mis@Seiden.com (Mark Seiden) (04/09/91)
i've just gotten a vast pile of unfamiliar sql dropped on me which i have to somehow assimilate. this is around 40k lines of sql, split into almost 600 files (and growing...), and a equal number of lines of sqr (sql solutions' report generator) for generating reports. even better, there's a pile of sql embedded in c++, around 1200 sql statements stuck in there somewhere. while this is probably larger than most systems folks have to deal with, this can't be a new problem. any suggestions for tools for better understanding (or for that matter, performance analysis or debugging) of sql and embedded sql? specifically, things like the sql equivalent of c TAGS or anything that will help me understand what calls what, where things are defined, etc. anyone interested in profiling/execution counting for sql and stored procedures? i've been thinking about some scripts that insert (and noop out) counting and timing code in stored procedures. has anyone yet figured out what the minimal useful code to do that would be? if you want to *understand* what is going on in an update (e.g.) there are so many things to look for (and in sybase, a trigger or a rule could be involved...) another small frustration is not having an sql-mode in emacs. has anyone already written it for any flavor of sql? i'm planning to summarize (if you'd prefer to send email) or if you have useful code to share, please post a followup. thanks for your interest and communication. mark seiden, mis@seiden.com, 1-(203) 329 2722 (voice), 1-(203) 322 1566 (fax)