ebook-convert error on PDF output without X

[hiperlink]

Hi all,

I'm running a headless server to convert some recipes to .mobi which works just fine.

Now I wanted to convert e.g. modoros.recipe to PDF on the same server, and I had got:

The cover image has an id != "cover". Renaming to work around bug in Nook Color
Traceback (most recent call last):
  File "site.py", line 58, in main
  File "site-packages/calibre/ebooks/conversion/cli.py", line 282, in main
  File "site-packages/calibre/ebooks/conversion/plumber.py", line 1033, in run
  File "site-packages/calibre/ebooks/pdf/output.py", line 90, in convert
  File "site-packages/calibre/ebooks/pdf/output.py", line 119, in convert_text
  File "site-packages/calibre/ebooks/pdf/output.py", line 122, in write
  File "site-packages/calibre/ebooks/pdf/writer.py", line 116, in __init__
Exception: Not OK to use Qt

Using this command:

ebook-convert "${RECIPE}" "${RECIPE_NAME}.pdf" \
 --smarten-punctuation  \
 --change-justification justify \
 --paper-size a4  \
 --pretty-print  \
 --preserve-cover-aspect-ratio  \
 --insert-blank-line  \
 --margin-bottom 72.0  \
 --margin-top 72.0  \
 --margin-left 72.0  \
 --margin-right 72.0  \
 -vv \
 --debug-pipeline debugdir/pdf \
 | tee debugdir/pdfdebug.log

What am I doing wrong? Or do I really need X and Qt for CLI conversions to PDF?

[ldolse]

You really need QT. X is not a requirement that I'm aware of, but the conversion engine uses some of QT's image handling libraries.

[Manichean]

Moderator Notice
Moved to appropriate forum.

As far as I know, Calibre needs X to be installed for just about everything because of some common Qt functions. There are some notes in the FAQ that may be of help to you.

[user_none]

You don't need a full X server installed you can use xvfb to allow Qt to run. PDF output for instance uses Qt for creating the PDF files. A number of other parts (MOBI output) will check for X / Qt and skip the processing steps that require it.