I’ve just increased the scrollback to 10,000 lines on the terminal and tested it with your capture, works great. That will probably be included in 2.2.11 today, if all goes well.
1 Like
The terminal window output already looks great for GPS sentences, so what are you looking for to add?
Sentence checksum verification and simple data validation are easy and something I would be willing to take a stab at (I’m not familiar with python), but once you start looking into assembling the drip-by-drip data into full cycles with 2D/3D fixes with date, time etc, the rabbit hole gets pretty deep pretty quickly - see gpsd.
I’ll like to break the GPS messages down into something a little easier to read. I haven’t started planning it yet, but I think that readability will be quite important.
It should also be a helpful example of a real-world HLA, people should be able to read the source code and learn how it works, so they could modify it or write their own.
We haven’t implemented this yet, but one of the next big things we’ll be doing with the terminal view is support analyzer-specific columns. For example, this would allow a GPS HLA to have a table display with a column for Latitude and a column for longitude.
There is a lot of information in GPS data, so it would be a good exercise for us to see what that’s like for the user. We might need to implement the ability to hide/show/reorder columns, or we might want to make it easy to select details of the table format in the HLA settings.
We’re also considering hooking the python print function to the terminal in the app, so you could write an HLA that just prints the GPS information you’re interested in, instead of seeing all of it all the time. (e.g. filter out all but one GPS sentence)
First of all, thanks for the feedback! We really appreciate it 
I’ll address only the HLA part here.
Good point! We can provide create some python unit tests.
It seems there is a bug in passing the first I2C frame data in a capture to the HLA
We’re looking into this.
With my Si4735 HLA, I notice that when zooming out, the frames that I use to convey a full transaction (write command, read response) become difficult to visually separate multiple transactions joined/squished together.
Looks like we have a bug there… we’ll fix it
As I was developing the Si4735 HLA I noticed the I2C DLL analyzer plugin seems to report “null” in Decoded Protocols.
We noticed that last week, it will be fixed soon.
If an HLA returns a frame data strings containing ‘=’
Are you referring to the results table or the output on the graph (the bubbles)?
We’ll start working on the community marketplace in a couple of weeks. I’m really excited to see some new HLAs there
I’ll address here the rest of your feedback (which is great btw!)
Specifically, V1 was able to set the filter in units of time (ms/us)
Yes, we’ll add that soon
It was also possible to set this on a per channel basis
Our backend supports that, we just need to add that to the UI. I added that to our feature requests board: Apply per-channel glitch filter - Logic 2 - Ideas and Feature Requests - Saleae
Also, V1 would reprocess the current capture when the filter settings were changed
Unfortunately, this is not possible in Logic 2, as we apply the filter during processing (for simplicity and performance).
If you sample at a really slow frequency then it becomes difficult to know where the “real data” is versus the “interpolated”
Good point. We’ll do that as part of supporting dot mode that we’re planning to add
Is there a possibility to allow the user to disable anti-aliasing filtering done on the Analog signals?
I’ll have to ask @markgarrison to answer that
it seems the “analog drawing process” has some aliasing issues
It’s a known bug. We’ll revamp the analog features in a few months including rendering and triggering, and we’ll fix that bug as well!
In the meantime, we’ll publish our bug tracker soon, would that be helpful for you?
Is there any plan to add the marker-dots that V1 had for displaying the sample points on the analog waveform?
Yes
This hamburger menu option on a loaded plugin does not seem to do anything unless you recapture. Is this intended behavior?
The terminal only works in realtime, as we don’t store the data. The main goal of the terminal is to support real-time debugging. Is it a showstopper for the terminal in your opinion? We can re-run the analyzer but it feels redundant
Again, thanks a lot for the feedback!
It is the output on the graph.
Below is an example where the “!=” is meant to be “!=” to indicate, in this case, that the response size is not what was expected (12 bytes) by the HLA.
My HLA’s set_settings returns the below where data.message is a sting containing the string shown in the image, but with “!=” instead of “!=”. I tried a few methods for escaping on my end, but I have not found anything that works, but I did not put too much effort into this yet.
return {
'result_types': {
'transaction': {
'format': '{{data.message}}'
}
}
}
Yea this sounds interesting. Are there any public details on this yet? Like what method of distribution you will use? (maybe this would be some pip repository which the software interfaces with?)
Absolutely, while none of my issues/request are deal-breakers nor am I in any rush for them I think having details on the what has currently been reported would at least avoid some potential for repeating issue reports that others have already submitted.
No this is not an issue at all for me. I had been fiddling with the UI/options to learn more about the features available and that ultimately provoked this question. My only thought for why this would be desirable is if the user had multiple serial streams that they wanted to monitor at the same time, having the ability to filter out one from the other might be nice. But I have never had such a need on my end. Usually I only have a serial terminal and the rest of the interfaces I may monitor are SPI/I2C/etc.
I am not sure how flexible your terminal software component is, but if it is multi instance, maybe there is a possibility to allow the user to create a terminal instance and then assign analyzers to that instance.
1 Like
It is the output on the graph.
We are using Mustache to generate the string and I’d assumed that you need to escape some symbols. Let me know if I can help with that.
Yea this sounds interesting. Are there any public details on this yet?
Nothing official yet. In general, the current plan is to allow users to publish their Github repo and we’ll track new versions automatically. Users will be able to install extensions via the app (similarly to VSCode). What do you think?
I am not sure how flexible your terminal software component is, but if it is multi instance, maybe there is a possibility to allow the user to create a terminal instance and then assign analyzers to that instance.
We’re talking about that option of having tab per analyzer, and then you won’t have to filter anything no matter what.
1 Like
Well that was simple =)
Just needed another set of curly braces around my format:
'format': '{{{data.message}}}'
Sounds good.
1 Like
I second this. Same problem, the SMBus analyzer is still unusable in v2.2.15 on Windows 10 for Logic Pro 8.
@reendael @zainahmed1559, we just released a new version (2.2.16) with a fix. Can you please give it a try?
@rani Just tested it out on MacOS 10.15.4, looks good!
1 Like