In keeping with our users' need to migrate their mission-critical mainframe applications to best-in-class open systems tools, CoSORT also provides an intelligent, standalone data conversion package. netCONVERT's combination of advanced file translation capabilities, cross-platform interoperability, and Java-based GUI, adds up to a superior, full-featured data conversion solution.


With netCONVERT, application data files can be ported with ease, readily converting them for use by ported COBOL applications. Data can also be translated to a "comma separated values (CSV)" format acceptable to most third party applications and advanced DBMSs. Data can travel in any direction, with the conversion process running on the most convenient platform for application batch streams. The conversions may be complex, and include any of the following: one mainframe format to a different mainframe format; mainframe format to workstation or desktop format; workstation or desktop format to mainframe format. 

netCONVERT has two basic components: a filter program to access and translate data; and, a graphical data mapping and conversion tool. The filter program works in all environments, and can be used with Micro Focus COBOL pipes to allow COBOL programs to use the output of conversions directly. The graphical tool is available for all platforms where a GUI is commonly available and Java is supported. netCONVERT's GUI can import COBOL File Descriptions ("FD"), preview data mappings, import test data, preview converted data, specify remapping associated with REDEFINEs, specify conversion parameters, run test conversions, and execute complete conversions. 

netCONVERT uses the COBOL FD as the basis for data mapping. If an FD exists for an input file, netCONVERT can use that file for the record map. If one does not exist, netCONVERT's graphical mapping tool or the mapping syntax of the filter program can be used. Other data mappings could also be converted to a COBOL FD. In fact, by adding a few date expansions to your COBOL FD as COBOL comments, netCONVERT will automatically expand pre-Year 2000 date fields as required. For conversion between platform types, netCONVERT includes an ASCII/EBCDIC translation table that produces reversible results. Users can also define specialized translation tables for data or applications beyond the standard table.

Input may be from disk files, tape files, or other sources piped to stdin. Output may be directed to disk, tape, or piped to other sources through stdout. Tape I/O includes support for both IBM and ANSI standard labeled tapes as well as for unlabelled tapes. A volume id, device name and file position can be specified for both input and output. A file disposition and expiration date can also be specified on output.

Regardless of the input or output source, netCONVERT users can specify record length, blocksize, and record format. Supported mainframe formats include Fixed: F, FB, FBS; and, Variable: V, VB, VBS. Workstation/desktop record formats supported include Fixed, Variable, Segmented (variable, but with a segment byte as in ANSI standard), Text (line-feed delimited), NT (carriage-return/line-feed delimited), Micro Focus COBOL special variable format, and Fortran CSV.


netCONVERT is available for the major mainframe (VM, MVS, OS/390) and UNIX (HP 9000, IBM RS/6000, Sun Solaris) platforms. netCONVERT's graphical component is not currently available for the mainframe. HP/9000 systems require HP-UX 10.10 or later for Java support. IBM RS/6000 systems require AIX 4.1.3 or later for Java support.