Alternative new page headers

The scenario: I don’t want headers on my chapter pages. I want it clean at the top. However, I have a section of “song lyrics” in the middle of a chapter that I want to display on its own page, so I’ve added new-pages to the beginning and ending of the compile/separators for “song”… but it doesn’t show the headers on the pages. I want author/book/page info on these pages, so they don’t just get copied/lost without relevant info… also, as the book picks up after with a section, that section page has no headers…

How do I get headers on the non-chapter (new-pages) but exclude them on the chapters?


(Chapter displays correctly)

Screenshot 2022-12-03 at 19.52.16
Lyric page not showing header on the left

Screenshot 2022-12-03 at 19.54.22
Following section text also doesn’t show heading on left.

How I’d like the lyric/section page to look:
Screenshot 2022-12-03 at 19.45.10
The way I’d like the song lyrics to display… but the left side is blank if the Chapter heading is too

Screenshot 2022-12-03 at 19.56.44
How I’d like the following section to look with header

I think, given what you’re going for, this isn’t the sort of thing Scrivener does very well. You’ll have to patch that level of detailing up in a layout tool, post-compile.

The options for this are all in Page Settings, maybe you can find a combination that will work for you, but I think it will be tough since chapter and song cuts are both triggering the same underlying behaviours.

Any way to submit a feature-request? I don’t use a layout tool.

As a paperback fiction novel, I’d like to think my books are probably nicer (typographically) than 80% of what’s out there (certainly self-published), but I need to keep compile->KDP simple because I have an epub, paperback, hardcover, reader-copy version for each of (7) books at present… so every time I need to do an update, I need to do it 4x in different compiles and remember all the “post-compile” stuff. I’ve largely got the epub figured out through CSS, but I still need to open it in sigil to embed fonts and to get it to start on the right page…

I was always about scripting/automating stuff like this when I had to publish websites/updates for 1800+ page technical schematics every week… we kept the data in xml, and then I wrote scripts that ran XSLT web-page-output and XSL-FO->PDF compilers to make it all look pretty… but I was publishing stuff at the bare-bones programmatic level, not as an “author” or high-level user per se. I’m a perfectionist who doesn’t like “ad hoc” in my processes, because it takes away from other things… like writing. :-/

You’re speaking with the right person for feature requests anyway, so there’s no need to go and make a separate report anywhere.

The main problem with adding this as a singular feature is what I implied above: there is nothing about your section layout’s functioning that has a strong difference from a chapter layout, other than extremely specific stuff, so specific it would probably make the checkbox useless to anyone but you.

So as you might conclude, the main issue with implementing such a thing in Scrivener elegantly is the separation between Layouts and page flow. They have nothing to do with each other, other than the former indirectly influencing page flow triggers: the checkboxes you see in Page Settings. It’s a system that works okay for a proverbial 90% of use cases, but we’re not going to get much higher than that without a large increase in architectural complexity. I.e. it’s not just adding a checkbox.

I’ve largely got the epub figured out through CSS, but I still need to open it in sigil to embed fonts and to get it to start on the right page…

You could easily automate the former. Take a look at the compile result you get from the Save source files in a folder with exported ePub file. This checkbox isn’t just a troubleshooting and webdev tool, it’s intended to be a post-processing workflow tool—hence the script we provide in that output folder to rebuild the source into .epub. A script could migrate font files, adjust the .opf accordingly (you’d only need to insert some lines into the Manifest element I think), and then trigger the script to rebuild. With a filesystem automation tool like Hazel this could even be automatically triggered so that it happens moments after compile finishes.

As for the latter, try the “<$ebook_start>” placeholder, although that is also just a line in the aforementioned source.

That said, the above is just the start of how Scrivener can tap into a more sophisticated workflow than the basic rich text formatting stuff. It sounds like you’ve already gone well down the path you’re on at this point—perhaps too far to consider switching gears for past projects. But take a look at the Scrivener user manual PDF. That comes “straight out of Scrivener” in the human-centric sense of me clicking the Compile button and getting a PDF. And it is a lot more aligned with what it sounds like you’ve invested time into developing.

Of course, if you look at the layout being done you’ll quickly see that there are many things it is doing that Scrivener flat out cannot do, no matter how hard you torture its features. So the technical reality is that “straight out of Scrivener” means more along the lines of what you described with. When I click compile, Scrivener is responsible for generating formatting instructions, in the form of augmented Markdown syntax, and it then invokes a Ruby script to manipulate that further, build a .tex file, clean that up, and can then go on further to invoke LaTeX and result in a PDF. Thus when I compile I’m depending upon gigabytes of design libraries and tools, outside of Scrivener—but again for me it’s just clicking a button. It took me a while to set that all up, but now a revision update is largely just waiting for the computer to finish working.

So maybe for the next project, take a look at installing Pandoc on your system. This will add three new file types to the compile menu: DOCX, DocBook and ePub.

Those latter two might be of interest to you in your existing systems. But even beyond that you can get into the type of stuff I speak of above, where you have a script in your compile settings, or being triggered by them, to execute all of the steps you need to build ePub, PDF and whatever else, using the best systems for those jobs—even conceivably in one shot.

At that point, setting up the compiler in Scrivener is decidedly different, since you’re building formatting instructions rather than being stuck inside of a design box that is imposed by the natural limitations of what one programmer can do toward that goal (and what one program can provide a front-end for). The advantages should be clear though, to anyone with the aptitude, and you will find that a lot of thought and design has been put into making the compiler really good at building instructions. You could design a compile format that builds XML to the same specifications you were using before.

If you want to see an applied example of that, take a look at the “General Non-Fiction (LaTeX)” project template’s default compile settings. With that template we have basically trained Scrivener to export .tex directly. This principle could be used to export XML, Markdown, HTML, ReStructuredText and probably just about any good system of instructions for building output from them.

Anyway—bit of a tangent. As you can see I’m a big advocate for using the best tool for the job, and that goes beyond matters of typesetting quality. Sure that is something important to me, but the bigger thing to me is being able to do what I want with a design, rather than running into feature walls constantly. I want to be able to build an index, or link directly to a figure, or design a chapter page freely, or a sidebar/callout box without ugly hacks… I don’t want to have to settle for what a writing program lets me do—and that means using that writing program to get me into the next tool as efficiently as possible (which as demonstrated, can become so efficient we might as well even think of Scrivener being the thing doing it).

Like I say, seven books deep into how you’re using Scrivener—it’s maybe not feasible to think of using it different for them. But maybe book eight can be a single click to produce all four variants to precisely the level of quality and design range you are willing to exert into the production of them.

I’m a perfectionist who doesn’t like “ad hoc” in my processes, because it takes away from other things… like writing. :-/

And once these workflows are developed, you can get back to that, which is what I do.

2 Likes