poslq-simple 0.2.1

A simple nosql json storage for small projects.

project

Looking for contributions.

What ?

really simple no-sql database ( for small projects ) using json.

Documentation

read the docs.md file.

Quickstart ( simple example )

from posql_simple import w_out_sql.py
my_password = "hello world" # for cipher coding ( key )
path = "./default.pst"      # use the pst extension
my_db = Database (__pass =  password, path = path)
my_collection_name = "clients"
my_collection = Collection ( collect_name = my_collection_name) # creating the collection
db.bind_new_collection(my_collection) # binding the collection to the db

client1 = {
    "name" : "hello world"
}
my_collection.add_record( client1 )  # adding a record to the collection
db.save()                            # saving ( writes data to the file )

Documentation

Database ( class )

init method requires pass ( for encrypting and decrypting ), optionally a dict full of collections, and the path of the file ( where to store the content ).

save method is saving the data in the file, takes one positional argument for debuging purpose ( default value : True ).

load returns a Database object. The arguments are path ( the path of the file ) and the decrypt_pass. Throws FileNotFoundError err if the file doesn't exist and ( or ) InvalidFileExtensionError ( if the file extension doesn't match the .pst ).

bind_new_collection binds the collection to the database object ( for data persistence ). Takes a collection as Param.

get_collection returns a collection object. Args : coll_name ( name of the collection ). get_collections returns a dict with all the collections.

Collection

init method requires a name ( collect_name ), and a list of the json records ( the data persisted ) which is optinal.

add_record method takes one positional argument which is the json obj that you want to store. Returns a boolean ( True, if the transaction is OK, and false if it's not ok ). Each object is padded with the "_id" entry ( uuid generated string ). You can add your own custom "_id" entry.

delete_by_id method takes one positional param which is the id ( making a linear search ). Like the add_record method, it returns a boolean.

find_by_id same as delete_by_id ; returning deep copy of the dict ( avoid mutability ) or else None.

update_obj : takes two positional args : new_obj ( the obj to persist, must contain an "_id" entry , or else throwing an error ), returning a bool.

search_by : takes two positional args : slot_name which is the name of the entry you want ( for instance the "name" entry of, each dict ) and the slot_value. Returns None if not found.

get_all_slot : returns a list of all the entries.

this lib supports multithreading queries by using synchronization.