Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Re: grc1leg


From   Chris Witte <eljefespeaks@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Re: grc1leg
Date   Tue, 2 Dec 2008 09:21:38 -0800 (PST)

Mike wrote "The path name error is a red herring.  The actual problem with your example is that you -name()- the graphs, but then do not use the named graphs in your -graph combine- or -grc1leg- commands."  That is the incorrect pathway reference that I meant to fix before I hit the send button.  I never said that I was too busy to fix the problem.  I said that I was distracted and missed correcting the problem.  Again, I screwed up and I apologize.  

The lack of an error message regarding the -ysize- option when using -grc1leg- was relevant both to my problem I was describing, and your mention of needing to "Say exactly what you typed and exactly what Stata typed (or did) in response."  It would have been nice if I had received an error message saying something along the lines of "ysize option not allowed with grc1leg".  Instead, the code produced a graph with the default dimensions without any error warnings (ignoring the -ysize- option)... a bug that should be fixed, but easily circumvented by using -graph display-, as Gary suggested.

I assumed that initially, Martin fixed my pathway mistake, ran the code, didn't see an error message and didn't catch that the graph was not altered from the default aspect ratio, looked at the -grc1leg- help file, and then replied with his original response.  This further emphasizes the need for a warning message, and mention in the help file that you can't use ysize or xsize options (and maybe others).

All's well since it ended well,
Chris



----- Original Message ----
From: Michael Hanson <mshanson@mac.com>
To: statalist@hsphsun2.harvard.edu
Sent: Tuesday, December 2, 2008 10:46:05 AM
Subject: Re: st: Re: grc1leg

Chris:

I did read your response to Martin.  In fact, you'll note I quoted it below.

But I suspect you do not understand my response.  The path name error is a red herring.  The actual problem with your example is that you -name()- the graphs, but then do not use the named graphs in your -graph combine- or -grc1leg- commands.  Instead, you use saved graphs, but your graph-generating commands do not contain the -saving()- option.  In other words, even with "correct" paths (or no paths at all), the example you provided is not self-contained, and cannot possibly produce the problem you cite.

Why does this matter?  For starters, it implies that you have executed other code outside of your example (such as previously creating and saving versions of your graphs) that may be critical to understanding why you are getting the undesired results you report, error message or not.  As such it becomes awfully difficult for anyone else to accurately troubleshoot your problem.  Hence my quoted reminder from the Statalist FAQ.

Your follow-up message is puzzling to me, as the path (noted above) and the lack of an error message are not relevant to the issues I raised regarding your original question and the sample code you sent to the list.  Moreover, to suggest that you're too busy in your office to submit a complete example or to correct one that you know is wrong -- to suggest that your time is more valuable than all those who voluntarily provide suggestions on this list -- strikes me as an odd attitude to have when you are the one seeking help....

I'm glad you were able to get an answer to your specific question this time.  Perhaps you will recognize as constructive friendly suggestions that increase the likelihood of others taking their time to answer your future requests.

-- Mike


On Dec 2, 2008, at 9:33 AM, Chris Witte wrote:

> Mike,
> I don't think you read my response to Martin.  I already appologized for the pathway screw up that Martin pointed out.  It was my original intent to remove the pathways from the code, but office distractions led me to overlook those lines in my example.  I also said that Stata did not return any kind of error message, and that it just ignores my -ysize- option.
> 
> Gary,
> Thanks for your constructive advice.  I'll try to get -graph display- to work for me.
> 
> 
> ----- Original Message ----
> From: Michael Hanson <mshanson@mac.com>
> To: statalist@hsphsun2.harvard.edu
> Sent: Monday, December 1, 2008 6:29:32 PM
> Subject: Re: st: Re: grc1leg
> 
> Unless I misunderstand your example (which the exchange below suggests it may fail the Statalist FAQ dictum: "Say exactly what you typed and exactly what Stata typed (or did) in response. N.B. exactly!"), it looks to me that you are not using your named graphs in your -graph combine- and -grc1leg- commands, but rather (old?) versions you have saved to disk.  If so, this could yield unanticipated results.
> 
> (As an aside:  why hard-code Windows-type paths into such a simple example?  It is completely unnecessary, even if you are -saving()- the graphs.  When people do this, they are implicitly saying they don't want any help from the many folks on Statalist who use a non-MS operating system.)
> 
> -- Mike
> 
> On Dec 1, 2008, at 6:21 PM, Chris Witte wrote:
> 
>> hmm, when I run the -gr combine- I get a very tall skinny combined graph, but the -grc1leg- does not do that for me.  It is like the ysize option is ignored, and doesn't return an error or anything.
>> 
>> sorry about the path mess...
>> 
>> 
>> 
>> ----- Original Message ----
>> From: Martin Weiss <martin.weiss1@gmx.de>
>> To: statalist@hsphsun2.harvard.edu
>> Sent: Monday, December 1, 2008 5:05:35 PM
>> Subject: st: Re: grc1leg
>> 
>> Seems to me that -grc1leg- takes all the options of the official -gr combine-. Your syntax runs without error for me (apart from the obvious path issue), with
>> 
>> . which grc1leg
>> c:\ado\plus\g\grc1leg.ado
>> *! version 1.0.4  08sep2005
>> 
>> 
>> 
>> 
>> HTH
>> Martin
>> _______________________
>> ----- Original Message ----- From: "Chris Witte" <eljefespeaks@yahoo.com>
>> To: <statalist@hsphsun2.harvard.edu>
>> Sent: Monday, December 01, 2008 11:29 PM
>> Subject: st: grc1leg
>> 
>> 
>>> I've ran into a problem using -grc1leg-, which puts a single legend on a combined graph. I am unable to use the -xsize- and -ysize- region options, which are available when using -graph combine-. here's an example:
>>> 
>>> sysuse auto, clear
>>> scatter price mpg, name(g1)
>>> scatter weight length, name(g2)
>>> graph combine "C:\_DATA\g1" "C:\_DATA\g2", ysize(10)
>>> grc1leg "C:\_DATA\g1" "C:\_DATA\g2", ysize(10)
>>> 
>>> can anyone fix this?
>>> 
*
*  For searches and help try:
*  http://www.stata.com/help.cgi?searchhttp://www.stata.com/support/statalist/faqhttp://www.ats.ucla.edu/stat/stata/



      

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index