Thread: C question about bitmasks in datetime.c
I see a few cases of this code in src/backend/utils/adt/datetime.c: else if ((fmask & DTK_DATE_M) != DTK_DATE_M) Wouldn't this be clearer as: else if (fmask & DTK_DATE_M) -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 2013-10-01 11:15:36 -0400, Bruce Momjian wrote: > I see a few cases of this code in src/backend/utils/adt/datetime.c: > > else if ((fmask & DTK_DATE_M) != DTK_DATE_M) > > Wouldn't this be clearer as: > > else if (fmask & DTK_DATE_M) That doesn't have the same meaning. The latter is trueif only one bit of DTK_DATE_M is set, while the former requires all bits to be set. Greetings, Andres Freund -- Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training & Services
On Tue, Oct 1, 2013 at 05:17:35PM +0200, Andres Freund wrote: > On 2013-10-01 11:15:36 -0400, Bruce Momjian wrote: > > I see a few cases of this code in src/backend/utils/adt/datetime.c: > > > > else if ((fmask & DTK_DATE_M) != DTK_DATE_M) > > > > Wouldn't this be clearer as: > > > > else if (fmask & DTK_DATE_M) > > That doesn't have the same meaning. The latter is trueif only one bit of > DTK_DATE_M is set, while the former requires all bits to be set. Oh, I see it now --- DTK_DATE_M is not a bit but rather a set of bits, and they are testing if _all_ are set. Thank you! -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +