Questions about liveupdate

Questions
#1

Little time is devoted to the amazon type in the liveupdate documentation. There are several questions:

  • In what form is the data loaded, will it be the same zip archive that will be automatically loaded into the basket or will each excluded resource be loaded separately? If this is a zip archive, then the following questions are not relevant.
  • Each downloaded resource in this case needs to be added via liveupdate.store_resource?
  • Is a unique folder created in the recycle bin for each build?
#2

For the AWS publisher, it uploads each separate file.

This is also an old (deprecated) workflow, which we don’t recommend anymore.
They might even disappear this year.

We recommend using the new workflow of downloading the entire .zip file, and mounting it using liveupdate.add_mount()

2 Likes
#3

Thanks for the clarification, Mathias. Regarding the new workflow with liveupdate.add_mount(), could you provide a bit more detail on how to structure the zip file for mounting? Are there specific guidelines or best practices we should follow for compatibility?

#4

Not sure what you mean here.
Currently, there is only one zip file, and you can mount it as-is. It has a manifest that specifies the excluded files in the zip.
You can also add other custom files to the zip.

Are there specific guidelines or best practices we should follow for compatibility?

The only supported way is to use the same runtime and .zip file that was generated at the time of bundling.

It is however possible to use older/newer .zip files.
Warning: This requires the developer to do thorough testing to make sure the runtime formats hasn’t changed. Given that complexity, it’s not something we want to officially support ourselves.

#5

I can add to this that it is possible for advanced users to split the zip archive into multiple archives but this currently requires a bit of scripting on your part:

We will in the future provide functionality so that developers can configure which files and folders go into which archive and then let the tools automatically split things for you. This will likely be implemented in the coming six months or so.