index page improvements

[raisjn]

We use the script in https://github.com/toltec-dev/toltec/blob/testing/scripts/repo-build-web to build the package web listing. This can potentially use some upgrades, like adding:

• timestamp (but in easier to read format than the ISO 8096 format in the package recipe)
• link to .ipk file
• filesize of .ipk file
• ???

To run the script, go to root of the repository and execute bash scripts/repo-build-web package tmp - it will then generate the package listing into tmp/ directory

[matteodelabre]

A link to the package file + its size would be great for example if you need to install a package from testing while being on the stable branch, or to inspect a specific package.

Do you have an use case in mind for adding the release timestamp to the list?

[Eeems]

We should also provide a file hash of the ipk for the paranoid who want to confirm that there is no MITM on the network they are using.

[raisjn]

Do you have an use case in mind for adding the release timestamp to the list?

People tend to look at recency of software as some sort of signal. Personally, I could see someone saying: "oh, I installed this package a week ago, the latest update is today - that means I can update it now".

This reminds me that it might be useful in the future for us to have an RSS feed or page feed showing a list of package updates. The feed would also double as a way for people to find new software, like freshmeat used to.

[matteodelabre]

Do you have an use case in mind for adding the release timestamp to the list?

People tend to look at recency of software as some sort of signal. Personally, I could see someone saying: "oh, I installed this package a week ago, the latest update is today - that means I can update it now".

Sounds good.

This reminds me that it might be useful in the future for us to have an RSS feed or page feed showing a list of package updates. The feed would also double as a way for people to find new software, like freshmeat used to.

That's interesting, we'll have to think about how we can automate this.