Options without values, like "smart" or "table_of_contents," fail

[wcaleb]

I have been unable to get Docverter to accept requests that include options such as smart and table_of_contents that have no values other than "true" or "false."

For example, this command

  curl http://c.docverter.com/convert -F from=markdown -F to=pdf -F smart=true -F input_files[]=@input.txt

returns the errors contained in this gist.

I have tried the same command with different to= values and different input files, as well as true/false options other than smart=, and it looks like I get the same result.

I'm testing this on Mac OSX Lion.

[chrisvfritz]

I'm also experiencing this problem in Ruby:

Docverter::Conversion.run do |c|
      c.from    = "markdown"
      c.to      = "#{params[:download_type]}"
      c.content = params[:included_chapters].split(',').map {|id| Chapter.find(id.to_i).content}.join("\n\n")
      c.table_of_contents
end

I've also tried the variations mentioned by the first poster. No luck.

[jldiaz]

Also hit this bug, and it is a real problem for me because most of my markdown documents are utf-8, and they cannot be converted to pdf (utf-8 chars are dropped, see this question in Stack Overflow for example), unless the option --ascii is passed to pandoc which is not possible because it is a boolean option and those options seem to be broken.

[peterkeen]

This is fixed in v1.0.3 which is running on c.docverter.com. Using the Docverter API you'd do this:

Docverter::Conversion.run do |c|
      c.from    = "markdown"
      c.to      = "#{params[:download_type]}"
      c.content = params[:included_chapters].split(',').map {|id| Chapter.find(id.to_i).content}.join("\n\n")
      c.table_of_contents = true
end

Using curl you'd say -F table_of_contents=true.

Please comment here if you have any more problems getting it to work. I'll close the issue in a few days if everything's cool.

[jldiaz]

Thank you very much! I can confirm that ascii option is working now (tested from python, using httplib).

[wcaleb]

Thanks for fixing this! I tested the "smart" option with my Pythonista script and it worked great!