Version 2.0: the splitting of the node table

[ UPGRADE WARNING for Unidatab 2.0:
The new version brings structural changes, in consequence older databases have to be migrated to the new standard (see this link). Besides, some functions have been improved in syntax and output. Therefore old documentation will reveal inappropriate in some cases and will be corrected gradually in the near future. This warning scopes the WordPress blog more than the software’s sql_help files now installed in standard locations. ]

New structure:

Keeping nodes of different kinds in the same table was known to be an avoidable weakness and that was the main reason of this structural update. Now records and symbols have their dedicated spaces. This releases the amount of runtime memory in many operations. But the structure of this new symbol table is identical to the node table and one can consider the original version as more genuine in some sense.

Another change brought by this release is the renaming of the tables, in order to let Unidatab install easily next to an existing database (where tables are supposedly not named “unid_*”). This step is justified by the new analytic tools developed, potentially used in relation with a non-unid database.

New functionalities:

  • db stats for gnuplotting graphical representations of data structures
  • record stats with tree-representation of content
  • time is now taken till the milliseconds (useful for the db analysis)
  • 2 new record-objects remaining in-memory
  • 2 connections at a time in unid-adm

New functions:

  • ana_base(): db analysis tool, can work on subsets
  • ana_record(): record analysis tool
  • size(): size evaluation of structures and data involved in a  record
  • find_format(): an analogon of find_record()
  • duration(): returns the time spent in the execution of an sql request
  • time_flip(): numerical or litteral expression of a given time
  • uid() and uname(): better handling of user info

.. various clone functions related to the splitting of the node table ( make_symbol() etc.. )

This upgrade helped circumscribe, with the ability to represent the development of an information system in time, a field where Unidatab could be used with results. This reflexion also brought a simpler way of describing Unidatab. The next programming tasks include translating several functions, like list_occs(), into table-valued versions of themselves. Another task is the management of copy and paste operations between unidbs, a step in this direction has been taken with a second db connection in unid-adm.

[2018 edit]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s