Greediness behaving differently in JavaScript?

Question

There was this question which made me realise that greediness of quantifiers is not always the same in certain regex engines. Taking the regex from that question and modifyi

Answer 1

This is not answering why grediness is behaving differently in Javascript but it shows that this is not a bug and it was tested to behave so. I will take as example google's v8 engine. I found a similar example in their tests.

/test/mjsunit/third_party/regexp-pcre.js:

line 1085: res[1006] = /([a]*?)*/;
line 4822: assertToStringEquals("aaaa,a", res[1006].exec("aaaa "), 3176);

https://code.google.com/p/v8/source/browse/trunk/test/mjsunit/third_party/regexp-pcre.js#1085

This code works well for Javascript http://regex101.com/r/nD0uG8 but it does not have the same output for PCRE php and python.

UPDATE: About your question:

It seems that JavaScript's regex engine is interpreting the second * to convert .*? from lazy to greedy

https://code.google.com/p/v8/source/browse/trunk/src/parser.cc#5157

RegExpQuantifier::QuantifierType quantifier_type = RegExpQuantifier::GREEDY;
if (current() == '?') {
    quantifier_type = RegExpQuantifier::NON_GREEDY;
    Advance();
} else if (FLAG_regexp_possessive_quantifier && current() == '+') {
    // FLAG_regexp_possessive_quantifier is a debug-only flag.
    quantifier_type = RegExpQuantifier::POSSESSIVE;
    Advance();
}