Compiler-based scanner characteristics

With the compiler-based scanners, you use IBM® Enterprise compilers. While the proprietary scanner analyzes source files, the compiler-based scanners process associated data (SYSADATA) generated by the COBOL and PL/I compilers. Typically, a scanner and compiler will recognize slightly different information because their parsing and analysis technologies are rarely identical. However, using the compiler-based scanners makes it possible for you to get analysis results more consistent with compiled assets.

Compilers older than the Enterprise level compilers cannot be used with the compiler-based scanner. But, in many cases, assets written for the older compilers are compatible with the Enterprise compilers' syntax, and analyzing them with the compiler-based scanner could be considered in order to avoid the proprietary scanner's limitation.

• For an asset written for Enterprise COBOL, the compiler-based scanner with regular COPYBOOK scanning support is the typical selection. Use the enhanced COPYBOOK scanning support option only if regular COPYBOOK support is not sufficient.
• Restrictive to syntax other than the Enterprise COBOL or PL/I language.
• Full support for new syntax of the Enterprise COBOL or PL/I language.
• Full PL/I macro support; use the compiler-based scanner with enhanced INCLUDE support to get full PL/I macro support.
• No restriction for PL/I data names.
• Compiler option/directive statement support (limited to subset of options effective for the analysis; see "User-specified compiler-based scanner options" in Related References for details).
• Support provided for setting common SQL options.
• SQL syntax support level including DB2® V8.
• CICS® syntax support level - application programming interface (API) including TS 3.1; system programming interface (SPI) including TS 2.3.
• Slower analysis time as compared to the proprietary scanners.
• More system resources required for processing when compared to the proprietary scanners.