Susan Lynch wrote: > Tony, there are a number of things that a source code analyzer can do > that would be very helpful if you found yourself coming into a > company as a consultant or as a new employee and they had no > technical documentation:
Without knowing more about what Herve wanted, I'm making this up as I go. It seems to me that anyone could do this, even making it open source, so I'll flesh it out just a little more. Regarding paragraphs, etc, one of the things I was alluding to was that separate items should be maintained which contain XML related to code which cannot itself contain the metadata. For example, Susan, it would be nice to have Where Used data in your HR Partner DesignBais front-end, so that you know which menus call to specific forms, which fields are used on which forms, which forms run specific selection processes, which programs use PROCESS.STACK to invoke other forms, etc. In DesignBais or with other utilities that allow GUI queries it would be helpful to know where various dict items are used. It may not be practical to put the XML in the dict item (extended dict items, paragraphs, etc) and these environments don't have a special/common place for metadata, so a custom file would need to be created that defines where dict items are used, how they're used, etc. As always, Garbage In, Garbage Out, and unless metadata is maintained the reports one derives from the data are going to be inaccurate. Manual procedures must be implemented which require update of the data, or version control software can prompt developers to ensure they have updated docs before checking in changes. T ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
