CONV versus XCNV and ICNVs

Sometimes it is difficult to figure out which command is the right one. Especially for the conversion commands (CONV/XCNV/ICNV) understanding the differences is important:

CONV:

• simple to use, more automatisms, fewer parameters
• the user must not know how the data is formatted or where it came from
• the conversion is done on a best-fit basis for platform specific formats
• there are no possibilities to do dirty things
• converted data can be written to only a single target
• the automated detection and conversion can result in more CPU usage
• only global pre- and post-processing at read and write is supported

XCNV:

• gives access to all capabilities of FLUC (more complex to use)
• the user must know how the inout data is formatted and where it came from
• the conversion is done as good as possible to the original data
• dirty things (e.g. manipulation of file modification time) are possible
• conversions are limited to a minimum of CPU usage
• output can be written to multiple targets via an array of file I/O specifications
• supports global pre- and post-processing and pre- and post-processing per file

ICNV:

• provides only character set conversion for block or record-oriented files
• is much simpler than CONV and XCNV