On Mar 27, 2009, at 4:22 PM, Stephen Fisher wrote:
Can we work unicode (UTF-8) support into these new string functions?
What does "Unicode (UTF-8) support" mean in this context? If a string
passed as the initial value, a string passed to ep_strbuf_append(), a
format string passed to ep_strbuf_append_vprintf() or
ep_strbuf_append_printf(), or a string corresponding to a %s format
item in ep_strbuf_append_vprintf() or ep_strbuf_append_printf() is
UTF-8 encoded, the resulting string will be UTF-8 encoded.
The only issue I'd see with UTF-8 support would be if we added an
option to set a maximum length for an ep_strbuf, so that somebody
could specify that they don't want a string generated from a list of
items to grow so big as to fill up the entire Info column or protocol
tree item representation (e.g., just stop adding items and just put
", ..." at the end if it gets too big), in which case we'd want to
make sure that we don't add some but not all of the bytes of a UTF-8
string - and so that the length is measured, perhaps, in characters
rather than bytes (not all characters are the same width, but I'm not
sure we want to find out how deep that rabbit hole goes).