Revise and improve publishing logic #1492
Conversation
| end | ||
|
|
||
| def publish?(thing) | ||
| can_be_published?(thing) && !hidden_in_the_future?(thing) |
There was a problem hiding this comment.
Awful naming for both, I know. Couldn't come up with any idea though. Can you?
There was a problem hiding this comment.
There's nothing wrong with the naming here. 😃
|
Seems ok, although I don't understand the difference between |
|
@mattr- |
|
i.e. for future dated posts? |
|
@mattr- Correct! :) |
|
Wow, three flags for this? Seems like overkill. Is there a way we can On Tue, Sep 3, 2013 at 1:03 PM, maul-esel notifications@github.com wrote:
|
|
Well, three flags for three similar yet different features. Can't really drop the features or the two existing flags because of incompatibility, so the only possible (though probably not recommendable) thing would be some option So it seems to me there's no good way to avoid having three flags. |
|
Something to keep in mind for Jekyll 2.0. 😃 On Tue, Sep 3, 2013 at 1:23 PM, maul-esel notifications@github.com wrote:
|
|
Should be ready to be 🚢ed. |
|
Re-reading the code, I think there's still a bunch of entanglement. What if we just passed metadata or something to a class method in a |
| Then the _site directory should exist | ||
| And the "_site/index.html" file should exist | ||
| And the "_site/public.html" file should exist | ||
| And the "_site/secret.html" file should exist |
There was a problem hiding this comment.
The published flag is also valid for posts. Could you please add a similar example with posts instead of pages?
|
Looks like there is a merge conflict here with current master, too. |
|
Ok, I think you might have missed what I meant. My mistake though, that link in my previous comment is basically invisible. So the re-done stuff is here: maul-esel/jekyll@master...publisher. |
|
This needs a rebase. Would you mind, @maul-esel? |
|
@parkr: No problem, done! |
Fixes 1034.
Also make step definitions a little more generic.
|
Oops, sorry, my master branch was out of date. Now it should be fine. |
This PR refactors the publishing logic by moving it out to a
Publisherclass. It also adds the possibility to mark a page aspublished: false, as already possible for posts (thus this supplements and closes #1080), and a new--unpublishedflag to render such posts and pages.