st: compressed data and named pipes on linux

[James Sams <sams.james@gmail.com>]

I wish to compress my stata datasets and use them via a named pipe. I was 
hoping that the gz* tools would do this, but it appears that they make use of 
a temporary file and write the entire compressed file out then read it in, 
slowing things down pretty significantly with the extra hard drive read/write 
cycle. My first instinct was to use a named pipe, and it seems like this should 
work according to past messages to the stata list and a stata FAQ. However, 
with stata 12 mp, I am getting a core dump. Can someone tell me where I am 
going wrong?

unzip.sh:

#!/bin/bash

gzip_file="$1"
pipe_name="$2"

if [ -e "$pipe_name" ]; then
    rm "$pipe_name"
fi
mknod "$pipe_name" p
zcat "$1" > "$pipe_name" &


usage in stata:

local gzip_file 2667.dta.gz                                      
tempfile pipe_file                                               
                                                                 
shell ./unzip.sh `gzip_file' `pipe_file' >& /dev/null < /dev/null
use `pipe_file'                                                  
shell rm `pipe_file'

Doing this manually also results in a core dump:
$ mkfifo tempfile
$ zcat 2667.dta.gz > tempfile &
$ stata -q use tempfile.
Segmentation fault (core dumped)

FWIW, Doing this works just fine:
$ mkfifo tempfile
$ zcat 2667.dta.gz > tempfile &
$ cat tempfile > 2667.dta && stata -q use 2667.dta
$ stata -q use 2667.dta



-- 
James Sams
sams.james@gmail.com
*
*   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/