Can't convert doc to html

[zhongjin616]

hi, I test unoconvert with libreoffice7.1.7

unoconvert 97html转换文档.doc 977.html

cause an error:

INFO:unoserver:Starting unoconverter.
INFO:unoserver:Opening 97html转换文档.doc
Traceback (most recent call last):
  File "/root/miniconda3/envs/docvert/bin/unoconvert", line 9, in <module>
    sys.exit(main())
  File "/root/miniconda3/envs/docvert/lib/python3.8/site-packages/unoserver/converter.py", line 246, in main
    result = converter.convert(
  File "/root/miniconda3/envs/docvert/lib/python3.8/site-packages/unoserver/converter.py", line 186, in convert
    raise RuntimeError(
RuntimeError: Could not find an export filter from com.sun.star.text.TextDocument to graphic_HTML

but i can do libreoffice --headless --convert-to html html转换文档.doc succeed.

[regebro]

Could you share the document? It doesn't seem to be a generalproblem.

[lublak]

@regebro i have the same issue with a pptx file.
Just create a simple pptx and use unoserver unoconvert somepresi.pptx somehtml.html.
Currently i can't send this file because there are some personal datas (author, last edit user etc.)
Could not find an export filter from com.sun.star.presentation.PresentationDocument to generic_HTML

[lublak]

I can force with: filtername = "impress_html_Export"
But than i only get a single html file. It would be nice to support a folder export with a complete html structur. Or with embed images

[regebro]

@lublak Libreoffice sees presentations as a graphical format, and html as document format, so there wouldn't be much to convert at all. It can only do useful conversations of presentations to PDF, IMO.

What is it you are attempting to do?

[lublak]

@regebro

I try to export presentation as full web pages. (And that automatically in the background.)

It is possible to convert a presentation to a complete html page via the interface. So LibreOffice already has the possibility. Only how it looks like via the command line I do not know.

Export:

as html

Standard-HTML

[regebro]

Yeah, that includes defining export formats, etc, which I don't know how to do. It's possible it can be done if we implement support for filter flags, but I'm not sure even then.

If you can figure out how to do it with LibreOffice from the command line, I can look at implementing support for that.

[felixble]

Hi @regebro,

first of all thanks a lot for your great work with this package!

We are having the same issue when using unoconvert to convert odt to html.

Executing unoconvert --convert-to html test.odt test.html causes the following error:

INFO:unoserver:Starting unoconverter.
INFO:unoserver:Opening //b577c1152d56441fa928bd54d914ee07.odt
Traceback (most recent call last):
  File "/usr/local/bin/unoconvert", line 10, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.7/dist-packages/unoserver/converter.py", line 248, in main
    inpath=args.infile, outpath=args.outfile, convert_to=args.convert_to
  File "/usr/local/lib/python3.7/dist-packages/unoserver/converter.py", line 186, in convert
    f"Could not find an export filter from {import_type} to {export_type}"
RuntimeError: Could not find an export filter from com.sun.star.text.TextDocument to graphic_HTML

We can do it with LibreOffice from the command line with the following command: soffice --headless --convert-to html test.odt.

This produces the following output on the cli: convert /data/test.odt -> /data/test.html using filter : HTML (StarWriter).

It looks like there is an issue when figuring out the correct filter in

unoserver/src/unoserver/converter.py

Line 87 in 3e30d67

def find_filter(self, import_type, export_type):

EDIT:
Setting the variable filtername hardcoded to "HTML (StarWriter)" with the following line in

unoserver/src/unoserver/converter.py

Line 183 in 3e30d67

filtername = self.find_filter(import_type, export_type)

works in our tests:
filtername = "HTML (StarWriter)"

Could you please have a look at this? Can this filter be added?

Thanks!

[asmundstavdahl]

Can be solved with #59