SQLite Database Encryption (2)

Databases
22

https://www.sqlite.org/see/doc/trunk/www/readme.wiki
Seems to imply that the key prefix would work in a PRAGMA key/hexkey statement if the full SEE library is included.

23

[quote=344347:@John A Knight, Jr]https://www.sqlite.org/see/doc/trunk/www/readme.wiki
Seems to imply that the key prefix would work in a PRAGMA key/hexkey statement if the full SEE library is included.[/quote]
That was what I was thinking/hopeing… only not sure how to really verify that…
because those same docs say if you get the syntax “wrong” or if it is not supported then the prefix just becomes part of the key

so in any case. it would look like it worked, but you don’t really know if it applied “myKey” using “aes256” or if it encrypted using the string “aes256:mykey”

24

[quote=344352:@Dave S]That was what I was thinking/hopeing… only not sure how to really verify that…
because those same docs say if you get the syntax “wrong” or if it is not supported then the prefix just becomes part of the key

so in any case. it would look like it worked, but you don’t really know if it applied “myKey” using “aes256” or if it encrypted using the string “aes256:mykey”[/quote]
Our plugin currently only supports AES128, and does not do anything special with the prefix. There’s a case in feedback for supporting it.

25

On the plus side is the data will still be encrypted, just not with the key/algorithm you expect.

The Bad side of silently using a special prefix as part of the key: after an upgrade to a system that honors the prefix.
The key that you’ve been using is now incorrect. Yikes! Where’s my data?

For Tadpole, it may be a good idea to check the key and display a firm warning on any attempt to encrypt until Xojo actually supports the alternate formats.