Here are some ideas for the future of node_import. Just my thoughts.

(1) develop it into an API for transforming nodes from 1 type to another. This might be useful for workflows, etc. We use similar field mapping methods to store how the fields should be populated for the new node, then delete the old node.
(2) permit import of XML. The XML could contain literal field mappings or not.
(3) permit automated importing via a cron process
(4) export as CSV, XML

In general, field mappings is the common thread.

Comments

TheWhippinpost’s picture

I know this is an aging thread now but nevertheless, all good suggestions above - I'd also love to see term hierarchy import incorporated too.

Whereas now, there is the ability to import multiple-select terms, given a field formatted:

multi-select
term1|term2|term3|

.. an option to preserve that same format but as an hierarchy.

Anyway, whilst I'm here, just wanna say a big thanks to Robert for taking-up the baton on this project - I consider this to be amongst the, 'should be in core' features, particularly as sites are getting larger these days thanks to bandwidth and hosts providing larger disk-space. Tools like this are going to grow in demand and users will naturally gravitate to those CMS's that provide it IMO.

Don't stop Robert ;)

geodaniel’s picture

I'm really interested in the XML import part of this, and the automatic refresh too, and it looks like others are willing to sponsor it...

geodaniel’s picture

Category: task » feature
csc4’s picture

Subscribing - this would be tremendous if it could be done

chrisroditis’s picture

subscribing!
"Permit automated importing via a cron process" is something I am willing to pay for!

mlncn’s picture

Subscribing.

Also interested in some sort of integration with User Import and import of user node profiles. Working on a hack now...

benjamin, Agaric Design Collective

Matthew Davidson’s picture

While everybody's wishing, it seems to me that the current process of defining which fields are available for node_import to use for each node type is a duplication of the views integration that every module worth it's salt has already done. The fields which are exposed for views to query are generally going to be the same fields node_import wants to be able to populate. Does anybody with views expertise have an opinion on whether it would be feasible to use this to get the list of fields in this way, and eliminate most if not all of the contents of the 'supported' folder?

It seems like the only other major function of the 'supported' files is field validation, and it looks to me like some love applied to _node_import_get_nodes() in node_import.module (see comments from line 624 on) would remove the need for this.

kingandy’s picture

(4) export as CSV, XML

Just for the sake of completeness, I'll note that the Views Bonus Pack adds the Export view type - if you can build a View of it, you can export it as a CSV/XML(/DOC) file (either by using an argument or by setting the entire thing to "be" an export). I'm using this a lot.

akaserer’s picture

Subscribing

zach harkey’s picture

kingandy is right. I've been doing this too. It can be tedious but it works great in many cases.

Robrecht Jacques’s picture

Status: Active » Postponed

Not in 6.x - maybe in 7.x.