(I'm on version 3.4.4 on Windows 10, non-Portable)
I'm playing with the weblog settings, and observing some weird inconsistencies and seemingly contradictory behaviour with document header behaviour. I'm posting to a Wordpress blog that I built today for the sole purpose of testing this out before integrating an existing website.
First off, this setting does the exact opposite of what it claims to do:
The help text says it does one thing, but the actual metadata command does the opposite:
When posted, with the "Strip Document Header" box checked, the header is not stripped, but if unchecked, the header is stripped. Looks like the second setting was incorrectly patterned after the previous one when naming, but not in functioning?
The second thing I'm noticing is that the "Strip Header" setting ignores H1 style headings created with the alternate method of equal-sign underlining. If I switch to using the Hash-notation method, it works (albeit still in reverse, as detailed above), but when using a series of equals symbols as the top header, it breaks either way.
Maybe this second behaviour is intentional, not sure. I only note it because the editor that I use when working with my notes on my Mac and Linux machine - QOwnNotes (which I also use on Windows but am trying to switch to MM instead) - automatically builds first-line headers in that fashion for new documents, so I've just gotten used to that flow.
(Update: I previously stated that the "don't infer featured image" setting was acting the same, but that was an incorrect observation, so I've revised the post)
(I'm on version 3.4.4 on Windows 10, non-Portable)
I'm playing with the weblog settings, and observing some weird inconsistencies and seemingly contradictory behaviour with document header behaviour. I'm posting to a Wordpress blog that I built today for the sole purpose of testing this out before integrating an existing website.
First off, this setting does the exact opposite of what it claims to do:
The help text says it does one thing, but the actual metadata command does the opposite:
When posted, with the "Strip Document Header" box checked, the header is not stripped, but if unchecked, the header is stripped. Looks like the second setting was incorrectly patterned after the previous one when naming, but not in functioning?
The second thing I'm noticing is that the "Strip Header" setting ignores H1 style headings created with the alternate method of equal-sign underlining. If I switch to using the Hash-notation method, it works (albeit still in reverse, as detailed above), but when using a series of equals symbols as the top header, it breaks either way.
Maybe this second behaviour is intentional, not sure. I only note it because the editor that I use when working with my notes on my Mac and Linux machine - QOwnNotes (which I also use on Windows but am trying to switch to MM instead) - automatically builds first-line headers in that fashion for new documents, so I've just gotten used to that flow.
(Update: I previously stated that the "don't infer featured image" setting was acting the same, but that was an incorrect observation, so I've revised the post)