On Thu, 2020-01-09 at 22:15 +0000, Marcus Bannerman wrote:
>
> I'm keen to extend sigrok to support the Agilent 16700 series
> of logic analyzer mainframes and I'm looking for a bit of
> support where best to start.
>
> [ ... ]
>
> I think there's two approaches to the implementation, either
> (1) learn to parse the file format that the systems use to
> export to a desktop pc, or (2) use the remote programming
> interface over the network. I'd prefer the latter as I'd like
> to be able to control the system via pulseview.
What makes you feel that these approaches would be strictly
alternative, and only one of them could get implemented? :)
Sure, one of them is more appealing to you, but that does not
exclude the other as a potential intermediate step to get there.
Experience from other device drivers suggests that vendors do
re-use details from communication protocols in file formats and
the other way around. You may learn something that's re-usable in
the process.
Can't speak more specifically (having neither access to the
device, nor knowing their physical attachment nor communication
protocol nor the software's file format). But: Pick one approach
that you think is easier to achieve for now, but don't discard
the other just because one of them exists, or is about to appear
at some time in the future. Whatever increases your chances of
exchanging data with that system is good, is it not? Even if you
achieve remote control of live sessions, you or other users still
may want to import previously recoded captures without repeating
the setup (if at all possible).
> Can you tell me if there's any work on agilent/keysight's file
> format already underway?
I'm not aware of any. You may consider adding an item to the wiki
page, the list of file formats. To raise awareness. Also consider
creating a wiki page for the file format itself, to make details
available which anyone would need to implement the support.
> Also, can you point me to a good example of a driver for a MSO
> or something similar using a network connection? The agilent
> just uses telnet/text so nothing complicated is needed there. I
> couldn't find SCPI-based drivers in my quick search which
> should also be a similar implementation.
The sigrok project provides abstractions for several layers of
communicating to devices. Which means that you may not strictly
have to look for specific examples of an ethernet connected logic
analyzer or scope (which reduces the set of samples that you can
find). Consider the option to "communicate a stream" of requests
and responses which "just happens" to be of serial nature and get
transported via ethernet by coincidence, while the physical
transport might as well be any other medium. Again, not being
familiar with that device of yours and its protocol, I cannot be
more specific.
As a general rule of thumb, you may find more developers and see
faster turnarounds on questions and answers of that kind in IRC
than the ML. And IRC is also where you can get a wiki account to
improve and extend available documentation, regardless of whether
you plan or are able to implement a driver or input module. Any
help is appreciated when it results in an improvement for other
users.
virtually yours
Gerhard Sittig
--
If you don't understand or are scared by any of the above
ask your parents or an adult to help you.
_______________________________________________
sigrok-devel mailing list
sigrok-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sigrok-devel
