Automating exports for the mobile app

[jaymzh]

OK so I'm looking into better automation for feeding the Schedule data into the app.

The easiest thing to do would just be to make the current CSV view non-authenticated like the current JSON view (https://www.socallinuxexpo.org/scale/23x/signs).

However, if we would like to collapse these, using the JSON is a simple change, except there's a bunch of changes that would need to happen:

1. I need a long abstract that isn't HTML escaped (currently it's the short abstract, but with HTML escaping)
2. I'd need a "Track" that's the same as the current "Topic" but again, without HTML escaping.
3. Not sure if "name" is HTML escaped as well, but if so, a "title" that is the name but without escaping.

We'd want to keep the current fields as-is for the existing users, so these would be additional fields.

I don't have a preference on which route we take, either is fine with me.

[sarcasticadmin]

I would suggest the non authenticated csv feed since the exisiting json feed isnt adequate for your needs.

We have https://www.socallinuxexpo.org/scale/23x/signs working just the way we want for signs and Id like to keep it that way 😁 #194 (comment)

[jaymzh]

@sarcasticadmin - to be clear, we'd only be adding fields, not modifying existing fields. And having one of these rather than several may be preferable. But, as I said, I'm fine with either.

[DrupalPhil]

I can create a feed quite easily. Do you want me to clone the signs feed and modify it?

1. Do you want the property to be snake case (lowercase with underscores)?
2. You mean you want me to rename "Topic" to "Track"? On the signs feed, the "Topic" property is the "Track".
3. You mean you want me to rename "Name" to "title" and remove HTML encoding?

Take a look at a draft here:
https://www.socallinuxexpo.org/scale/23x/app

[jaymzh]

The naming of the keys isn't super important I can adapt to whatever, but using the CSV headers is what the code expects today.

This sample looks great except for the HTML in the abstracts. Track and Name look good here.

[DrupalPhil]

I see. So you need CSV, not JSON?

[jaymzh]

no no no. Defintiely want JSON. Was just saying, it's the same code that was reading the CSV, so if the headers match, it's no change, but happy to adjust headers as well.

[jaymzh]

[syncing comments from slack]

Feed currently works great.

Currently you're doing work to strip out the HTML and I'm re-wrapping it in <p></p> - we can probably undo both of those, but it's no biggie just let me know.