instructions for adding html4publishers plugin?

[tomjohnson1492]

Can you provide current instructions for installing the html4publishers plugin into Oxygen? I found a set of detailed steps here, http://straygoat-technicalauthor.co.uk/ ... ml-editor/, but they didn't quite work. When I ran the transformation, it said failed. I'm not sure I configured it right.

Here's what I did:

1. Downloaded the main plugin from the release candidate folder here: http://sourceforge.net/projects/dita4publishers/files/.
2. Unzipped it and moved all the files into Oxygen's plugin folder.
3. Shut down and restarted Oxygen.
4. Ran the OT Integrator transform.
5. Copied the DITA XHTML transform and added a new parameter, setting the Name to transtype and the value to d4p-html5.
6. Ran the transform.

It failed. What am I doing wrong?

Also, just curious, but I thought there would be better instructions for installing this plugin into Oxygen. Isn't not that cool of a plugin?

[Radu]

Hi Tom,

This step is wrong:

Unzipped it and moved all the files into Oxygen's plugin folder.

You should have copied the DITA OT plugins to the OXYGEN_INSTALL_DIR/frameworks/dita/DITA-OT/plugins folder.

The Oxygen application also has a plugins folder in which somebody can add plugins which enhance editing with Oxygen (add toolbar buttons, custom actions, etc).
But the DITA for Publishers plugins are plugins for the DITA Open Toolkit which comes with Oxygen.

If this still does not work for you, please attach to an email or post here the entire output received from the DITA OT ANT process in the DITA OT console view after the transformation fails.

Regards,
Radu

[tomjohnson1492]

Thanks. But I actually did mean the OXYGEN_INSTALL_DIR/frameworks/dita/DITA-OT/plugins[/b] folder when I referred to the plugins folder in my initial post..

I'm just confused about the whole file architecture of the dita4publishers folder. When I download the latest release candidate and unzip it, and move it into that frameworks/dita/DITA-OT/plugins folder, this is what it looks like: http://www.screencast.com/t/O9u6QfXaD

I don't know what the net.sourceforge.dita4publishers.common.html and other similarly named files are. Were they there by default, or did I put them there in some earlier attempt?

The file structure of the release candidate that I moved in doesn't match the file structure of the other folders. It seems odd.

I feel like there should be clear documentation on how to integrate this plugin, but I haven't found it. If you could provide more detail, I would really appreciate it. I think it's just that I'm unfamiliar with adding plugins to Oxygen, and I assume most people who work with plugins already know how to install plugins, so they don't bother to explain the steps in detail. Thanks,

Tom

[Radu]

Hi Tom,

I'm just confused about the whole file architecture of the dita4publishers folder. When I download the latest release candidate and unzip it, and move it into that frameworks/dita/DITA-OT/plugins folder, this is what it looks like: http://www.screencast.com/t/O9u6QfXaD

The initial ZIP you downloaded was expanded to a folder called "dita4publishers-0.9.19RC12". In that folder there is an extra ZIP called "dita4publishers-toolkit-plugins-0.9.19RC12.zip". That ZIP contains the actual plugin folders which should be copied to the frameworks/dita/DITA-OT/plugins folder.

I don't know what the net.sourceforge.dita4publishers.common.html and other similarly named files are. Were they there by default, or did I put them there in some earlier attempt?

The DITA OT distribution which comes bundled with Oxygen already has some DITA 4 Publishers plugin installed in it (we were mostly interested in the plugins for producing EPUB). But they do not contain the HTML5 plugin you are interested in, it contained some resources with shady licensing and we had some doubts if they could be included in a commercial application.

When you copy add the new ones, you will need to overwrite some already existing folders.

Regards,
Radu

[tomjohnson1492]

Thanks, that clarifies things a lot and now I think I'm doing all the installation steps correctly. However, when I try to run the DITA OT Integrator, it fails.

See this screenshot: http://www.screencast.com/t/bH3F3jVeYq

Do you know why the OT Integrator might be failing?

[adrian]

Hi Tom,

The DITA OT Integrator may be failing because it doesn't have write access over the files involved. If you have installed Oxygen in 'Program Files' (default location), that's a common problem.
To work around this, you should quit Oxygen, run it as administrator (right click on its shortcut and pick Run as Administrator), then run the DITA OT Integrator transformation. If it is successful, quit Oxygen and start it again in normal mode.

If that's not the problem and it's still failing, please have a look in the DITA OT Transformation view from the results panel (bottom of the window) and send us the output log found there.

Regards,
Adrian

[tomjohnson1492]

I'm on a Mac, so there isn't a Run as Administrator option.

The error says F [ANT] Transformation failed. There aren't any more log files to send.

Can you try integrating the HTML5 plugin and see if you can get it to work?

[Radu]

Hi Tom,

When a DITA OT transformation or an ANT transformation fails, besides the view which shows that the transformation failed you should also have another view which shows all the output received from the ANT process. I uploaded an image to show you where the view can be found:

http://www.oxygenxml.com/forum/files/ta ... etails.png

Also, for DITA OT 1.8.4 which comes with Oxygen you should try to install directly the DITA For Publishers 0.9.19RC12 plugins which can be downloaded from here:

https://sourceforge.net/projects/dita4p ... candidate/

The plugins are in a beta stage but the stable version from what I remember might only work with DITA OT 1.7.

I tested on my side installing the plugins from the 0.9.19RC12 ZIP, running the integrator worked for me. Then I edited the default DITA to XHTML transformation scenario and added the transtype parameter with the value html5. The first run of the transformation seems to require network access to bring down a compressed ZIP, at the end in the output folder you should obtain a file called generated-index.html which should be opened manually and should work.

Regards,
Radu