首页>
外国专利>
Compiling an application program written in legacy source code according to stack or legacy semantics based on equivalence
Compiling an application program written in legacy source code according to stack or legacy semantics based on equivalence
展开▼
机译:根据堆栈或基于等价语义的旧版语义编译用旧版源代码编写的应用程序
展开▼
页面导航
摘要
著录项
相似文献
摘要
A computer system (1200) prepares an application program (1210) comprising legacy source code written in, for example, COBOL for deployment into a processing environment (1260). The system has a control flow analyser unit (1242) operable to determine equivalence in control flow between source code if compiled using stack semantics and if compiled legacy semantics. Determination of equivalence involves generating and inspecting a directed call graph to identify strongly connected components. In the absence of such, tuples comprising ranges of functions, end points of functions and reaching endpoints of direct/indirect predecessor nodes are created and a union of the range and the reaching end point calculated. If the union is NULL and there are no strongly connected components then equivalence exists. The system has a compiler unit (1243) to compile the legacy source code using stack based semantics if equivalence is determined and using legacy semantics if non-equivalence is determined. This compilation may be to intermediate code such as Java Bytecode or common intermediate language (CIL) code.
展开▼