Re: Page and Column breaks #PDF
Using either FOP or Antenna House Formatter it is technically possible to implement a multi-pass process using the "area tree" XML that both processors can emit (they are similar but not the same but the programming principles are the same for both).
Using this kind of approach it would be possible to provide a much higher degree of layout awareness to the page composition, for example, detecting the case where a title and paragraph followed by a figure occurs at the bottom of a page. Likewise, it is possible to dynamically size figures and tables based on the layout details of the page.
The main challenge is that the logic to apply will be specific to a given page layout and design coupled with specific content, so it's difficult to have a generic processor, although one could imagine having some set of layout hints that could be embedded in content or applied via style that would then guide a multi-pass pagination process.
For typical full-page layouts the keep facilities in XSL-FO and CSS pagination are sufficient, but when you get into more-constrained layouts, like typical consumer product manuals, it's much harder for a simple keep-with-next or keep-together to produce the correct result.
I've done enough work to date using XSLT to post-process AHF's area tree to know that it would be possible, if even not that difficult, but of course it's not something I personally can do nights and weekends (especially since I seem to have fewer of those available for programming than I used to....)
On 11/25/19, 9:16 AM, "John Piechowski" <firstname.lastname@example.org on behalf of email@example.com> wrote:
I'm guessing many of us using DITA receive similar feedback from reviewers that they don't like how text flows in documents.
Often, they relate back to when we were using DTP, and we could insert a page/column break to force the section header or headline that renders on the bottom of one page to the next page with it's content. Or sometimes there are one or two lines of text left at the bottom on a column that they prefer are with the content on the next column.
Sometimes it is a headline and short paragraph that stays on one page with the image that it references on the next page, making the two disconnected.
For many years, it was easy (maybe too easy) to give them exactly what they wanted using DTP. Now as we promote content reuse, we lost this control without negatively impacting other documents that use the same topic.
We've tried to explain the great benefits we are getting through content reuse: consistency, reduces translation, improved quality. But often, DITA is seen outside the team as a step back in quality.
I'm interested in what others are doing to balance the positive outcome of content reuse against the negative effect document layout.