Should we have a version of bin_test that runs the installed code?

2018-12-12 Thread Hal Murray via devel
The current code knows the version time-stamp and compares against that. That time-stamp comes from configure and gets lost after you configure again or delete that directory. I think I'd be happy if it just sanity checked the output. Extra credit for testing to see if all the time-stamps a

Re: Why is ntpclients/ntploggps.py in ntpsec rather than gpsd?

2018-12-12 Thread Hal Murray via devel
> Is there any value in my adding a waf check for the "gps" python module and > skipping the checking/installation of ntpclients/ntploggps.py if it's not > found, similar to what I did for the "curses" python module and ntpmon? Maybe, but please wait until we reach some sort of consensus. --

Re: Why is ntpclients/ntploggps.py in ntpsec rather than gpsd?

2018-12-12 Thread Matthew Selsky via devel
Is there any value in my adding a waf check for the "gps" python module and skipping the checking/installation of ntpclients/ntploggps.py if it's not found, similar to what I did for the "curses" python module and ntpmon? It would take only a few minutes. Thanks, -Matt _

Re: Why is ntpclients/ntploggps.py in ntpsec rather than gpsd?

2018-12-12 Thread Gary E. Miller via devel
Yo Hal! On Wed, 12 Dec 2018 14:25:17 -0800 Hal Murray wrote: > >> Is there reason not to move it to gpsd? And most gpsd could care > >> less? Its only use is with ntpd. > > > Maybe because gpsd does not use it? > > Maybe somebody would use it if they knew about it. I can't see any reaso

Re: Why is ntpclients/ntploggps.py in ntpsec rather than gpsd?

2018-12-12 Thread Hal Murray via devel
>> Is there reason not to move it to gpsd? And most gpsd could care >> less? Its only use is with ntpd. > Maybe because gpsd does not use it? Maybe somebody would use it if they knew about it. Is there any problem with putting it in gpsd? How stable is the API between ntploggps and gpsd rela

Re: Why is ntpclients/ntploggps.py in ntpsec rather than gpsd?

2018-12-12 Thread Richard Laager via devel
On 12/12/18 2:59 PM, Gary E. Miller via devel wrote: > On Tue, 11 Dec 2018 22:52:07 -0800 > Hal Murray via devel wrote: >> Is there reason not to move it to gpsd? And most gpsd could care >> less? Its only use is with ntpd. > > Maybe because gpsd does not use it? >> If we want to keep it here,

Re: puthon curses on NetBSD

2018-12-12 Thread Richard Laager via devel
On 12/11/18 1:37 PM, Gary E. Miller via devel wrote: > First we need to start including better dependency info for different > distros in the INSTALL file. Without that we don't know which packages > to look at. I wasn't aware of, until this thread, or had forgotten the ntpmon curses requirement.

Re: Why is ntpclients/ntploggps.py in ntpsec rather than gpsd?

2018-12-12 Thread Gary E. Miller via devel
Yo Hal! On Tue, 11 Dec 2018 22:52:07 -0800 Hal Murray via devel wrote: > James Browning said: > > I think because it started out as a helper script for ntpviz. You > > should blame me for the name though I moved it last. > > Is there reason not to move it to gpsd? And most gpsd could care >