st: auto-create dictionary file

["Buzz Burhans" <buzzb3@earthlink.net>]

I have a large comma delimited text file I need to read into Stata. I need
to do this repeatedly with similarly structured files I previously split out
from a large master file.  The subfiles I need to import are nonetheless
large in terms of both number of observations and number of variables. If I
-insheet- the file I get a portion of it and then I run out of memory.
-insheet- does not allow an -in- option, so I can't restrict the amount it
tries to import, and I can't specify the variable types to conserve memory
(use of -compress- after the import decreases the file size tremendously).
I'd like to use -infile- to take advantage of being able to use  -in-  and
to specify the new variable types.

Is there any way to produce a Stata .dct dictionary file automatically, so
that I could then modify it for use with -infile- or -infix-, instead of
having to create a dictionary from scratch?  Using -findit- I found a very
old STB program (STB9) called -creatdct- by Bill Gould which automatically
creates Stata .dct dictionaries. That program seems to have limitations in
terms of file size, and I could not get it to work. It appears that that
program was a precursor to later functionality built into the -insheet- and
-infile- commands, but I am not aware of a way to get a dictionary file
after running  those commands.

I'm guessing Stat Transfer would work easily here, but this is not a task I
have to do often enough to justify owning it, so I'd like to come up with an
alternative "manual" method.


Thanks for any insights.

Buzz
 

Buzz Burhans, Ph.D.

Dairy-Tech Group
So. Albany, VT / Twin Falls ID

Cell: 208-320-0829
ID Fax: 208-735-1289
VT Fax: 802-755-6842

Email: buzzb3@earthlink.net



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/