Date.parse(0) returns midnight of 2000, why?

Question

When I try to Date.parse() an integer or string 0, it returns 946681200000, which translates to a date of:

Sat Jan 01 2000 00:00:00 GMT+0100 (CET)

Why?

I would assume that the parser interprets the single zero as a year 2000, but the specs say nothing about single-character year definition - both RFC 2822 and ISO 8601 require a four-character year in the string.

I would like to better understand how the string '0' is parsed into a Date, why is it accepted as a valid Date (should it not be NaN or some such?) and why the year 2000 is chosen instead of for example 1900.

Update

After some trial & error, I discovered that the single number is in fact interpreted differently in different numeric ranges.

• 0 - 12: A month of year 2000
• 13 - 31: NaN
• 32 - 49: A year + 2000, with all the other values set to defaults
• 50 - 99: A year + 1950, with all the other values set to defaults
• 100 - ??: A year, with all the other values set to defaults

Answer 1

the specs say nothing about single-character year definition

The spec says:

If the String does not conform to that format the function may fall back to any implementation-specific heuristics or implementation-specific date formats.

For V8 specifically, see this bug report on unpredictable results when called with a single number. You can also read the source directly (dateparser.cc, dateparser.h, dateparser-inl.h).