Tabular network database

Tags: tabular network database, hypertable, HT, HTMS, object-oriented software, nosql, python

This project is a new, original approach to building data models and implementing software at the junction of the concepts of relational DBMS, network DBMS and the object-relational mapping (ORM) concept. The new data model was called a "Tabular Network". Tabular network database may be of interest in subject areas in which existing SQL and NoSQL systems are not always effective: in AI, in big data processing, in adaptive applications with self-modifying database structures, in embedded systems etc.

Tabular network database (TNDB) or the hypertable (HT) - is a new type of databases.

The hypertable is a system of cross-referenced tables:

• records are grouped into tables, and all records (rows) in the same table have the same set of fields (columns);
• tables do not have headings: the set of columns of any table is a certain subset of common hypertable's attributes, i.e tabular structures in HT are defined as projections of a common attribute set;
• there are two disjoint attribute classes:
  • attributes, called data type attributes - DTA, define fields with self-defined data values;
  • attributes, called reference type attributes - RTA, define fields with values that are interpreted as links to rows in tables and represent their physical address pairs (table number, row number), i.e. the values of RTA fields in the rows of the tables are explicit links to any rows in any tables in the hypertable;
• field values can be not only atomic, but also sets.

The first functionally complete implementation of the TNDB application programming interface (API) was called the HyperTable Management System - HTMS.

Simplicity was one of the main goals of creating HTMS. Using the HTMS is no more complicated than the APIs of well-known ORMs such as Django ORM. But the tabular network model is much closer to the real world than the relational model, so to design the databases easier.

Full description of the tabular network data model see on medium.com website: Tabular network data model series

The HTMS includes the following levels and packages:

• The "object-level" of the HTMS includes HT_Obj class for creating the HT objects, and Obj_RAM class for mapping (transforming) objects to/from the rows of tables. The main functional class at this level is Obj_RAM, the methods of which provide the basic operations for manipulating, searching and filtering data, which are similar in results to the objects methods in the ORM;
• The "mid-level" of the HTMS API is a set of classes and functions that form the basis for object-level: for organizing HT structures in general as well as models of individual tables (an analog of the models in Django ORM) and support common operations with the attributes, tables, rows and fields (create, change, delete etc.) and for formation search waves;
• The "low-level" HTMS API is a set of classes and utility functions - the basis for the mid-level and object-level;
• The "file-level" HTMS API is a subsystem Cage of remote cached access to database files on servers on the network (description see github.com/Arselon/Cage).

The HTMS user interface created as a universal HT screen editor - HTed, implemented as a website. It can connect to any data server regardless of application. HTed is used to create, design, and edit database files compatible with the HTMS.

For most applications, using object-level and mid-level classes and functions is sufficient. Both of them together are a logical level of the HTMS. Packages are available on PyPI: htms-obj and htms-mid-api. 

Low-level and file-level - for advanced developers if they want to get inside the HTMS technology. Both of them together are a physical level of the HTMS. Packages are available on PyPI: htms-low-api and cage. 

Copyright 2018-2021 Arslan S. Aliev

Software licensed under the Apache License, Version 2.0 (the "License"); you may not use this software except in compliance with the License. You may obtain a copy of the License at www.apache.org. Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.