RSS Results Structure Changes Depending on Normalization

[JulianWebb]

Hello!

I am glad to have found your module, it looks like it will make handling feeds easy.

---

The structure of the results from fetching an RSS feed depends on if the normalization option is set.

If it is false then the object containing the feed items is called item and if it is true then it is called entries.

I do not know if this is intended behaviour, however the documentation doesn't mention it either way.

[ndaidong]

@JulianWebb thank you. When you don't use normalization, this lib will keep the structure as it is. That's is expected behaviour.

In real world, it's more complicate than you said, here is what I known:

• RSS uses tag name item, ATOM uses tag name entry
• item and entry can be an array or single object (when there is only one item)
• JSONFeed uses items. Despite the fact that it's plural, it can be single object too!

So I recommend to enable normalization option if you need something more consistent.

[JulianWebb]

If I enable normalization though, I lose data from the item's fields because it gets stripped.

I understand that the names are different, I just think it makes sense to expand the documentation around the normalization option to inform people of this.

[ndaidong]

@JulianWebb I agree that it is better to described more clear. The documentation is quite sketchy, not fully telling the capabilities of this library 🔔

Note that 2 methods getExtraFeedFields and getExtraEntryFields can help you get any data from feeds when you need, similar to #66

[JulianWebb]

Do those methods stop the normalization from stripping HTML and CDATA?

[ndaidong]

@JulianWebb no, they don't stop normalizing, but allow to choose which content you want to pick from the original data and how you customize these pieces.