Re: [gpsd-dev] [PATCH 1/2] gpsprof: fix replot

[Fred Wright]

On Sun, 28 May 2017, Robin H. Johnson wrote:

> gpsprof's replot functionality feeds JSON back to a session, but relies
> on the oldstyle keydict to read fields out for analysis.
>
> Refactor gps.gps.read() to have common code for handling JSON input &
> using the shim-setter, then wire gpsprof to feed JSON into the new
> function.
[...]

This change breaks xgps (and perhaps other things):
========================================================================
MacPro:gpsd fw$ xgps -r95 time
Traceback (most recent call last):
  File 
"/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/gi/overrides/GLib.py",
 line 735, in <lambda>
    func_fdtransform = lambda _, cond, *data: callback(channel, cond, *data)
  File "/usr/local/bin/xgps", line 867, in handle_response
    if self.daemon.read() == -1:
  File 
"/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/gps/gps.py",
 line 264, in read
    self.interpret_json(self.response)
AttributeError: 'gps' object has no attribute 'interpret_json'
========================================================================

Also, when testing gpsprof, I noticed some questionable lines with "nan"
in the output.

Fred Wright