rails invalid byte sequence in UTF-8

[ghost]

Hello, i got this error trying to OCR this pdf document: https://www.dropbox.com/s/ko76kalp5p59hwc/contrato%20de%20fianza%20prueba%2010.pdf?dl=0

The code which fails is: Docsplit.extract_text(attachment.path, :output => output_dir, :language => 'spa').

I have tried using:

• Docsplit.extract_text(attachment.path, :output => output_dir, :language => 'spa', :no_clean => true)
• Docsplit.extract_text(attachment.path, :output => output_dir, :language => 'spa', :no_clean => false)
• Docsplit.extract_text(attachment.path, :output => output_dir, :no_clean => true)
• Docsplit.extract_text(attachment.path, :output => output_dir, :no_clean => false)

but non of the above is helping, still fails. A lot of other pdf documents works great.

My environment: Rails 4.2 Ruby 2.2 Docsplit 0.7.6 tesseract-ocr 3.03 tesseract-ocr-spa 3.02

Any help please?

[tbk303]

Check PR #134 that might fix your problem.