Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: RE: Features for Stata 14
From
Robert Picard <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: RE: Features for Stata 14
Date
Tue, 10 Sep 2013 12:57:22 -0400
And apologies for repeating myself but -project- (SSC) can return both
the name of the current do-file AND the path to the directory that
contains the current do-file. When do-files are nested, the concept of
the currently running do-file is ambiguous as there can be more than
one do-file running at the same time. To correctly track the path to
the "top-most" do-file, you have to maintain a LIFO stack and update
it as a new do-file starts and when it exits. That's what -project-
does (and much more).
Of course StataCorp could track each -do- and -run- call and implement
the same functionality.
On Tue, Sep 10, 2013 at 12:09 PM, Eric A. Booth <[email protected]> wrote:
> <>
>
> An important note on my previous post --
> I took out a preserve/restore in the last part of my .ado file before
> I pasted it to the list b/c I was looking at what it was doing, but
> that can destroy data in memory if you used it later in your do-file
> (which I caution against anyways), so you can put the preserve/restore
> back in by using the ado below instead.
>
>
> Also, the reason I use -file write- to save the one command -#review
> 4- and then run it separately, is that when I tried to just include
> #review in the program and log it's output it was not logging
> properly. So a quick solution for me at the time was to save it in a
> separate .do file and then run that file and log the output. There's
> likely a faster/better way to do all this , but it served it's purpose
> for me at the time.
>
>
> *********************!updated dofilename.ado
> cap program drop dofilename
> program def dofilename
> syntax
>
> *-write #review to a do file**
> cap file close j1
> tempfile j2
> file open j1 using `j2'.do, replace write
> file write j1 `" #review 4 "' _n
> file close j1
>
> *-run j2 to get the #review output
> tempfile t
> cap log close _t
> qui log using `t', replace text name(_t)
> do `"`j2'.do"', nostop
> log close _t
>
>
> *-parse the j2 file output to get the .do file name
>
> preserve
> insheet using `t', clear nonames
> tempvar i
> g `i' = 1 if index(v1, ".do")
> cap keep if `i'==1
> cap keep in 1
> if _N!=1 di as err "Make sure your current dofile is saved, and run
> -dofilename- as first command in file!"
> global dofilename `"`=v1[1]'"'
> restore
>
> end
> ***************************! END
>
> - Eric
>
> On Tue, Sep 10, 2013 at 10:52 AM, Eric A. Booth <[email protected]> wrote:
>> <>
>> On Mon, Sep 9, 2013 at 9:55 PM, Timothy Mak <[email protected]> wrote:
>>
>>> 1. A system macro that gives the directory of the executing do-file. This feature has been asked for in a number of recent posts, and it would make organizing do-files so much easier. see e.g. http://www.stata.com/statalist/archive/2013-07/msg00667.html
>>
>>
>> Hi Timothy - Though probably rarely needed, I agree that a system
>> macro containing the current do file name would be useful.
>> FWIW, I had a situation where I needed this, so I came up with the
>> kludge solution below to store the do file name. I havent fully error
>> checked it, but basically it used the #review interactive command (see
>> -help #review-) to display the do-file name you just ran and then
>> store that output in a global macro for later use. I've pasted this
>> program below JIC it's useful for others.
>>
>>
>> So, you need to save the following program as an .ado file, type
>> discard, then make sure the do-file you are running is saved
>> (otherwise it is just given a tempfile name), and that this program
>> runs (just include the command -dofilename-) as the first line in your
>> do-file. ((Really, you should be able to put the -dofilename-
>> anywhere in the file since the last command you ran in the 'command
>> window' for Stata was to -do- that dofile ...but I put it at the top
>> just to be sure)) Those are some pretty inconvenient
>> limitations/requirement for some, but it worked for what I needed.
>>
>>
>>
>> *********************! save below as dofilename.ado
>> cap program drop dofilename
>> program def dofilename
>> syntax
>>
>> *-write #review to a do file**
>> cap file close j1
>> tempfile j2
>> file open j1 using `j2'.do, replace write
>> file write j1 `" #review 4 "' _n
>> file close j1
>>
>> *-run j2 to get the #review output
>> tempfile t
>> cap log close _t
>> qui log using `t', replace text name(_t)
>> do `"`j2'.do"', nostop
>> log close _t
>>
>> *-parse the j2 file output to get the .do file name
>> insheet using `t', clear nonames
>> tempvar i
>> g `i' = 1 if index(v1, ".do")
>> cap keep if `i'==1
>> cap keep in 1
>> if _N!=1 di as err "Make sure your current dofile is saved, and run
>> -dofilename- as first command in file!"
>> global dofilename `"`=v1[1]'"'
>> end
>>
>> ****************************! END
>>
>>
>> then run something like this in your do-file:
>>
>>
>> *********************! sample do-file
>> dofilename
>>
>> sysuse auto, clear
>> desc, sh
>> ...
>> di `"${dofilename}"'
>> *********************! sample do-file
>>
>> HTH,
>>
>> Eric
> *
> * 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/
*
* 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/