PDFFont#getEncodingFromFont and incorrect usage of StringTokenizer

["martijn.list" <ma...@gmail.com>]

The PDF document attached to bug report
https://issues.apache.org/jira/browse/PDFBOX-816 (TaxReturn-1.pdf)
throws a NumberFormatException.

#getEncodingFromFont uses a StringTokenizer to split a line into
separate tokens:

StringTokenizer st = new StringTokenizer(line);

The following line however results in a NumberFormatException because
0/NUL is read as one token.

dup 0/NUL put

The StringTokenizer only accepts the following chars as line delimiters:
" \t\n\r\f".

I think this is not correct because it seems that some delimiter chars
are missing like (, ), <, >, [, ], {, }, /, and %

Is this a bug?

Kind regards,

Martijn Brinkers

Re: PDFFont#getEncodingFromFont and incorrect usage of StringTokenizer

[Andreas Lehmkuehler <an...@lehmi.de>]

Hi,


Am 01.12.2010 22:40, schrieb martijn.list:
> The PDF document attached to bug report
> https://issues.apache.org/jira/browse/PDFBOX-816 (TaxReturn-1.pdf)
> throws a NumberFormatException.
>
> #getEncodingFromFont uses a StringTokenizer to split a line into
> separate tokens:
>
> StringTokenizer st = new StringTokenizer(line);
>
> The following line however results in a NumberFormatException because
> 0/NUL is read as one token.
>
> dup 0/NUL put
>
> The StringTokenizer only accepts the following chars as line delimiters:
> " \t\n\r\f".
>
> I think this is not correct because it seems that some delimiter chars
> are missing like (, ),<,>, [, ], {, }, /, and %
Hmm, the problem are not the unsupported delimiter chars, it's the missing space 
character in the input line.

dup 0/NUL put -> dup 0 /NUL put

> Is this a bug?
Yes, it is. I filed an issue on JIRA [1] and fixed the problem by replacing each 
"/" with " /" to ensure that there will be a delimiter at the right place.

Thanks, for reporting!


BR Andreas Lehmkühler
[1] https://issues.apache.org/jira/browse/PDFBOX-921