Re: [O] org-ref exports nicely to html but not to odt

[Ken Mankoff]

On 2019-01-04 at 22:37 +0100, Uwe Brauer <address@hidden> wrote:
> I exported the org file in question to latex (I could also start
> directly with latex for that matter.)

Yes.

> Could I stick with bibtex?

Yes. No need to use biber + biblatex, you can stick with bibtex.

I only run that command because my massive 3000+ bib file contains things that 
pandoc has trouble with (accents? Not sure). That command extracts the bibtex 
entries used in the current document.

As I write this I realize this means you don't even need to compile the PDF 
and/or run bibtex. If you have a TeX file and an appropriate accompanying 
BiBTeX file, that should work.

> pandoc -f latex -t docx -o test.docx --bibliography 
> /home/oub/texmf/bibtex/bib/bibgraf.bib  test.tex
>
> /home/oub/texmf/bibtex/bib/bibgraf.bib is my main bibtex date base.
>
> But this did not work pandoc gets frozen.
>
> What do I miss.

I don't know. Try with a small bib file with 1 entry (and a TeX file that only 
cites that one entry). Does that work? Then maybe your bibgraf.bib has the same 
problem mine does (although I recall pandoc issuing an error and crashing, not 
just freezing).

If this method works with a MWE, then the issue my be your full bib file. In 
this case you need to extract the entries used in this document (e.g. my 
method). Presumably this subset does not cause the pandoc issue. I also find 
extracting the entries useful for archiving a project.

You can extract entries as I showed, or "bibexport <file>.aux" (requires a 
compiled PDF), or 
http://korrekt.org/page/Note:How_to_Extract_Used_Entries_From_BibTeX_Files or 
other methods...

  -k.