Encrypt can also be a Dictionary

[samalloing]

Hi @carlwilson

Encrypt can be also be a PdfDictionary (also used Arlington PDF model to verify)

A testfile: https://github.com/mozilla/pdf.js/blob/master/test/pdfs/issue6010_1.pdf

Sam

[codecov]

Codecov Report

Base: 46.33% // Head: 46.32% // Decreases project coverage by -0.01% ⚠️

Coverage data is based on head (7f84e7b) compared to base (eadc199).
Patch has no changes to coverable lines.

❗ Current head 7f84e7b differs from pull request most recent head 2f22a39. Consider uploading reports for the commit 2f22a39 to get more accurate results

Additional details and impacted files

@@                Coverage Diff                @@
##             integration     #783      +/-   ##
=================================================
- Coverage          46.33%   46.32%   -0.02%     
+ Complexity          1054     1053       -1     
=================================================
  Files                 57       57              
  Lines               9079     9079              
  Branches            1612     1609       -3     
=================================================
- Hits                4207     4206       -1     
- Misses              4333     4335       +2     
+ Partials             539      538       -1     

Impacted Files	Coverage Δ
...a/edu/harvard/hul/ois/jhove/NisoImageMetadata.java	75.99% <0.00%> (-0.20%)	⬇️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[carlwilson]

This is a solid approach, but the if/else leaves the possibility that the instance returned here isn't "tested" at all. Put another way, what happens when encryptObj ISN'T an instance of PdfIndirectObj OR PdfDictionary? This may be an error that gets picked up elsewhere, but I wonder if there shouldn't be a default condition here that marks an error?

[carlwilson]

@samalloing I've also created PR #803, which sets things up for this change. I've added you as a reviewer anyway, but the two want merging before trying to merge into integration. I'm good to handle that but want to keep them separate until the review is resolved. I actually merged them up instead, I'll fix the tests once the changes are made.

BTW this appears to be a duplicate of #743 ? Please confirm and I'll close that PR.

[carlwilson]

@samalloing I think I've fixed these issues in #810 where I've added the changes needed to report encryption in the new case. Finally, I've added a couple of examples to the corpora and patched the tests to work.

If you could review #810 we can get these merged up.

[carlwilson]

Closed as contained in #810 now, thanks @samalloing