sscan scatter plot has no data

[prjemian]

expected data not shown in the scatter plot from the sscan screen

• upper: PyDM
• lower: MEDM

[prjemian]

Detector PV did not connect. Macros?

[prjemian]

One problem: $() should use curly braces: ${}

Not the real problem, though. The *PV addresses are connected while the *CA addresses are not. Unexpanded macros?

[prjemian]

Also expand macros in the yLabels (and xLabels)

[prjemian]

All channels connect, still no data:

[prjemian]

Of course! The data here are waveforms so we need to be using the PyDMWaveformPlot widget instead.

How can a screen translator be smart enough to know this a priori? MEDM handles both scaler PVs and waveform PVs from the same cartesian plot widget.

@hhslepicka, @mattgibbs , @timmmooney: Any insight how a display screen translator can decide if given PVs come from a waveform PV (so need to use the PyDMScatterPlot widget) or a scalar PV (need to use the PyDMScatterPlot widget)?

[prjemian]

This would be easy if the translator had access to an IOC with those PVs available. Since the translators run in an automation (likely even in CI), can't count on an IOC to be available.

[prjemian]

@MarkRivers : Might you have some ideas how to resolve the problem here?

[MarkRivers]

MEDM handles both scaler PVs and waveform PVs from the same cartesian plot widget.

I think I've only used it with waveform PVs. How does it handle scaler PVs?

[prjemian]

See https://epics.anl.gov/EpicsDocumentation/ExtensionsManuals/MEDM/MEDM.html#CartesianPlot Look for section titled *Kinds of XY Traces*

…

On Wed, Mar 18, 2020, 9:51 AM Mark Rivers ***@***.***> wrote: MEDM handles both scaler PVs and waveform PVs from the same cartesian plot widget. I think I've only used it with waveform PVs. How does it handle scaler PVs? — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#45 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AARMUMCJKEO5VWATNSKFXBLRIDNXTANCNFSM4LOSJ3XQ> .

[prjemian]

not a section, that's a table - very detailed explanation there

[prjemian]

I'm thinking that we'll probably need a new plotting widget in PyDM (for compatibility with MEDM screens) that provides this capability. Until then, the translator should default to PyDMWaveformPlot instead of PyDMScatterPlot.

[prjemian]

Switching (manual edit) to the PyDMWaveformPlot widget, the plot data shows and is the same as MEDM

[MarkRivers]

Until then, the translator should default to PyDMWaveformPlot instead of PyDMScatterPlot.

I agree, I think the Cartesian plot is used much more with waveforms than scalers.

Since most windows have at most one plot window (or if there is more than one they are usually all the same type) then you could have a command line option for the translator that tells it to use PyDMScatterPlot rather than the default PyDMWaveformPlot.

[prjemian]

Good suggestion.