Decrypt does not work on outlines

[JorjMcKie]

In the process of establishing a more formal testing approach for PyMuPDF, I ran into the following problem:

I password protected a test pdf. Then I used the demo program removePass.py to create a non-protected version. This version was readable allright, but all outline data were still encrypted!!.
If I open the encrypted pdf with a standard viewer like Nitro, the outline looks fine.
Also no problem occurs with SumatraPDF.exe (which is based on MuPDF).

[JorjMcKie]

here is what removePass.py does with test_encrypted.pdf:
x.pdf

[JorjMcKie]

If I may guess: it is probably a timing issue.
I believe @rk700 you told me once that we instantiate the outline when the Document object is created. If we later decrypt a document, these outline data (and maybe metadata too) do not get updated.

[rk700]

Yes I think it's very likely to be true. Maybe we should update the outline when the doc is decrypted.

I'll try later and see if that works.

[rk700]

I just updated fitz.i and removePass.py. Now the outline of an encrypted document will not be initialized until it is decrypted. Hope that I didn't break other things.

[JorjMcKie]

I tested the new version:
It works ok for my test cases with non-encrypted files, but decryption seems to no longer work.
What happens is:

• doc.authenticate() returns 1
• doc.save() does create a copy which is not flagged as encrypted, but it in fact is still encrypted.

[rk700]

@JorjMcKie could you please provide an encrypted pdf and password as well?

[JorjMcKie]

Unencrypted:
test.pdf
encryption password: "encrypt_pw"
encrypted pdf:
encrypted.pdf

[rk700]

I did make a mistake:(

Now the updates are:

• doc.authenticate() returns 1(True) if authenticated successfully, that is, the password is correct. I made this update for consistency with the mupdf upstream and our document.
• I found that besides outline, metadata like author name, document title, are not decrypted either. So I move all those stuffs in a new function called initData, which will only be invoked either the doc is plain or when it is decrypted.

[JorjMcKie]

I executed my tests again with the new version - everything works fine again!
I will document the implied changes to the API.
I will also do a few minor things to prevent unexpected behaviors in case of encrypted and not yet authenticated files, like e.g. doc.metadata, doc.outline attributes exist, but lead to exceptions when being used, etc.