Bug#948891: licensecheck --encoding utf8 exits on error when parsing binary files

[Jonas Smedegaard]

hi again, Dominique,

Quoting Dominique Dumont (2020-01-14 12:06:27)
> I'd suggest to either:
> - skip binary files
> - read binary file without utf8 encoding (even if --encoding utf8 is used to run licensecheck)
> - for image files, use exiftool or Image::Exif to extract license information from copyright tags See Image::ExifTool::TagNames for tag list (which unfortunately depend on file format)

I forgot to comment on your suggestions previously - and apologize if I 
came across as rude.

1st suggestion I considered a wrong approach at first when I read this 
bugreport, but now that you discovered it as a bug (missing "use 
Try::Tiny") I thought again and it obviously makes sense to not _halt_ 
the program but continue with other files.

2nd suggestion feels wrong to me.  Please do try again and argue your 
case if you feel it makes sense to ignore an explicit encoding request.

Your 3rd suggestion is sensible in a broader sense - i.e. not 
specifically tied to use of the "--encoding utf8" option: I agree that 
licensecheck should ideally parse structured data in addition to its 
core task: Sifting through unstructured scriblings in comments and do a 
best-effort job of making sense of it.  I have ideas on how to approach 
that (starting from "licenseminer" script now part of CDBS).  If you 
want to discuss that further, then I suggest to file a separate wishlist 
bugreport about support for parsing structured data.


Kind regards.

 - Jonas

-- 
 * Jonas Smedegaard - idealist & Internet-arkitekt
 * Tlf.: +45 40843136  Website: http://dr.jones.dk/

 [x] quote me freely  [ ] ask before reusing  [ ] keep private
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: signature
URL: <http://alioth-lists.debian.net/pipermail/pkg-perl-maintainers/attachments/20200131/d811e08e/attachment.sig>