p RUBY_VERSION
r = /(?<order>\d+)\n(?<start>[\d:,]+)\s+-{2}\>\s+(?<end>[\d:,]+)\n(?<text>[\s\S]*?(?=\n{2}|$))/
DATA.read.scan(r).each do |match|
p %i[order start end text].zip(match).to_h
end
__END__
1
00:01:35,418 --> 00:01:43,418
In 2015 an American archeological company
bought a military manuscript from the Han Dynasty
2
00:01:45,293 --> 00:01:49,501
It said
Roman soldiers went to China 2,000 years ago
3
00:01:49,918 --> 00:01:53,084
and the ancient city of Regum was bulit
4
00:01:53,918 --> 00:01:57,084
The archeological community believed
it was a fake