You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When implementing cmdipass, I noticed that if you submit a query that doesn't have a scheme (e.g. http://), you get the entire database back. I assume that getting the entire database will be fixed by #30. But the official implementation of KeePassHttp doesn't seem to care about the protocol.
That is, if I have a database with an entry called test-token and I run cmdipass get test-token against KeePassHttp, I'll just get back the one entry. If I run that against MacPassHTTP, I'll get the whole database.
If the user alters their command to add a protocol, e.g. cmdipass get http://test-token the command will execute as expected (only getting one entry) against both implementations. However, since the protocol doesn't seem to be strictly necessary by the reference implementation, I'm opening this Issue since it probably shouldn't be required by MacPassHTTP either.
The text was updated successfully, but these errors were encountered:
Fitzsimmons
changed the title
Support for queries that don't have a protocol prefix
Support for queries that don't have a scheme
Feb 13, 2017
When implementing cmdipass, I noticed that if you submit a query that doesn't have a scheme (e.g.
http://
), you get the entire database back. I assume that getting the entire database will be fixed by #30. But the official implementation of KeePassHttp doesn't seem to care about the protocol.That is, if I have a database with an entry called
test-token
and I runcmdipass get test-token
against KeePassHttp, I'll just get back the one entry. If I run that against MacPassHTTP, I'll get the whole database.If the user alters their command to add a protocol, e.g.
cmdipass get http://test-token
the command will execute as expected (only getting one entry) against both implementations. However, since the protocol doesn't seem to be strictly necessary by the reference implementation, I'm opening this Issue since it probably shouldn't be required by MacPassHTTP either.The text was updated successfully, but these errors were encountered: