We need a handler for the database messages, that will manage local databases and perform operations on them, based on received messages.
It needs to have it's own config file, where it lists:
How often must database be saved on the disk (while it is loaded in memory);
Where it is stored;
Where/how often backup copies are stored. How much to store at once.
It must provide following commands:
Create/remove/exists check for database's groups;
Create/remove/exists check for files inside database;
Create/remove/exists check for values in side files;
Set/Get commands for values in files that overwrite them;
Aside from a simple set of commands, we will need to support a number of value-changing commands such as add to int/float, add(append) value to JSON object/array.
These names do not have to be preserved, but they have to be documented.
It also must implement and document a number of response messages that report the result: success/failure (different type of failure per reason?).
Adding more is possible if deemed necessary during the implementation.
We need a handler for the database messages, that will manage local databases and perform operations on them, based on received messages.
It needs to have it's own config file, where it lists:
1. How often must database be saved on the disk (while it is loaded in memory);
2. Where it is stored;
3. Where/how often backup copies are stored. How much to store at once.
It must provide following commands:
1. Create/remove/exists check for database's groups;
2. Create/remove/exists check for files inside database;
3. Create/remove/exists check for values in side files;
4. Set/Get commands for values in files that overwrite them;
5. Aside from a simple set of commands, we will need to support a number of value-changing commands such as add to int/float, add(append) value to JSON object/array.
These names do not have to be preserved, but they have to be documented.
It also must implement and document a number of response messages that report the result: success/failure (different type of failure per reason?).
Adding more is possible if deemed necessary during the implementation.
We need a handler for the database messages, that will manage local databases and perform operations on them, based on received messages.
It needs to have it's own config file, where it lists:
It must provide following commands:
These names do not have to be preserved, but they have to be documented.
It also must implement and document a number of response messages that report the result: success/failure (different type of failure per reason?).
Adding more is possible if deemed necessary during the implementation.
Do we even need several DBs?
Corrected text a little, removed requirement for multiple databases.