While it is possible to trace dofiles, I have never been able to see
proper macro expansion the way it occurs with adofiles.
Imagine the following dofile "example.do" with a single line:
local xmas `"Days until Christmas: `=td(25 dec 2010) - td(`c(current_date)')'"'
if you -set trace on- and -do example.do- (or using Ben's program -tr
1:do example.do- )
you will see simply:
. local xmas `"Days until Christmas: `=td(25 dec 2010) -
td(`c(current_date)')'"'
Can you tell me the value that is in the local macro xmas ??
If you use my program -dodebug using example.do- you get the following:
- local xmas `"Days until Christmas: `=td(25 dec 2010) -
td(`c(current_date)')'"'
= local xmas `"Days until Christmas: 319"'
Here you clearly see the value stored in the local. I think this is
an important advantage over other methods when debugging.
Thanks to Eric for directing me to -appendfile- and I used Julian
Reif's method to sandwich the dofile code into the adofile wrapper
resulting in this new version:
x------------- begin code -------------x
program define dodebug
version 9.0
*! version 1.0.1 2010.02.09
*! Run a do-file with trace
*! by David C. Elliott
syntax using/ [, Depth(integer 1) Hilite(string)]
// cleanup if program has been run previously
capture program drop __dotemp
capture erase `"`c(sysdir_personal)'__dotemp.ado"'
file open top using `"`c(sysdir_personal)'__dotemp.ado"', write append
file write top "program define __dotemp" _n ///
"version 9" _n ///
`"noisily di "{res:{hline 10} begin debug {hline}}""' _n
file open middle using "`using'", read text
file read middle line
while r(eof)==0 {
file write top `"`macval(line)'"' _n
file read middle line
}
file write top `"noisily di "{res:{hline 10} end debug {hline}}""' _n ///
"end" _n
file close _all
set tracedepth `depth'
set tracehilite "`hilite'"
set trace on
__dotemp
set trace off
* viewsource __dotemp.ado // Uncomment to look at temporary adofile
erase `"`c(sysdir_personal)'__dotemp.ado"'
end
x------------- end code -------------x
DC Elliott
*
* 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/
