Re: [Groff] problem with preconv and sample_docs.mom

[Bertrand Garrigues]

Hi Branden,

On Fri, Nov 03 2017 at 11:56:26 PM, "G. Branden Robinson" <address@hidden> 
wrote:
[...]
> $ ./preconv -d contrib/mom/examples/mom-pdf.mom
> default encoding: `UTF-8'
> file `contrib/mom/examples/mom-pdf.mom':
>   no encoding tag
>   len: 21946
>   uchardet read: 21946 bytes
>   charset: 
> ./preconv: encoding `' not supported, not a portable encoding

The result of `uchardet_get_charset' gives nothing, which is not normal.
The `uchardet' lib is an optional dependency used by preconv to detect
the encoding of the input file.  I've added it several months ago
because `mom' example files could fail badly (preconv would default to
latin1).

On HEAD I don't have any problem to generate mom examples pdf, and

  preconv -d mom-pdf.mom

gives me:

default encoding: `UTF-8'
file `mom-pdf.mom':
  no encoding tag
  len: 21946
  uchardet read: 21946 bytes
  charset: ASCII
  encoding used: `US-ASCII'

Which version of `uchardet' do you have on your system?  On mine I have
version 0.0.6.

Another thing is that if `uchardet_get_charse' returns NULL we use the
default encoding, but we could also test that this function returns an
empty string (which should be your case).

Regards,

Bertrand Garrigues