Re: b32ff1a86 Breaks OSX and FreeBSD

[Gary E. Miller]

Yo Greg!

On Fri, 20 Dec 2019 21:47:29 -0500
Greg Troxel <address@hidden> wrote:

> > The Linux man page says to include time.h, not sys/time.h.
> >
> > "sys/" stuff is for system dependent stuff.  time.h is the standard
> > include.  
> 
> There's a lot of historical stuff in posix, for lots of reasons.  You
> are saying "standard", but you are meaning "what the man pages on the
> linux system in front of me stay", which is not what standard means.

No, I mean the POSIX standard:

    https://www.unix.com/man-page/posix/3/clock_gettime

It says <time.h>

> See POSIX.  That is the standard, and if Linux is different Linux
> should be fixed.

The Linux man page:

    https://linux.die.net/man/2/clock_gettime

Also says <time.h>

> Absolutely gpsd needs to follow POSIX.  If that
> works ok on Linux, it should just be like that.   If not, there needs
> to be a Linux workaround.

I changed it this morning.  <time.h> works for all the pipelines, and
for osX.

No workarounds, just works.

bzed thinbks he has an idea why some of the pipelines randomly fail
"scons check".  Long standing problem, so I would not hold up release
for it, but a fix would be great for release.

As he said on  IRC:

bzed: But I actually think I found a bug in fake.py
bzed: send() on a socket does not care if the data was actually send. it 
returns the number of bytes that were sent, the appication has to take care of 
it
bzed: it should be sendall()
bzed: I'll give that a try in my fork

Fingers crossed.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can't measure it, you can't improve it." - Lord Kelvin

pgpDeaMyNKcth.pgp
Description: OpenPGP digital signature