Read database is slow

[Alireza Khorshidi]

Probably the reason is that, reading of one database happens multiple times. Should try to reduce it as much as possible:

https://bitbucket.org/andrewpeterson/amp/src/f8612f4754c60060c2c3f7498589c4b5a754edcc/utilities.py?at=master&fileviewer=file-view-default#utilities.py-545

[Jacob Boes]

The following code performs a single iteration over the database:

https://github.com/jboes/amp/commit/43313f5aaa0f1112ba35e8d4c9a642f26976af23

[Jacob Boes]

The previous code did not pass all of the write tests in the newer commit: 8b43104 (not using the the most recent Fortran update since it doesn't appear to be working).

The following succeeds for multi-processing, but does not enforce ordering of the lists, I'm not sure if that is a critical feature: https://github.com/jboes/amp/commit/b0b118ea767a7f282e9201cbfdd7473d51a5c3c5

[Alireza Khorshidi]

Thanks Jake! There was a bug in your implementation, as well as the previous version of the code for reading databases. I fixed the bug, extended the tests folder, and sped up reading databases (now it reads within a single iteration only). These changes are performed in the commits 22a0c4d and d77911f.

[Alireza Khorshidi]

Reading databases is still a bit slower than reading json files, but this is probably due to the nature of db reading data line by line, as compared to json loading data all at once.