Yes, I understand that stuff is under development, and that is a period during which a) things are not always dependable and b) things are known by the developer and no one else.  So eventually this primitive or the more high-level tool discussed below.
 
Paul Thompson

From: Joseph Wright <[log in to unmask]>
To: Mailing list for the LaTeX3 project <[log in to unmask]>
Cc: Paul Thompson <[log in to unmask]>
Sent: Sun, January 2, 2011 2:23:11 PM
Subject: Re: LaTeX3 and engines

On 02/01/2011 15:27, Paul Thompson wrote:
> There is really a somewhat deeper issue.  Like many who use LaTeX for more than
> superficial stuff (I wrote a class file newlfm and several sty files, most of
> which are not much used), I can read documentation and figger it out somewhat.
> However, knowing of the EXISTENCE of a tool is really the key, not how it is
> used.  How are people to learn that specific primitives exist?  It's not in the
> LaTeX Companion, which is my starting point.

As Arno has already commented, this is a primitive, so it belongs in the pdfTeX manual and not in anything from LaTeX. I'd also note that it was introduced relatively recently (compared at least to the age of LaTeX2e), so it's not surprising that it's not there.

As an aside, one of the points of the expl3 work is to provide a LaTeX programming language in which everything is documented if it's available. That doesn't mean that all of the primitives are documented in expl3: the other point is that many of them are not really ideal for direct use. \pdfstrcmp is a case in point: the expl3 interface is called \str_if_eq:xxTF (and related functions), as it's not excatly obvious that a string comparison yields a number which can then be used in an \ifnum :-)
-- Joseph Wright