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: Need help with regular expression
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: Need help with regular expression
Date
Sun, 17 Feb 2013 14:08:42 +0000
The regular expression being sought is different.
http://www.stata.com/support/faqs/data-management/regular-expressions/
explains Stata's syntax here. $ has a special interpretation as the
end of a string.
By the way an entire line such as
di regexm("show us", " us$")
is a command, not a function. Commands can include function calls;
indeed that is the only way that Stata functions can be used.
Nick
On Sun, Feb 17, 2013 at 1:22 PM, Michael Stewart
<[email protected]> wrote:
> Hi Nick and Alison,
> One quick question.(sorry, I am a novice to regular expressions)
> The two functions di regexm("show us", " us$") & di regexm("show us",
> " us&") have different results.
> what is the difference between these functions ??
> Thank you very much .
> Sincerely
> Mike
>
> On Sun, Feb 17, 2013 at 5:03 AM, Nick Cox <[email protected]> wrote:
>> Alison's suggestion is surely right. If your condition is not
>> restrictive enough, you need to make it more restrictive.
>>
>> If it is a separate word, it occurs surrounded by spaces or at the
>> beginning or end with one space after or before. I tend to build up
>> syntax using -display- and little examples.
>>
>> . di regexm("us marines", "^us ")
>> 1
>>
>> . di regexm("show us", " us$")
>> 1
>>
>> . di regexm("invited us to dinner", " us ")
>> 1
>>
>> You could also use -strpos()-
>>
>> ... if strpos(var, " us ") | strpos(var, "us ") == 1 | strpos(var, " us") == -3
>>
>> Note that -regexm()- is a function, not a command. in Stata functions
>> and commands are quite separate in definition and documentation.
>> -strpos()- is another function. Fixing _function_ firmly in mind as
>> the key term leads to consultation of the documentation on functions,
>> which as here can suggest alternatives.
>>
>> There are also solutions with -word()- but they all seem to entail a loop.
>>
>> Nick
>>
>> On Sun, Feb 17, 2013 at 8:04 AM, Alison Connor
>>
>>> I am sure there is a much better way to do this, but until someone who
>>> actually knows responds, could you try something like:
>>>
>>> regexm(var, " us ") | regexm(var, " us&") | regexm(var, "^us ") ?
>>
>> On Sun, Feb 17, 2013 at 1:00 PM, Michael Stewart
>>
>>>> I am trying to use regexm command (regexm(var,"us"))to see if the
>>>> values of variable have a particular string "us" like in "us thyroid"
>>>> .Unfortunately , it is picking up "us" from other string values like
>>>> "venous".If there a way to formulate command so that it can pick up
>>>> only values where "us" is a separate word and not part of word like
>>>> venous.
*
* 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/