[oXygen-user] oXygen to WordPress
Wed Jan 6 12:20:25 CST 2016
On 5/01/2016 6:52 pm, Oxygen XML Editor Support (Radu Coravu) wrote:
> Hi Ben,
> We have not (yet) implemented such a "Publish To WordPress" feature.
> And also right now this is not on our todo list.
> Oxygen has lots of API so hypothetically speaking an Oxygen plugin could
> be built for this.
> Or you can just publish to XHTML and then upload that XHTML content to
> the WordPress web site.
Yeah, I suspect I'll opt for this or a variant of it. Probably
something along the lines of DITA to [X]HTML 5 (by file rather than
map) and then run a command line script to do the rest (or call via
the external applications feature).
> A problem would be with referenced images. Usually when I blog, I blog
> in DITA and I embed the images in the XHTML content using base 64
> encoding, thus the XHTML which I paste in the uploaded blog content
> contains all the information inside it.
I did something similar for a userstyles.org theme (i.e. just a CSS
file) for the ICCF website (iccf.com). I've also got scripts that let
you feed a CSS file with image references in at one end and get a
rewritten CSS with base64 encoded image data in them out the other.
They're over on Github somewhere, I can't remember which repo off the
top of my head.
The only thing is my WordPress site is sitting on an AWS EC2 instance
and is configured to move all the media files to an S3 bucket and then
rewrite the references to load from there (to improve loading and
response times as well as reduce traffic costs). Still that can be
side-stepped entirely by uploading straight to an S3 bucket and
setting the references directly in the HTML (or it might be able to
use the current plugin method on the blog via the API, I haven't
peered that deeply under the hood yet).
I think for now, though, I'll focus on the problem that is really
bugging me: how to get DITA (specifically D4P) to ePub 3.0 with only a
minor amount of post build editing (I expect to have to check and/or
mess with the metadata after generation, but not every file). Though
the webhelp transformation looks like it's about 90% of the way there,
it just needs to generate an NCX file and manifest from that context
XML file and a TOC Nav file and spine from the index file, then piece
those together with the metadata for the OPF and bundle the whole
thing up and it's done, more or less.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 630 bytes
Desc: OpenPGP digital signature
More information about the oXygen-user