Tvheadend

Flole Systems wrote:

Right now basically all build are broken for various reasons. I won't merge anything until someone has time to do adapt Github actions to test all PRs on all targets so such issues can be noticed before merging. I am tired of code that makes it in and then randomly breaks some targets....

Sorry, I'm no expert at GitHub. I know that when I submit a TVH PR, GitHub automatically compiles it and tells me if/where I went wrong. I see this happen when I submit a PR for Kodi as well.

When GitHub compiles a new PR for Kodi, it does so for a variety of target architectures automatically. I have run afoul of this in the past whereby a Linux change inadvertently broke something in the Windows build.

Is this what you are referring to? Having GitHub automatically build for multiple target architectures and report failures.

• What target architecture is TVH currently built for ? (Looks like just Ubuntu)
• What target architectures are missing from the automated process?

I did some reading on GetHub workflows:

In the TVH Repository, I see that there are a number of 'workflows' under the 'actions' tab. One of these workflows is 'Build Tvheadend CI' and its actions are defined in 'test-compile.yml'.

In 'test-compile.yml' there seems to be a number of instructions for building TVH for 'ubuntu-latest'.

GitHub seems to have a number of hosted runners that are available:

https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#choosing-github-hosted-runners

• windows-latest or windows-2022
• windows-2019
• ubuntu-latest or ubuntu-22.04
• ubuntu-20.04
• macos-13 or macos-13-xl
• macos-latest, macos-12, macos-latest-xl or macos-12-xl
• macos-11

I suspect that this may not cover all of TVH's preferred build targets.

• Which TVH target architectures are missing from the list?
• Which of the missing target architectures would the most helpful to try first?

I volunteer to do some research into adding the most useful target architecture and we can see how that turns out. I will try to make a copy in my private repository and see if I can get it running there along side the existing 'ubuntu-latest' target. It may take a while becasue I will have to teach myself how to do it first.

We are actually compiling various targets, all that's needed now is to add a "if no cloudsmith key is defined/if it's empty" condition to the upload script and add an entry specifying "env: Cloudsmith" (see the coverity file for an example) and allow it to run on every PR. It should then run. If it does we can merge that and then I can remove the cloudsmith api key and then it should only upload master-builds but build everything on all PRs.

I believe Oliver is already on this though.

Flole Systems wrote:

I believe Oliver is already on this though.

Flole Systems - OK, so Oliver is already working on a build solution. I will stand down from my previous offer regarding build targets.

Will Oliver's solution apply to all ongoing future builds? Or is this a one-off audit to see if the current portfolio of PRs cause issues?

Assuming that Oliver's efforts resolve the build issue, what's the next biggest roadblock that we can help you with?

It will affect all future PRs.

What you can do is look again at the changes from your previous PR. I left a comment somewhere where a declaration in a "for"-statement was introduced which now causes builds to fail on some platforms. Then at least there's a working master again once that and the time_t issue is solved.

I still have no clue how the documentation stuff is supposed to work (I believe I edited an automatically generated file a while ago cause I had no clue how to do it properly).

Flole Systems wrote:

What you can do is look again at the changes from your previous PR. I left a comment somewhere where a declaration in a "for"-statement was introduced which now causes builds to fail on some platforms. Then at least there's a working master again once that and the time_t issue is solved.

I still have no clue how the documentation stuff is supposed to work (I believe I edited an automatically generated file a while ago cause I had no clue how to do it properly).

I will look into my previous PRs for the issue that you mention.

Can you please provide some clarity regarding the documentation issue? Are you referring to the documentation that gets built into TVH that you can access via the WebUI?

I don't think I have access to the translations, also the documentation needs to be "compiled".

Can you also please provide clarification on your previous statement regarding the translations?

I found this:
https://explore.transifex.com/tvheadend-foundation-cic/tvheadend/

Are you referring to the translation process or how those translations get incorporated into TVH?

There is https://docs.tvheadend.org/ but I don't think anybody has access to it anymore, so it doesn't really matter how it's built. We could use GitHub pages though, and automatically build it and upload it to some GitHub-page.

As for Transifex, I don't have access to that so no way for me to actually use it.

Delta Mike Charlie wrote:

I will look into my previous PRs for the issue that you mention.

OK, I found it. You left a comment on a merged PR (#1544).

How would you like to proceed? Would you like to try to un-merge the PR and have me update that, or would you like me to create a new PR to fix that 1 line?

The new PR would probably be easiest for both of us.

New PR, definitely. But let's wait for Oliver to get the tests running, if you fork now and open a PR it won't be tested if the tests get added after that.

Flole Systems wrote:

New PR, definitely. But let's wait for Oliver to get the tests running, if you fork now and open a PR it won't be tested if the tests get added after that.

OK. If you let me know when Oliver is finished. I will raise a new PR with the required fix.

Guys we are now seeing real debate and collaboration instead of moaning and criticism. Hopefully this behaviour gets some traction and gets TVH back on track again.

Those who “can do” have identified themselves so hopefully this is the beginning of new approach to the project.

Remember there is a considerable user base who are members of the forum and I am sure would be only too glad to assist in testing if invited to do so. Do bear in mind that they are not developers or coders but do have something to offer. Their time too.

Hopefully together we can unravel the hidden mysteries of the TVH project together. Thanks again for the input.

Delta Mike Charlie wrote:

When GitHub compiles a new PR for Kodi, it does so for a variety of target architectures automatically. I have run afoul of this in the past whereby a Linux change inadvertently broke something in the Windows build.

I have, coincidentally, just submitted a PR to Kodi. Here are the targets that it automatically builds for:

• android-arm-docker (android-arm-docker)
• android-arm64-docker (android-arm64-docker)
• android-x86-docker
• webos-docker
• LINUX (LINUX-64-GLES)
• FreeBSD
• IOS-ARM64
• WIN-32
• OSX-64
• TVOS
• WIN-UWP-ARM
• OSX-ARM64
• WIN-64
• LINUX (LINUX-64-GL)
• WIN-UWP-64
• WIN-UWP-32
• LINUX (LINUX-aarch64-GLES)
• LINUX (LINUX-arm-GLES)

@ Delta Mike Charlie see this
https://tvheadend.org/issues/6288#note-3

pipelines are now enabled for all MR's.

What would be nice is Mac OS builds and FreeBSD builds since people complain about breaking exotic systems, but we have no proof thereof.

Flole Systems wrote:

There is really no point in trying to maintain 2 branches if there aren't even enough resources to maintain a single one. Also I have absolutely no clue how a new release needs to be done. I assume it somehow involves branching, but that's probably not all that's needed.

I have seen lots of projects dying. Usually the last phase starts with quick-fix-wins which are introducing new and fancy features. With people telling that the others are "slow" and yesterday. Next step are commits that leads to surprising sideeffects - which noone is going to fix: the experience ramp up takes a while.

I’ve been following this item on the GitHub with great interest. https://github.com/tvheadend/tvheadend/pull/1569

This item started off with a report of the TVH forum being down. It then progressed to trying to determine who had access/privileges to the forum in order to get the forum back on line. Then it became a debate on whose message that was sent reached the person that could bring the forum back on line.

From there it appeared that that many of the contributors seemed more interested in gaining access control over the forum rather than the forum itself. I thought with all these contributors commenting on the loss of the forum and what could be done to get it back on line we’d see a renewed interest in participating in it. Well so far I was clearly mistaken.

I’ve mentioned this before. Users new and current drop by the forum for assistance from time to time. There’s no automated query answering system active on the forum to my knowledge.

Is there any chance that the persons mentioned in the link above or indeed any other experienced members here could start contributing by answering some of the queries being posted here if they are not already doing so. This will give us a fighting chance to get some meaningful discussion restarted here again.

Jonas Lang wrote:

From there it appeared that that many of the contributors seemed more interested in gaining access control over the forum rather than the forum itself.

I entered that discussion late and perhaps I’m over simplifying things, but to me, it was mostly about “How did this happen?” and “How do we stop it from happening again?”

In order for these two questions to be answered, we need to understand who has ownership/admin rights over the various components that make the forum work, namely: DNS and hosting. This also leads to asking “How is all of this funded?”

I agree with you that answering user questions on the forum is very important, but if the forum platform goes offline again and we can not get it back, then there will be no forum to visit and no questions to answer. I also think that keeping the Wiki alive is important because it can serve as a source of self-help if it is kept up-to-date.

To prevent the forum from going offline again and keep it running, we need to know the basic facts about how it operates. As far as I know, as at now, the following questions have not been answered publicly:

• Why the forum went down?
• Why the forum come back?
• What do we do if it happens again?
• How do we prevent it from happening again?

Delta Mike Charlie wrote:

Jonas Lang wrote:

From there it appeared that that many of the contributors seemed more interested in gaining access control over the forum rather than the forum itself.

I entered that discussion late and perhaps I’m over simplifying things, but to me, it was mostly about “How did this happen?” and “How do we stop it from happening again?”

In order for these two questions to be answered, we need to understand who has ownership/admin rights over the various components that make the forum work, namely: DNS and hosting. This also leads to asking “How is all of this funded?”

I agree with you that answering user questions on the forum is very important, but if the forum platform goes offline again and we can not get it back, then there will be no forum to visit and no questions to answer. I also think that keeping the Wiki alive is important because it can serve as a source of self-help if it is kept up-to-date.

To prevent the forum from going offline again and keep it running, we need to know the basic facts about how it operates. As far as I know, as at now, the following questions have not been answered publicly:

• Why the forum went down?
• Why the forum come back?
• What do we do if it happens again?
• How do we prevent it from happening again?

Firstly I don’t include you in that list of non contributors. Clearly you’ve been doing what you can for TVH being a late comer to the forum. I’d also include myself as a late comer too but one who is trying to help fellow TVH users where I can.

The consensus is that there is no development happening in TVH so I’m assuming the commits since the last official release could be termed maintenance spearheaded by Flole Systems Systems. You clearly have the required skills to possibly rejuvenate the development side of things but in fairness I’ve seen a number of other possible candidates just melt away after promising the same.

I’ve also done a bit of background investigation into the TVH project and you can clearly see the point at which the final named project developer disengaged from the project. Six years ago would be a fair assessment.

It has been suggested that Andreas Smas at [email protected] may have a solution but as of yet that’s not been confirmed. His last access to the forum was 2019, which is a while back and if he’s chosen not to respond to any correspondence I for one would respect that.

That leaves us in a situation where for the moment we have a usable forum where issues are being posted that do warrant a response. That’s a good starting point for the moment.

We have a GitHub where PRs are being submitted and where warranted are being committed. Unless otherwise advised all the original developers have stepped away from the project so for the moment we make do with what we have and hope that one of the original project members are checking in to the forum to remedy the situation at some stage.