Some of you who have upgraded to SEG 3.6 and SWG 3.0 may have noticed that the processing of PDF documents occasionally generates some issues that you have never seen before.
Previously we used PDF libraries that would occasionally have issues with text detection and the extraction of embedded objects (such as images).
These libraries did not offer the means to inspect meta-data such as document properties.
They also did not provide us with one of the key requirements for the Adaptive Redaction features, that being the ability to modify content in the PDF according to policy.
To support content inspection and provide a means to modify it according to policy, we have spent over a year building our own PDF handler.
As a result we have been able to improve the detection of words and phrases in PDF and while we have tested this to great lengths, it has not been possible to test against the multiple variations of PDF documents in existence. Therefore, it is possible that the occasional PDF document will cause a problem.
We have had a few reports of PDF processing issues from customers running SEG 3.6 or SWG 3.0, all of which have been fixed in 3.6.1 and 3.0.1.
However, there is still a chance that you might find others PDF documents that fail to process correctly.
If this happens, even if the error message returned has been seen before, please contact Support and supply the file to help us fix the problem as quickly as possible.