Bug #65547 [Opn->Ver]: Default value for sunrise/sunset zenith still wrong

Edit report at https://bugs.php.net/bug.php?id=65547&edit=1 ID: 65547 Updated by: cmb@php.net Reported by: df1986 at aol dot com Summary: Default value for sunrise/sunset zenith still wrong -Status: Open +Status: Verified Type: Bug Package: Date/time related Operating System: All PHP Version: 5.4.19 Block user comment: N Private report: N New Comment: Yes, you're right. <http://aa.usno.navy.mil/faq/docs/RST_defs.php> explains why it should be 90° 50'. Previous Comments: ------------------------------------------------------------------------ [2013-08-25 09:31:04] df1986 at aol dot com Description: ------------ I have discovered https://bugs.php.net/bug.php?id=49448 this bug report filed several years ago which, quite correctly, points out that php's default zenith is 90 + 35/60. However, the official algorithm from the 1990 Almanac for Computers published by the Nautical Almanac Office at the United States Naval Observatory (which can be found at http://williams.best.vwh.net/sunrise_sunset_algorithm.htm) quite clearly says that the correct value is 90+50/60. The old bug was closed as bogus with a message saying that "the algorithm code we use clearly says 35/60". That statement is nothing more than a tautology. It says "the code should be 35/60 because if you look at the code, it says 35/60", and does nothing to answer where this value came from, or why it differs from the canonical value. Some commenters had noted this, but there was no response. As such, I'd like to reopen this bug and see it correctly fixed, rather than simply swept under the carpet. Test script: --------------- echo ini_get("date.sunset_zenith"), "\n", ini_get("date.sunrise_zenith"), "\n", (90 + 50/60), "\n"; Expected result: ---------------- 90.833333333333 90.833333333333 90.833333333333 Actual result: -------------- 90.583333 90.583333 90.833333333333 ------------------------------------------------------------------------ -- Edit this bug report at https://bugs.php.net/bug.php?id=65547&edit=1