Re: [Emacs-diffs] emacs-25 f8208b6: Document the user-level features of the Xref package

[Dmitry Gutov <dgutov@yandex.ru>]

On 01/18/2016 08:31 PM, Eli Zaretskii wrote:
> I tried to improve the manual according to your comments, please have

Thank you, Eli. I think it's good now, but see a couple of nits below.

> a look.  (I suggest to read the Info manual, not look at the diffs.)

I did that too, after some usual (for me) flailing around, to find the 
needed node in the manual. The diff turned out to be more useful, 
though, because I was able to quickly see what was removed, and what was 
changed.

The promised nits:

 > If there is a tags table loaded, this command can use it to
+generate completion candidates more intelligently.

That implies that we have some "dumber" completion sources than etags in 
Emacs. I don't think we do, currently. tags-completion-at-point-function 
is the default value of completion-at-point-functions, and we use it as 
the last resort if the major mode (or any minor mode) don't provide any 
specialized completion functions. Then, if no tags table is loaded, we 
don't provide any completions at all, not even stupid ones. I'd suggest 
to simply remove "more intelligently".

 > +  A @dfn{tag} is a synonym for identifier reference.  @xref{Xref}.

Maybe that's technically true, but as employed by Emacs usually, tags 
are elements of TAGS file, generated by the 'etags' program, and they 
only reference identifier definitions. This is the reason we had to 
resort to other means to implement xref-find-references anyway.

And in CEDET, there is a notion of "current tag", which is the function 
definition around point. Not variable reference or the method called at 
point, as one might expect to be the default input for `semantic-symref'.

Call it an "identifier definition reference", maybe? Or "identifier 
definition", or "definition reference".