About strings, I firmly believe that C's zero-terminated strings should be avoided and only used when necessary to communicate with existing code. They're really not simple and fast. You can't take a null-terminated string and extract a null-terminated sub-string without modifying the original - which is very often needed during various kinds of parsing. Also, null-terminated strings are unable to represent the zero byte, so in general, for example, you can't use them to store the contents of an arbitrary file, and if you do use them for purposes where null bytes can appear, you have to be careful and handle errors where nulls would implicitly truncate your string.
Just use (pointer, length) strings instead. It'll save you a whole bunch of trouble you may not see coming.
[1] https://code.google.com/p/badvpn/source/browse/#svn%2Ftrunk%... (CAvl_)
[2] https://code.google.com/p/badvpn/source/browse/#svn%2Ftrunk%... (cavl_test_)
Most of the time, the amount of memory you'll need to do something is predictable. When dynamic arrays are always at hand, you lose the habit and end up always reaching for one.
Currently I'm using this code just for a string buffers and other arrays where the individual elements are stored in-place in the array.
CPP macros, entirely. No, thanks.
https://github.com/pmj/genccont
Anyway, these are largely intrusive and don't do reference counting, but that can be seen as an advantage or disadvantage, depending on the situation. (I use these heavily in kernel code) The hash tables (chaining and open addressed linear probing) use function pointers to be type-generic, which might not be to everyone's taste either.
I don't really understand the fear of macros. void pointers are much scarier than macros, since they open you up to the threat of bugs going undetected by the type system.
You're not by any chance maintaining one of those much-maligned "driver portability" layers, are you?
If you look at the slist, slist_queue and dlist, you'll find there are typed macros (via a container-of macro) that call down to the generic functions. I might try to extend this to macros that generate a full set of shim inline function for a particular type of list element, if it turns out to be less error prone. You won't see a void pointer being passed around on any of these.
Where function pointers come into play (binary tree, hash tables) I'll admit it's maybe not quite so clear cut as the indirection messes with branch predictors, etc. and the callbacks all take void pointer arguments. Maybe for those containers it would be better to have generator macros that produce function prototypes and implementations.
And no, I'm not doing anything particularly focused on driver portability. I mostly write OS X kexts, and the built-in data structures in xnu aren't terribly helpful. (the OSContainer family uses blocking memory allocations, can only hold refcounted objects and the map and set use linear search) Some kernel code I write is designed to be portable, and I use these data structures for that code, but it's pretty much just a standard factored design - general core and platform specific setup and API use. Nothing with any pretence of being a general solution.
With regard to OBString: don't. Really. C strings are very flexible and powerful with just what's in the standard library. You don't need charAtStringIndex, for example: you have rindex and index. You don't need splitString; you have strtok_r. You don't need findSubstring; you have strstr.
If you think you need regexes, consider fnmatch, a libc function which matches shell wildcard patterns. It's always there, it's simple, and it's fast. Or you could bring in something like PCRE. In general, if you need full regexes in a small program, it might be worth rethinking whether your program should be in C anyway.
For some reason, I don't find myself using resizable array utility code much in C. It's pretty easy to do it yourself with realloc-- that's really one of the first things they should teach people learning practical C. Perhaps a small, minimal set of macros could wrap the reallocs and make it a little bit nicer.