The Design And Implementation Of Script Language For General Relational Database

Oracle, SQL Server, DB2, MySQL, Sybase, Interbase are the most popular database systems nowadays. These systems running on windows, Linux and Mac system, occupy more than 90% of the relational database market. These systems with a series of forms have a wide range of constantly changing business. They have been used in the railway, banking, communications, power and other industries. They are stored in the database as procedures or triggers, and the others are directly embedded in the source codes. Once the business logic changes or upgrading the database, some changes need to be made to the business logic. But it is difficult to be changed for the extended SQL syntax provided differently by databases.This paper designs a common database scripting language based on C language, removes the underlying functions such as the pointer, the address-operation, the pretreatment, the function pointers, assembly language support, etc. It keeps the basic functions such as numerical operations, logic operations, statements, functions, structures, enumerations, arrays and so on. It also increases the database manipulation functions as well as the standard SQL statements translation function, which provides a unified interfaces for the database SQL statements of a variety of database systems and makes all the business logic "write once, apply several times".The source code compiled syntax tree as a binary file is stored, which makes it possible to be called quickly and saves the time. The file uses relative offset addresses, which requires a continuous address spaces to compile syntax in the memory in order to calculate the offset easily. The difficulty is that how to estimate the space occupied by the target file, a compromise approach is adopted to compile twice, the firsttime is to calculate the space and the second time is to allocate it. The script language does not apply to the systems required a higher processing speed and real-time systems, because the script run outside the database system with a slower speed.