Re: st: getting part of strings

[Daniel Marcelino <dmsilva.br@gmail.com>]

That technique is impressive and make me think about if I could do the
opposite job, namely strip out all alpha characters. However, just
changing to i = a/z not seem to work.

Daniel


On Sat, Mar 26, 2011 at 9:31 PM, Nick Cox <njcoxstata@gmail.com> wrote:
> Another technique that might be helpful is to strip out all the
> numeric characters first, e.g. by
>
> clonevar copy = var1
> qui forval i = 0/9 {
>            replace copy = subinstr(copy, "`i'", "", .)
> }
>
> Nick
> On Sat, Mar 26, 2011 at 8:50 PM, Daniel Marcelino <dmsilva.br@gmail.com> wrote:
>> Thanks for helping, I'll work on the code for required output. I
>> thought Eric Booth's example pretty insightful for my needs. My data
>> is not delimited by dash "-", rather it is by ";". However, the
>> original source has a variable with many things nested, so, I want to
>> split those names, parties, offices and numbers ids into different
>> variables.
>>
>> Best
>> Daniel
>>
>> On Sat, Mar 26, 2011 at 4:42 PM, Eric Booth <ebooth@ppri.tamu.edu> wrote:
>>> <>
>>> Daniel:
>>> I missed the part in your post where you want to capture PB and PP as well.
>>> You could grab these from the var1? that contains this information from my previous example, or another approach entirely is to use the string functions (see -help string_functions-)  subinstr() or strpos() to generate indicators if var1 contains the substrings of interest -- this allows you to skip the -split- or regex* approaches completely if this is what you need from var1:
>>>
>>> ***********************!
>>> clear
>>> inp str200 var1
>>> "155 - VITAL DO REGO FILHO - PB - Senador"
>>> "1111 -  - PP -  - Deputado Federal / 25888 - ATAIDES MENDES PEDROSA -PB - Deputado Estadual"
>>> "1111 -  - PP -  - Deputado Federal / 22333 - EDNALDO PEREIRA DESANTANA - PB - Deputado Estadual"
>>> "151 - JOSE WILSON SANTIAGO - PB - Senador"
>>> "45123 - ANTONIO HERVAZIO BEZERRA CAVALCANTI - PB - Deputado Estadual"
>>> "1212 - DAMIÃO FELICIANO DA SILVA - PB - Deputado Federal"
>>> end
>>>
>>> g DF = 1 if strpos(var1, "Deputado Federal")
>>> g DE = 1 if strpos(var1, "Deputado Estadual")
>>> g S = 1 if strpos(var1, "Senador")
>>> g PP = 1 if strpos(var1, "PP")
>>> g PB = 1 if strpos(var1, "PB")
>>> order D* P* S
>>> ***********************!
>>>
>>> - Eric
>>> __
>>> Eric A. Booth
>>> Public Policy Research Institute
>>> Texas A&M University
>>> ebooth@ppri.tamu.edu
>>> Office: +979.845.6754
>>>
>>>
>>>
>>> On Mar 26, 2011, at 2:30 PM, Eric Booth wrote:
>>>
>>>> ***********************!
>>>> clear
>>>> inp str200 var1
>>>> "155 - VITAL DO REGO FILHO - PB - Senador"
>>>> "1111 -  - PP -  - Deputado Federal / 25888 - ATAIDES MENDES PEDROSA -PB - Deputado Estadual"
>>>> "1111 -  - PP -  - Deputado Federal / 22333 - EDNALDO PEREIRA DESANTANA - PB - Deputado Estadual"
>>>> "151 - JOSE WILSON SANTIAGO - PB - Senador"
>>>> "45123 - ANTONIO HERVAZIO BEZERRA CAVALCANTI - PB - Deputado Estadual"
>>>> "1212 - DAMIÃO FELICIANO DA SILVA - PB - Deputado Federal"
>>>> end
>>>>
>>>> **using split**
>>>> replace var1 = subinstr(var1, " / ", " - ", .)
>>>> split var1, p("-")
>>>>
>>>> **trim spaces in new vars**
>>>> ds var1?
>>>> foreach v in `r(varlist)' {
>>>>       replace `v'  = trim(`v')
>>>>       }
>>>>
>>>>
>>>> **it looks like the substr you want are in vars14, var15, var19:
>>>> l var14 var15 var19
>>>>
>>>> **grab the title or subtitle or gen an indicator if they are present**
>>>> g str50 title = var14 if  !mi(var14)
>>>> replace title = var15 if mi(title) & !mi(var15)
>>>> g str50 title2 = var19 if  !mi(var19)
>>>> l var1 title title2
>>>> **or
>>>> g titleind = 1 if !mi(var14) | !mi(var15)
>>>> g title2ind = 1 if !mi(var19)
>>>> order *ind
>>>> ***********************!
>>>
>>>
>>>
>>> *
>>> *   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/
>>>
>>
>> *
>> *   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/
>>
>
> *
> *   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/
>

*
*   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/