f.write(to_srt(text, fn[-4:]))
+            f.write(to_srt(text, fn[-4:], a.delay))
 
 
 if __name__ == '__main__':