_parse_pdf_to_string: catch and ignore PDFException.

This function is a best effort attempt to get text from a PDF file -
if the file is malformed (PDFSyntaxError), this function returns an
empty string.  However, there are other exceptions such as
PDFPasswordIncorrect that can occur even if the file is well-formed.

Although it would be better to handle these exceptions at a higher
level, this is a temporary fix to allow training applications
containing encrypted files to be rejected.

physionet-django/console/services.py

@@ -4,7 +4,7 @@
 from typing import Optional
 
 from pdfminer.high_level import extract_text
-from pdfminer.pdfparser import PDFSyntaxError
+from pdfminer.pdfparser import PDFException
 from django.conf import settings
 
 from user.models import Training
@@ -29,7 +29,7 @@ def _get_regex_value_from_text(text: str, regex: str) -> Optional[str]:
 def _parse_pdf_to_string(training_path: str) -> str:
     try:
         text = extract_text(training_path)
-    except PDFSyntaxError:
+    except PDFException:
         text = ''
         logging.error(f'Failed to extract text from {training_path}')
     return ' '.join(text.split())