Pathauto shouldn't create paths for unpublished nodes

why not?

Ok.

I have a couple thoughts:

1. I don't care about this problem enough to write the code. So, if someone else makes a patch I'll consider committing it.
2. If you have a small url_alias table, a few more entries won't impact performance becuase it will be fast. If you have a large table, then a few more entries for the handful of unpublished nodes won't matter much in the larger scheme of things.
3. the someduplicate_0 probem seems somewhat valid to me

If someone else makes a patch for this, please provide a benchmark that proves the need for this extra code.

And the software works as designed, so it's a feature, not a bug.

I can see how this would be troubling in your situation, but you have what's called a "corner case". A situation that is not common, so it's not a priority for me to work on it.

If you make the patch (or hire someone to or...) then I'll consider applying it if the benchmark shows a performance improvement in a more common situation (e.g. a ratio of 10 published nodes per 1 unpublished nodes).

The problem is real and I do not think this is limited to bulk mode.

Whenever you have path include the date and the scheduler module enabled (very common for professional blogging setup) paths are generated with "created" instead of "published" date. This is very confusing both to people and to search engines.

I might even get around to fixing this for 4.7 soon enough.

Useful feature.

I think we will see more and more use of Drupal for (new) projects that involves collaborating on more content in draft mode than what is published at any given time.

Example: systematic revisions of existing published content:
(quality control, keeping things up to date - coming features in other modules will support this)

A group of people may collaborate on several parallel suggestions for new versions of existing nodes before reaching a conclusion for which to publish in place of the old one. In such situations it will be completely normal to have more unpublished nodes than published ones.

So I think this feature will become increasingly important.

Folks - I appreciate your thoughts, but we don't need more advocacy or discussion about which version this should go into.

We need (reviewed) code and (reproducible) benchmarks.

ultraboy - consider that we simultaneously have issues to speed up bulkupdates and this issue which will almost certainly make bulkupdates slower (as a result of more php code).

I'm not about to commit something based upon a claim of "it's obvious".

Every optimization and feature has a tradeoff somewhere else. That's why we use benchmarks.

I think the fix is very simple. Skip non-published nodes whether the update is bulk or not. I cannot think of a reason not to do this and may be able to put some time into making a patch.

csevb10 is working on the ideas of this issue in the issue http://drupal.org/node/110223.

So, I'm marking this as a duplicate of that.

Hi Greggles,

I have a situation in which this can be handy as well, let me explain.

On a site I'm working on I have a page called promotions. On this page I promote a montly promotion.
What I would like to do is to set up all pages for the whole year and schedule them to be published/unpublished each month.
Yet I want each page, when published, to be accessable via domain.com/promotions but I can't set them up as an error message says: path is already in use (or something similar).

Hope you understand what I'm trying to say :) And maybe you know of a workaround or now are willing to address this issue :)

Cheers

PS. working on D5

I review a few companies on my site. There are a few comments which I have to unpublish since they seem defamatory, but keep them in unpublished list bcoz I have the email ids.. and also the comments are useful if the company requests for feedback.. yeah may be another corner case.. but by default page title should not have displayed these comments..