[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Invalid file specification

From   "Hendri Adriaens" <>
To   <>
Subject   RE: st: Invalid file specification
Date   Sun, 8 Jul 2007 22:16:28 +0200

Hi David,
> I had provided a rather detailed answer yesterday which seems to have
> gone off into Statalist limbo as I don't see it in the Archive.  One
> of my suggestions to address the need to have both bestand and zk
> defined would be as follows:
> local fnames parsed_cfd.txt parsed_qds.txt parsed_vfd.txt 
> parsed_tsa.txt
> local sub cfd qds vfd tsa
> forvalues x=1/4 {
>     clear
>     /* Inlezen */
>     local bestand :word `x' of `fnames'
>     local zk :word `x' of `sub'
>     insheet a b c d e using `"`bestand'"', tab
>     /* other commands */
>     }
> The loop simply scrolls through the values of the parallel lists.
> From your comments above, although the zk appears as a substring of
> bestand in the example, it may not always be so.

Thank you, that is a very nice suggestion. I will implement it.

> As a programming tip - you don't need the = sign to assign a string to
> a local macro, in fact, it limits the length of the string you can
> assign.  A string expression can have only 244 characters (-local
> bestand = "..."- is a string expression) versus a string macro limit
> of 67,784 characters (1,081,511 in SE and MP).  In your example there
> should be no problem, but if you were to assign a long macrolist to a
> macro, you might get bitten.

Ok, thanks for the warning. I will remember it.

> Also, if you don't want all the output generated by a -set trace on-
> for debugging you can surround just the suspect statements.
> Alternatively, if you want to check that a macro assignment went as
> intended, dropping in a -macro dir- is helpful.

Thanks, will keep that in mind!

> As for the discussion above - I agree with Hendri that it is often
> valuable to discuss the debugging process, error messages and their
> limits, and the processes one must go through in order to diagnose and
> correct one's error(s).  We can all learn from these discussions.

I agree. Thanks again David for so many helpful suggestions!

Best regards,

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index