Task #603

Task #600: Move node content from devcloud to pantheon

Create and migrate devcloud feeds

Added by Lippe Lippe about 1 year ago. Updated about 1 year ago.

Status:Closed Start date:04/21/2012
Priority:Normal Due date:
Assignee:Lippe Lippe % Done:

100%
Category:- Spent time: 1.00 hour
Target version:0.1.0

Description

  • Create exportable feature of all devcloud feeds, to be reviewed for quality and activity on pantheon, with the benefit of a working cron job.
  • Test local
  • Push to dev.

History

#1
Updated by Lippe Lippe about 1 year ago

  • Target version set to 0.1.0

#2
Updated by Lippe Lippe about 1 year ago

  • Status changed from New to In Progress
  • Assignee set to Lippe Lippe

As noted in #602, I've had various issues with both node_export & uuid_features.

1st try at featurizing Feeds with uuid_features created feed nodes with no source URL, blank taxonomy (feed node tag) fields, but also included the required tag field (channel matching tags) as is used on nodes. So basically, totally f'ed.

Giving a few more tries with node_export, but no glory so far.

I've already given a quick shot at using feeds to import from a CSV, but had issues with the fact that these would need the feeds feed node importer, which uses the FeedsFeedNodeProcessor.inc, and has a limited set of targets for mapping incoming data. I'm thinking that either a custom mapper or variant of FeedsFeedNodeProcessor would be needed to add the feed node tags and source data - which are fairly crucial.

For the time being, I'm going to see if I can create an OPML, because that functionality is at least built-in.
BUT, that would mean channel node tags would have to be hand-applied on every fracking import, and seriously limit workflow.

#3
Updated by Lippe Lippe about 1 year ago

OPML workflow:

Using this online OPML builder it's possible to take the URL column from the CSVs already in the repo (#619) and generate an OPML.

Downsides:

FeedsOPMLParser has limited mapping:
  • title -> Feed title
  • xmlUrl - > Feed URL

The OPMLs generated by this tool use the feed URL for the title, so feeds would come in without human readable titles, and without feed node tags.
Managing feeds via a GUI in OPML is cumbersome, requiring specific software like serna-free or oxygen.
Saving this as a last ditch way to have feeds in a repo, with a proven import method.

#4
Updated by Lippe Lippe about 1 year ago

  • Status changed from In Progress to Closed
  • % Done changed from 0 to 100

Feed importer for Feed Nodes was created (#709).
..so there's a (crummy) workflow for importing these feeds.

Also available in: Atom PDF