Re: st: somersd error message

["William Gould, StataCorp LP" <wgould@stata.com>]

Roger Newson wrote, 

RN> This is very curious INDEED. Do you get exactly the same error
RN> message as before, or another one? Because -lsomersd.mlib-
RN> should DEFINITELY include the -tidottree()- function. If it
RN> doesn't, then it is not the correct -lsomersd.mlib-.

when Barry Quinn <brotherbarry@me.com> replied most recently about
-mata query- showing that lsomersd was included among lsomersd.mlib and
yet he is still getting the error message that tidottree() is not being
found.

There are some Mata commands that might help debug the issue. 
They are 

           : mata mlib query 

           : mata which tidottree()

           : mata describe using lsomersd

I recommend running them, and in that order specified.

-mata mlib query- merely repeats the list of libriries that Mata
searches; the information provided should be the same as that provided
by -query-, but before we go looking for tidottree(), it's useful to 
verify that lsomesd really is in the libaries-to-be searched list.

If it is not, type -mata mlib index- and then issue the 
-mata mlib query- again. 

-mata which tiddottree()- will cause Mata to report where 
tidottree() is, if it is anywhere.  If lsomersd is in the search list, 
and if function tiddottree() exists in it, the result will be 

        : mata which tiddottree()
          tiddottree():  lsomersd

Alternatively, the result might be

        : mata which tiddottree()
        function tiddottree() not found
        r(111);

Assuming the function is not found, it would be intersting to know 
what is in the library lsomersd:

        : mata describe using lsomersd
          <list of contents appears>

-- Bill
wgould@stata.com
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/