Mystery of the docs in the PDF transformation

Here should go questions about transforming XML with XSLT and FOP.
manthony
Posts: 22

Mystery of the docs in the PDF transformation

Tue Jul 18, 2017 9:45 pm

I have a four book maps in the same directory.

Code: Select all

docs/admin_guide/admin_guide_book.ditamap
docs/admin_guide/advanced_searches_guide_book.map
docs/admin_guide/internal_POC_guide_book.ditamap
docs/admin_guide/user_guide_book.ditamap
 



They are all using the same, project-level transformation and the transformation is specified at the Project level. For some reason, 1 book (admin_guide_book.ditamap) is always output to a different level than the other books.

Code: Select all

├── admin_guide
│   ├── advanced_searches_guide_book.pdf
│   ├── internal_POC_guide_book.pdf
│   └── user_guide_book.pdf
├── dita.list
├── dita.xml.properties
├── docs
│   └── admin_guide
│       └── admin_guide_book.pdf


This is extra docs directory for one book is a problem because we plan to automate further. Different books should not result in a different transformation output behavior. So, I tried a couple of more things to pinpoint the reason. I wanted to rule out the our custom PDF transform itself, so I tried using the base DITA Map PDF transformation. The output was this:

Code: Select all

./out/pdf
./out/pdf/admin_guide/advanced_searches_guide_book.pdf
./out/pdf/admin_guide/internal_POC_guide_book.pdf
./out/pdf/admin_guide/user_guide_book.pdf
./out/pdf/docs/admin_guide/admin_guide_book.pdf


The docs directory was still there. Could it be the base PDF transform cared somehow about whether the directory and the file had similar names? I tried changing the name of the book map again. I got this output

Code: Select all

./out/pdf/docs/admin_guide/admin_book.pdf


So, not that. At this point, it looks like the problem could be in the book's map file but the map file in VI and out looks identical to the others in its structure. I'm at a loss what to try next to debug this. I could code my build around the mysterious output path issue but it is driving me crazy at this point

Anybody got any ideas why this one particular book picks up an extra docs directory level when it outputs?
Radu
Posts: 5088

Re: Mystery of the docs in the PDF transformation

Wed Jul 19, 2017 9:11 am

Hi Mary,

Oxygen uses the DITA Open Toolkit publishing engine to publish DITA content to various output formats.
The publishing engine has always had publishing problems when you have references to resources outside of the folder where the main published DITA Map is located. For example in your case all the published DITA Maps are in a subfolder and all the other DITA or image resources are in folders outside of that subfolder.
In this topic we propose two solutions for this:

https://www.oxygenxml.com/doc/versions/19.0/ug-editor/topics/dita-ot-external-refs.html

As you want to automate the output (maybe have an integration server publishing the documentation) the second option (that specific parameter) only works when publishing from Oxygen.
So ideally you would move your top level DITA Maps to the "docs" folder so that all referenced resources are underneath them.

Regards,
Radu
Radu Coravu
<oXygen/> XML Editor
http://www.oxygenxml.com
manthony
Posts: 22

Re: Mystery of the docs in the PDF transformation

Fri Jul 28, 2017 2:03 am

Thanks Radu. I discovered this was the case with a bit more experimenting.

Return to “XSLT and FOP”

Who is online

Users browsing this forum: No registered users and 0 guests