Orange Contributions

[sbernard31]

I create a dedicated issue to help cooperation with Orange contributors.

The discussion starts with #1034 (comment) :

(from @Przem83)

For now Orange will focus mainly on verifying the already implemented LWM2M 1.1 functionality and on the Observe-Composite implementation. After the holidays we also intend to start some work on the NIDD transport for Lora and 3GPP - as it seams that no body is working on it currently. For the remaining LWM2M 1.1 topics please let us know were you could use some more manpower and we'll do our best to help. Our main goal is to speed up the Leshan 2.0 release as much as possible - so we'll do our best to not disturb the ongoing works unless our help is welcome.

@MichalWadowskiOrange you could be interested too.

[sbernard31]

For now Orange will focus mainly on verifying the already implemented LWM2M 1.1 functionality and on the Observe-Composite implementation.

👍

After the holidays we also intend to start some work on the NIDD transport for Lora and 3GPP - as it seams that no body is working on it currently.

It could be a big part, I don't have clear idea about how this should be implemented, so help is very welcomed. I create a dedicated issue : #1046.

Our main goal is to speed up the Leshan 2.0 release as much as possible

With the recent 2.0.0-M4 release, I opened a lot of issues and updated the roadmap.
You should look what feature is missing you the most , because the scope of the 2.0.0 is not clearly defined, so this could help to define it.
All issues tagged with milestone-2.0.0 will not necessary be part of the release. If some of them are too expensive and/or not needed, they could be postpone to a 3.0.0.

Could I ask why you seems hurry about having a 2.0.0 release ?

For the remaining LWM2M 1.1 topics please let us know were you could use some more manpower and we'll do our best to help.

• so we'll do our best to not disturb the ongoing works unless our help is welcome.

Help is welcome 🙂

Testing and giving feedback help a lot.
Code contribution is very welcome too, you could maybe look at the opened issue and see if there is something which seems accessible to you. Just some ideas but :

• Maybe not so fun but there is some JSON parser/serializer to rewrite (see Keep only one JSON dependency #1045), it could be maybe an easy way to contribute.
• If you like frontend UI dev (vuejs/vuetify), there is a lot of missing feature in the demos. (I didn't created I issue for all of them so you should create one before to start)
• there is also some housekeeping issue like Enhance CoRE link format parsing. #1022.

[sbernard31]

@Przem83, @Michal-Wadowski just a ping in case you missed this (as I didn't get any feedback)
if there is nothing to add it's perfectly OK too 🙂

[Michal-Wadowski]

Could I ask why you seems hurry about having a 2.0.0 release ?

It's because we have internal project which requires Leshan with some functionalities (e.g. Observe-Compose, NIDD). And we have to use stable release, so it's the reason why we waiting for 2.0.0 😄

Maybe I will try to improve CoRE link for now - issue #1022

[sbernard31]

And we have to use stable release, so it's the reason why we waiting for 2.0.0

You are not allowed to use milestones release ?

[Przem83]

Actually we are using milestones - at least temporarily - so it's not like it's forbidden or something. It's just that we are getting some pressure to move to a full release ASAP to avoid rewriting code (the one that uses Leshan) in case of any major changes between milestone releases. You know, we're are not accustomed to use opensource on production very often - at least not without a full fledged support from a third party :). But anyway here is some more information on our current activities related with Leshan:

• We started the work on NIDD, though for now we're more like focusing on studying the standard documentation then writing an actual code, just to make sure that we know what we're doing :).
• We are working on a usable version of a Leshan server that would use a shared session store based on Redis instead of the in-memory one. So if you would be interested in making this part of the Leshan demo server in release 2.0.0 we can provide you with the code when we're finished (currently we're still sorting out some serialization issues and doing some tests). I know that you have already done some test in this regard in the past so it might also be interesting to compare our results.

[sbernard31]

It's just that we are getting some pressure to move to a full release ASAP to avoid rewriting code (the one that uses Leshan) in case of any major changes between milestone releases.

I understand but I guess once we will have a leshan version for LWM2M 1.1 you will integrate it.
But you will also need new feature of LWM2M 1.2 as soon as possible and so you will need to use milestone again.

Using milestones as drawback but this as also benefits to help to get a better final version :)

We started the work on NIDD, though for now we're more like focusing on studying the standard documentation then writing an actual code, just to make sure that we know what we're doing :)

Yep this is a very big part. To not hesitate to let me know what you learn about that at #1046 because this is a topic that I didn't investigate at all.

We are working on a usable version of a Leshan server that would use a shared session store based on Redis instead of the in-memory one.

Yes maybe this is something we could even share in leshan-server-redis ? if you need to talk about this please use : #1096