SRT-Link

SRT-Link is a SubRip cleaner that allows to filter, merge, and order SRT file sections. It is written in Python, requires only standard libraries and no 3rd party dependencies.

Installation

Tested and developed on python-3.10

git clone:

git clone https://github.com/EliaOnceAgain/SRT-Link.git
cd SRT-Link/
pip install .

Usage

Provide a SubRip file as input to run with default configs and output to stdout

python -m srt_link input.srt

Check the --help section for more information and custom configs

$ python -m srt_link --help

usage: srt_link [-h] [-o OUTPUT_FILE] [--parentheses] [--curly-brackets] [--angle-brackets] 
                [--square-brackets] [--max-digits MAX_DIGITS] [--min-duration MIN_DURATION] 
                [--faces FACES_TO_SKIP] [--text TEXT_TO_SKIP] [--debug] input_file

SRT-Link: filter, merge, and order SubRip file sections

positional arguments:
  input_file                            srt input file path

options:
  -h, --help                            show this help message and exit
  -o OUTPUT_FILE, --output OUTPUT_FILE  output file [default=stdout]
  --parentheses                         filter parentheses [default=True]
  --curly-brackets                      filter curly brackets [default=True]
  --angle-brackets                      filter angle brackets [default=True]
  --square-brackets                     filter square brackets [default=True]
  --max-digits MAX_DIGITS               max number of digits allowed [default=10]
  --min-duration MIN_DURATION           min section duration in seconds [default=0.3]
  --faces FACES_TO_SKIP                 comma separated faces to filter
  --text TEXT_TO_SKIP                   comma separates text to filter
  --debug                               print debug logs

Example usage

To filter a sections with face face-1 or face 2, or contains the text does not support, and remove any and all parenthesis and brackets, run:

python3 -m srt_link input.srt --faces "face-1,face 2" --text "does not support"

Provided the following example input.srt

$ cat input.srt
1
00:00:01,000 --> 00:00:02,000
<font face="face-1">Try 1</font>

2
00:00:02,000 --> 00:00:03,000
<font face="face 2">Try 2</font>

3
00:00:03,000 --> 00:00:04,000
<font face="pass">this media player does not support some media file</font>

4
00:00:04,000 --> 00:00:05,000
<font face="pass">{\an8}Lorem ipsum dolor sit amet(584.445,26.5,584.445,26.5)(0,240,\clip(562.797,8,606.094,45)</font>

The result would be:

$ python3 -m srt_link --faces "face-1,face 2" --text "does not support" test.srt
1
00:00:04,000 --> 00:00:05,000
Lorem ipsum dolor sit amet

Issues

Not working as expected? Open an issue and upload a minimal SubRip file that reproduces the issue and the output with debug logs using the cli flag --debug

How it works

The solution is based on a doubly linked list, here's an overview of the steps to add a new section to the linked list:

• A new section is read from the input SubRip file
• The section is filtered based on the provided configs
• Add the section to the end of the linked list
• If the new section starts before the tail section ends (tail = last section in the linked list)
  • Resolve the overlap issue based on one of the cases below
  • If the new section begins before tail, delegate the new section to the previous section (see last case below)

In all cases below, (A) is tail, and (B) is the new section to be added.

### New section starts at the same time as tail
# Case 1: Exact overlap
> Initial state
|-------------|     (A)
|-------------|     (B)
> Final state
|-------------|
     (A+B)

# Case 2: New section ends before tail
> Initial state
|-------------|     (A)
|--------|          (B)
> Final state:
|--------|----|
   (A+B)   (A)
   
# Case 3: new section ends after tail
> Initial state
|--------|          (A)
|-------------|     (B)
> Final state:
|--------|----|
   (A+B)   (B)

### New section starts before tail ends
# Case 1: same end
> Initial state
|-------------|     (A)
       |------|     (B)
> Final state
|------|------|
  (A)    (A+B)

# Case 2: new section ends after tail
> Initial state
|---------|         (A)
     |---------|    (B)
> Final state
|----|----|----|
 (A)  (A+B) (B)
 
# Case 3: new section ends before tail
> Initial state
|------------|      (A)
    |----|          (B)
> Final state
|---|----|---|
 (A) (A+B) (A)

### New section starts before tail
> Initial state
          |------------|    (A)
|--------------|            (B)
> Final state
|--------|------|-----|
 (prev+B)  (A+B)  (A)