Being Russian, I'm concerned with non-ASCII characters and their support in Lisp implementations. CLISP, SBCL, ACL and other implementations support Unicode, but provide slightly different APIs to convert strings from and to octet arrays; Araneida, Portableaserve, CLSQL, CLORB and other applications use these APIs via slightly different Compatibility Layers or introduce ad-hoc solutions that often do not work for non-ASCII text.
My suggestion is to make a Text library to unify this. To tell the truth, such a simple library already lives on my hard drive, but I have to hack every new release of aforementioned great software.
Maybe such a library already exists? kmrcl has functions usb8-array-to-string and string-to-usb8-array,
but they'll work incorrectly for non-ASCII strings (because they assume that char-code will always fit into a 8-bit byte), and even CLSQL doesn't use them, resorting to "#ifdef"s...
Update: It seems that Arnesi provides such functionality, along with other great utilities. I urge everybody to mend their evil ways and convert to Arnesi :-)
CLiki pages can be edited by anyone at any time. Imagine a fearsomely comprehensive disclaimer of liability. Now fear, comprehensively