-charlist- update available from SSC [was: Re: st: charlist syntax error]

[Nick Cox <njcoxstata@gmail.com>]

Thanks to Kit Baum, a revised -charlist- is now available from SSC.
Stata 9 is required. Use -ssc desc charlist- for an executive summary
and -ssc inst charlist- to install if interested.

Nick
njcoxstata@gmail.com

On 28 February 2014 13:50, Nick Cox <njcoxstata@gmail.com> wrote:

> We've resolved the problem. I'll send a revised -charlist- to Kit Baum
> for SSC, but as advertised he's away for a while.
>
> Thanks to Martyn for his bug report, which prompted a cleaner
> rewriting of -charlist-.

On 28 February 2014 13:14, Nick Cox <njcoxstata@gmail.com> wrote:

>> From the output the culprit is evidently the backtick or left
>> quotation mark (char(96)).
>>
>> The occurrence of the character `  (char(96) was problematic for
>> -charlist- given its uses for local macro references.
>>
>> I found in my files a -charlist- 1.2.0 from 2008 which dealt with this
>> problem without crashing, but not well. Evidently I never sent that to
>> SSC.
>>
>> I'll send a -charlist- 1.2.1 to Martyn directly to see if that helps
>> his problem

On 28 February 2014 12:48, Martyn Sherriff <statams48@gmail.com> wrote:

>>> I regularly use charlist (SSC NJC 1.1.0 17 Dec 2002 ) to screen data
>>> for strings in supposedly numeric data and I am getting a syntax error
>>> with the following variable:
>>> .d x35
>>> x35             str3    %9s
>>>
>>> . charlist x35
>>> invalid syntax
>>> r(198);
>>>
>>> I can detect the errant characters:
>>> . list x35 if missing(real(x35))
>>>
>>>      +-----+
>>>       | x35 |
>>>       |-----   |
>>>  53. | 29n |
>>>  63. |  6`   |
>>>       +-----+
>>>
>>> I would be grateful if somebody could suggest what I am doing
>>> incorrectly with charlist (it works with other variables in the
>>> dataset).
>>>
>>> Stata 13.
*
*   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/