Make a regular expression considering the conditions

I'm very bad at regular expressions and don't really understand them. But I want to make my code work great! I have a task to get all the DOI from a large text, but immediately there are problems, namely the difference in the display of the DOI.

DOI: 10.1...
DOI:10.1...
http(s)://dx.doi.org/10.1...
http(s)://doi.org/10.1...
DOI: 11.11111/aaa.111&1111&11

Now I have a regular expression that handles the code just 10.1.... $doiPattern = "/\b(10\.[0-9]{4,}(?:\.[0-9]+)*\/(?:(?![\"&\'])\S)+)\b/";

Unfortunately, my code can't process all the options that I presented. But just pulls out the 10.1... from the full links. https://doi.org/ 11.11111/aaa.111111111

The code I have tried:

$element = "Lorem Lorem Lorem Lorem Lorem Lorem Lorem `https://dx.doi.org/11.11111/aaa.111111111` Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem `https://doi.org/11.11111/aaa.111111111` Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem `DOI: 11.11111/aaa.111&1111&11` Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem `DOI:11.11111/aaa.111&1111&11` Lorem Lorem Lorem "

$doiPattern = "/\b(10\.[0-9]{4,}(?:\.[0-9]+)*\/(?:(?![\"&\'])\S)+)\b/";

$doiData = preg_match_all($doiPattern, $element, $doiMatches);

foreach ($doiMatches[0] as $doiMatch) {
       //...
}

how do I update the regular expression so that it takes into account the conditions. For example, a space between the words DOI:, as well as a links?

Solution

Here is a working solution with an updated regex pattern and PHP code.

<?php

$string = "Lorem Lorem Lorem Lorem Lorem Lorem Lorem https://dx.doi.org/11.11111/aaa.111111111 Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem https://doi.org/11.11111/aaa.111111111 Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem DOI: 11.11111/aaa.111&1111&11 Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem Lorem DOI:11.11111/aaa.111&1111&11 Lorem Lorem Lorem";

// Define regex pattern for DOI
$pattern = '/((https?:\/\/(?:dx\.)?doi\.org\/|DOI:\s*)(\d+(\.\d+)*\/\S+))/i';

// Perform the match
preg_match_all($pattern, $string, $matches);

// Print all matches
print_r($matches[0]);

Output

Array
(
    [0] => https://dx.doi.org/11.11111/aaa.111111111
    [1] => https://doi.org/11.11111/aaa.111111111
    [2] => DOI: 11.11111/aaa.111&1111&11
    [3] => DOI:11.11111/aaa.111&1111&11
)

Explanation of the Regex Pattern Used Above

Protocol Matching:

https?:\/\/: This part checks for a URL starting with either http:// or https://.
Domain and Path:

(?:dx\.)?doi\.org\/: This section matches the domain doi.org optionally preceded by dx..
OR Operator:

|: This symbol means "or". It's like saying "match either this part or that part".
DOI Identifier:

DOI:\s*: This part matches the literal string DOI: followed by optional whitespace characters.
DOI Value:

\d+(\.\d+)*\/\S+: This part captures the actual DOI value: \d+: Matches one or more digits. (\.\d+)*: Allows for zero or more occurrences of a dot followed by one or more digits. This is for version numbers. \/: Matches a forward slash. \S+: Matches one or more non-whitespace characters, which typically represent the remainder of the DOI.
Case Insensitivity:

/i: This flag makes the pattern case-insensitive, so it matches both uppercase and lowercase letters.