On Dec 1, 2009, at 12:17 PM, Ralf Junker wrote: > > always made me believe that all options to omit features are > supported. > Now I am not sure which ones I can really trust. > > Could you clear up my confusion? >
There are several dozen individual compile-time options. We obviously cannot test every combination of compile-time option since there is an exponential explosion in the number of test cases. And so, our strategy is to test only those compile-time configurations that are in common use or which are used by companies that provide material support to SQLite. So, for example, SQLITE_SECURE_DELETE was unsupported for a long time. Then we discovered that SQLITE_SECURE_DELETE is used in Firefox. So now we are very careful to test SQLITE_SECURE_DELETE before each release. Actually, we know exactly which compile-time options Mozilla uses, and we test that exact combination of compile- time options. A different set of compile-time options are used by Symbian/Nokia, and by Bloomberg, and so forth. We test all of these configurations carefully. But those are only a handful of configurations out of the billions and billions of possible ways of configuring SQLite. As it happens, we are not aware of any major product using SQLITE_OMIT_SUBQUERY. Nor do any of our sponsors use or desire that option, as far as we are aware. Nor is it an especially useful option, in most peoples view. And so testing and maintaining SQLITE_OMIT_SUBQUERY is pretty low on our list of priorities. I'm sorry to disappoint, but the fact is that we do have to prioritize things. The SQLite developers do not have any idle cycles at the moment. D. Richard Hipp [email protected] _______________________________________________ sqlite-users mailing list [email protected] http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
