too long

Question

I am looking to loop through existing .vtt files and read the cue data into a database.

The format of the .vtt files are:

WEBVTT FILE

line1
00:00:00.000 --> 00:00:10.000
‘Stuff’

line2
00:00:10.000 --> 00:00:20.000
Other stuff
Example with 2 lines

line3
00:00:20.00 --> 00:00:30.000
Example with only 2 digits in milliseconds

line4
00:00:30.000 --> 00:00:40.000
Different stuff

00:00:40.000 --> 00:00:50.000
Example without a head line

Originally I was trying to use ^ and $ to be quite regimented with the lines along the lines of: /^(\w*)$^(\d{2}):(\d{2}):(\d{2})\.(\d{2,3}) --> (\d{2}):(\d{2}):(\d{2})\.(\d{2,3})$^(.+)$/ims but I struggled to get this working in the regex checker and resorted to using \s to deal with line start/ends.

Currently I am using the following regex: /(.*)\s(\d{2}):(\d{2}):(\d{2})\.(\d{2,3}) --> (\d{2}):(\d{2}):(\d{2})\.(\d{2,3})\s(.+)/im

This partially works using online regex checkers like: https://regex101.com/r/mmpObk/3 (this example does not pick up multi-line subtitles, but does get the first line which at this point is good enough for my purpose as all subtitles are currently 1 liners). However if I put this into php (preg_match_all("/(.*)\s(\d{2}):(\d{2}):(\d{2})\.(\d{2,3}) --> (\d{2}):(\d{2}):(\d{2})\.(\d{2,3})\s(.+)/mi", $fileData, $matches)) and dump the results I get an array of empty arrays.

What might be different between the online regex and php?

Thanks in advance for any suggestions.

EDIT--- Below is a dump of $fileData and a dump of $matches:

string(341) "WEBVTT FILE

line1
00:00:00.000 --> 00:00:10.000
‘Stuff’

line2
00:00:10.000 --> 00:00:20.000
Other stuff
Example with 2 lines

line3
00:00:20.00 --> 00:00:30.000
Example with only 2 digits in milliseconds

line4
00:00:30.000 --> 00:00:40.000
Different stuff

00:00:40.000 --> 00:00:50.000
Example without a head line"

array(11) {
    [0]=>
        array(0) {}
    [1]=>
        array(0) {}
    [2]=>
        array(0) {}
    [3]=>
        array(0) {}
    [4]=>
        array(0) {}
    [5]=>
        array(0) {}
    [6]=>
        array(0) {}
    [7]=>
        array(0) {}
    [8]=>
        array(0) {}
    [9]=>
        array(0) {}
    [10]=>
        array(0) {}
}

Answer 1

The problem with your regular expression is poor line-ending handling.

You have this at the end: \s(.+)/mi.
This only matches 1 whitespace, but newlines can be 1 or 2 whitespaces.

To fix it, you can use \R(.+)/mi.

It works on the website because it is normalizing your newlines into Linux-style newlines.
That is, Windows-style newlines are (2 characters) and Linux-style are (1 character).

---

Alternativelly, you can try this regular expression:

/(?:line(\d+)\R)?(\d{2}(?::\d{2}){2}\.\d{2,3})\s*-->\s*(\d{2}(?::\d{2}){2}\.\d{2,3})\R((?:[^
]|?
[^
])*)(?:?
?
|$)/i

It looks horrible, but it works.
Note: I'm swapping between \R and because \R matches the literal R inside [].

The data is captured like this:

1. Line number (if present)
2. Initial timestamp
3. Final timestamp
4. Multiline text

You can try it on https://regex101.com/r/Yk8iD1/1

You can use the handy code generator tool to generate the following PHP:

$re = '/(?:line(\d+)\R)?(\d{2}(?::\d{2}){2}\.\d{2,3})\s*-->\s*(\d{2}(?::\d{2}){2}\.\d{2,3})\R((?:[^
]|?
[^
])*)(?:?
?
|$)/i';
$str = 'WEBVTT FILE

line1
00:00:00.000 --> 00:00:10.000
‘Stuff’

line2
00:00:10.000 --> 00:00:20.000
Other stuff
Example with 2 lines

line3
00:00:20.00 --> 00:00:30.000
Example with only 2 digits in milliseconds

line4
00:00:30.000 --> 00:00:40.000
Different stuff

00:00:40.000 --> 00:00:50.000
Example without a head line';

preg_match_all($re, $str, $matches, PREG_SET_ORDER, 0);

// Print the entire match result
var_dump($matches);

You can test it on http://sandbox.onlinephpfunctions.com/code/7f5362f56e912f3504ed075e7013071059cdee7b