Unified Software for Probe Testing and, eventually, deployment

Used to announce major milestones for WMT projects.
hamishmb
Posts: 1891
Joined: 16/05/2017, 16:41

Unified Software for Probe Testing and, eventually, deployment

Post by hamishmb »

Hi everyone.

I have recently created a github repository for us at https://github.com/wmtprojectsteam/rivercontrolsystem, and will add people as soon as possible after they have made accounts. This allows better version tracking of our source code, and easier collaboration.

While thinking about the test software for the solid-state relay (allows a Pi to turn a pump on and off), I realised I could save us a great deal of time and effort later if I combine our existing pieces of software into a "library" of sorts, with many python modules and several standalone programs. So far I have adapted, converted, and tested the resistance probe software, and the hall effect device software, and they now work with this package/module system, saving a lot of effort and allowing me to refactor and de-duplicate our code.

Standalone Programs so far:
  • Hall effect device software (slightly improved feature-wise).
  • Resistance probe software (fault detecting is now fixed and enabled, though only partially functional).
  • Capacitance Probe software (slightly improved feature-wise).
Classes/Structures so far:
  • Motor class to control motors (tested).
  • Resistance Probe class (tested).
  • Hall effect device class (tested).
  • Capacitance Probe class (tested).
  • Monitor threads for hall effect, resistance probe and capacitance probe (tested).
  • Test code for relay (tested).
I am yet to write the code for the relay, but that'll be coming in the next few days. UPDATE: Added - see above.

Because of the major changes to the software layout, all versions have been re-set to 1.0, and they will all now share a version number in keeping with the rest of the software suite. Old versions of the software won't be maintained by me any more, so we should try to bring the Pi up to date soon, though I'd like to finish the fault detection code for the resistance probe first.

I think you'll all like these changes, but if you object to anything or find the layout confusing I'm happy to resolve it with you :)

Hamish
Last edited by hamishmb on 20/07/2017, 21:18, edited 2 times in total.
Hamish
TerryJC
Posts: 2616
Joined: 16/05/2017, 17:17

Re: Unified Software for Probe Testing and, eventually, deployment

Post by TerryJC »

hamishmb wrote:I have recently created a github repository for us at https://github.com/wmtprojectsteam/rivercontrolsystem, and will add people as soon as possible after they have made accounts. This allows better version tracking of our source code, and easier collaboration.
I've created an account at Github and will read the Guide later this afternoon. My username is TerryJColes.
Terry
TerryJC
Posts: 2616
Joined: 16/05/2017, 17:17

Re: Unified Software for Probe Testing and, eventually, deployment

Post by TerryJC »

I've had a few responses to my query this morning on using the Issues tool in Githubas a bugtracker. I told the List members the size of our team and the fact that (ultimately) staff and other volunteers may want to raise issues. I also mentioned our Forum. No-one is against the tool (so far) and everyone thinks it would suit what we are doing. Here are a couple of the more detailed responses.

From Natalie (an Android developer);
At Google, I worked on a few open source projects hosted on Github, in particular this one https://github.com/googlesamples/android-architecture. We used Github issues as a bug tracker and it was working quite well. As you can see, we were using labels to sort the issues, we found this
necessary but we had a lot of issues coming in (as you can see, it's quite a popular project), you may not need to do this for yours. But yes, I would
recommend using it - you can comment on an issue, close an issue etc It does the job of bug tracking (not as powerful/flexible as paid for bug tracking software but much simpler to use so probably much better for your project).
From Ralph, (who is a Linux developer and the List Guru):
You four could have accounts so you can comment, and do whatever else you configure it to allow. And reports of possible problems from users
could filter through you lot in the existing ways, email and forum, to save them the bother of registering and learning yet another thing. They can always open the issue's link posted on the forum to read discussion and progress, and register if motivated.
I think that, now we have the Repository up and running, it would seem to be the right way to go to use the Issues tool to track bugs. If you click on Natalie's link to the Google project you can see how they were able to categorise the Issues by the use of labels. It's all very clear and straightforward.

If anything else useful comes in over the next few days, I'll pass it on.
Terry
hamishmb
Posts: 1891
Joined: 16/05/2017, 16:41

Re: Unified Software for Probe Testing and, eventually, deployment

Post by hamishmb »

Sounds good :)

What list is this? The people sound super helpful, and it may be good for me to connect with some of them.

I'll try and get that set up as well then :) I've not yet added you to github but I'll find some time to do it later :)
Hamish
TerryJC
Posts: 2616
Joined: 16/05/2017, 17:17

Re: Unified Software for Probe Testing and, eventually, deployment

Post by TerryJC »

Hamish,

This is the Dorset Linux User Group http://dorset.lug.org.uk/wiki/, otherwise known as DLUG. It's a bit like the Raspberry Jams that have sprung up all over the country in the last 4 years, but it's aimed at professional developers, IT specialists and Linux enthusiasts, rather than the more Edu-oriented Jams. It also only covers software and system admin topics; queries about hardware are usually confined to routers, etc.

There are around 70 people registered on the list at any one time, including yours truly and Clive. For the most part it's a technical resource that can be a bit more responsive and focused than the online forums, but it does also have a social side to it, with a monthly Pub meeting at the Broadway in Bournemouth.

Have a browse through the info at the link above. If you'd like to go to the Pub Meet, the next one is on the 1st August; I might even be able to organise a lift for you, depending where you live.

End of shameless DLUG plug. :lol:
Terry
hamishmb
Posts: 1891
Joined: 16/05/2017, 16:41

Re: Unified Software for Probe Testing and, eventually, deployment

Post by hamishmb »

Thanks. I will definitely be joining it :)

I would come, but I'll be on holiday, so I'll try for the next one instead. I may be able to drive there so lifts shouldn't be a problem, but thanks for the offer :)

Hamish

Hamish
Hamish
TerryJC
Posts: 2616
Joined: 16/05/2017, 17:17

Re: Unified Software for Probe Testing and, eventually, deployment

Post by TerryJC »

One thing that I should have mentioned :) The list can be quiet for days and then someone will post a query and it will be very active for days. When I first joined the list, Linux on the desktop was quite a new thing and we used to have a lot of traffic from people interested in getting their PC up and running. As you know, Linux is probably easier to install than Windows these days, so traffic is a bit more specialised.
Terry
hamishmb
Posts: 1891
Joined: 16/05/2017, 16:41

Re: Unified Software for Probe Testing and, eventually, deployment

Post by hamishmb »

Okay, you have been invited Terry. The bug tracking called "Issues" seems to be working too.

Sounds good, I'll have a look over the weekend :)
Hamish
TerryJC
Posts: 2616
Joined: 16/05/2017, 17:17

Re: Unified Software for Probe Testing and, eventually, deployment

Post by TerryJC »

I'm in.
Terry
hamishmb
Posts: 1891
Joined: 16/05/2017, 16:41

Re: Unified Software for Probe Testing and, eventually, deployment

Post by hamishmb »

Okay, you should have pull and commit access to our repo then :)

I have now written the basic test for the solid state relay probe, testing now...
Also, the capacitive probe software has been ported and tested.
Also (:P), the new code has been made to run on Python 3 only because there are some features I want to use that aren't on py2. The RPi.GPIO library has a py3 version too, so it seems fine :)

Hamish
Hamish
Post Reply