Hi,

my pull request for the APPA driver contains also changes to the output/csv component.

See repo https://github.com/Cymaphore/libsigrok commits

  - 0aa944dbd900d672d75374cc88bbaeee42845884
  - e34d814bd12887ae976555f326973e259e39e1c0
  - 9de55cf5ca48a9d101373b36143599a1f982aa94

These three commits are about the following issues:

  - Adding support for timestamps independent of samplerate based on the
    hosts RTC
  - Adding support for sample_interval as an automatic alternative to
    samplerate
  - Adding support to use SR_DF_META as altarnative data source for the
    samplerate

This is because samplerate / sample_interval works very different in DMMs.
In most cases none is reasonably available anyways, unknown or only
available under certain conditions (log download) or it can even change unanounced during acquision.

I hope these changes bring many advantages for DMM users, they should (please
correct me, I don't use it) not significantly impact the feature in
PulseView.

In PulseView on the other hand, config options time=now_abs and time=now_rel are more or less completely useless (and optional anyways), as far as I can
tell.

However, a couple of issues remain:

  - CSV is not fully locale independent. samplerate works around it by
    scaling to int, but a change for locale independent float formatting
    should be included at some point, I think.
- Time output should be changed to use sr_datafeed_analog instead of manual formatting. This way, it would also be possible to remove insignificant
    digits from time more easily.
- Header can be created before the samplerate/sample_interval arriaves, so
    the format of time should be either configurable as well (pre-set
    scaling) or be able to confirm with anything that follows (by using
    seconds as unit)
- Timestamp is created too late and it would be better to create them at the point we know that the data with the reading now will arriave, but
    for slow devices like DMMs this is ok.

So my questions are:

How do you see this issue?

Is my assesment about pv correct or did I miss something?

What approach would you think I should take on that?

Should I continue with the rest of the points? Any other input or suggestions
on this? Or would it be ok like this for now?

Best regards,
Martin


_______________________________________________
sigrok-devel mailing list
sigrok-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sigrok-devel

Reply via email to