Importing disregards title

ChrisWiegman commented 11 years ago

The importing of the OPML file seems to disregard the custom title field. Is this intended behavior?

joewiz commented 11 years ago

+1 to preserving the original title. I had to manually rename my google reader imports - seems unnecessary.

benubois commented 11 years ago

Feedbin will now use the title attribute in OPML for a custom feed title.

joewiz commented 11 years ago

Great - thanks, Ben!

Zegnat commented 11 years ago

Feedbin will now use the title attribute in OPML for a custom feed title.

Isn’t the against the spec?

The text attribute:

[…] should initially be the top-level title element in the feed being pointed to, however since it is user-editable, processors should not depend on it always containing the title of the feed.
The title attribute:

Is always ‘derived from information in the feed itself’ and ‘contains the top-level title element from the feed.’

This seems to suggest that title always matches the feed’s title element, while text can be a user-defined name for the feed.

I have no idea how the current RSS readers treat these attributes though, different export files from different readers would need to be compared.

benubois commented 11 years ago

@Zegnat that's a good point. I never tried editing subscription names in Google Reader so I don't know what attribute changes. The request here is for the title field though.

@ChrisWiegman and @joewiz which attribute has the custom name? text or title?

Zegnat commented 11 years ago

I don’t have access to a lot of different services, so here is what I can test.

On export:

Google Reader: both text and title are changed to the custom value.
NewsBlur: both text and title are changed to the custom value.
FeedWrangler: does not allow for renaming feeds.
Feedly: both text and title are changed to the custom value.

Everyone included type, text, xmlUrl, htmlUrl, and title. NewsBlur also includes version.

On import 1:

<outline type="rss" text="Pinboard Blog" title="Pinning" xmlUrl="https://blog.pinboard.in/feed/" htmlUrl="https://pinboard.in/"/>

NewsBlur: uses ‘Pinning’ as name.
Feedly: uses ‘Pinning’ as name.

On import 2:

<outline type="rss" text="Punning" title="Pinboard Blog" xmlUrl="https://blog.pinboard.in/feed/" htmlUrl="https://pinboard.in/"/>

NewsBlur: uses ‘Pinboard Blog’ as name.
Feedly: uses ‘Pinboard Blog’ as name.

On import 3:

<outline type="rss" text="Punning" xmlUrl="https://blog.pinboard.in/feed/" htmlUrl="https://pinboard.in/"/>

NewsBlur: uses ‘Punning’ as name.
Feedly: uses ‘Punning’ as name.

Conclussion:

NewsBlur and Feedly seem to break from the specification on import by preffering title over text.
Everyone seems to break from the specification on export by changing title to match the user supplied name.
Feedbin could follow the specification on import and prefer text over title without breaking anything, most OPML files will have equal values in those fields.
Feedbin should break from the specification on export and set both text and title to comply with other services’ import systems.

feedbin / support