Peter Eisentraut <peter_e@gmx.net> writes:
> On mån, 2011-09-26 at 10:08 -0400, Tom Lane wrote:
>> No, it's a hundred times worse than that, because in collations other
>> than C there typically *is* no total order. The collation behavior of
>> many characters is context-sensitive, thanks to the multi-pass behavior
>> of typical "dictionary" algorithms.
> Well, there is a total order of all strings, but it's not consistent
> under string concatenation.
> But there is a "largest character". If the collation implementation
> uses four weights (the typical case), the largest character is the one
> that maps to <FFFF> <FFFF> <FFFF> <FFFF>. If you appended that
> character to a string, you would get a larger string. (Unless there are
> French backwards levels or other funny things in place, perhaps.)
But the problem is not "make a string greater than this given one".
It is "make a string greater than any string that begins with this
given one". As an example, suppose we are given "xyz" where "z" is
the last letter in the collation. We can probably find characters
such as "~" that are greater than "z", but no string x-y-nonletter
is going to be considered greater than x-y-z-z by a dictionary
sorting method. This is why make_greater_string has to be prepared
to give up and go increment some character before the last one:
the only way to succeed for this input is to construct "xz".
regards, tom lane
