Question

I’m having a problem wherein xmlValue strips the <br /> tags that I need kept (or transformed to some other character that I can then strsplit on.

Here’s an example:

> f <- htmlParse(getForm("http://sites.target.com/site/en/spot/store_locator_popups.jsp", ajax="true", storeNumber=1889), asText=TRUE)
> xpathSApply(f, "//div[@class=\"sl_results_popup_address\"]", xmlValue)
[1] "1154 S Clark StChicago, IL 60605(312) 212-6300"

Versus the HTML it’s parsing:

<div class="sl_results_popup_address">
1154 S Clark St
<br/>
Chicago, IL 60605
<br/>
[phone]
</div>

I’ve tried , recursive=FALSE but that doesn’t seem to help.

If they were <p> and </p> line breaks then it would be easier since I could just grab them individually, but with <br/> not wrapping the text I really can’t go that direction. Hoping there’s just an option to reduce the level of stripping done within xmlValue (or maybe the <br/>s are being stripped at the parsing-of-document phase?).

Answer 1

two things may help

app.data<-getForm("http://sites.target.com/site/en/spot/store_locator_popups.jsp", ajax="true", storeNumber=1889)
app.data<-gsub("<br>","\n",app.data)
f <- htmlParse(app.data, asText=TRUE)
out<-xpathSApply(f, "//div[@class=\"sl_results_popup_address\"]", xmlValue)
> xpathSApply(f, "//div[@class=\"sl_results_popup_address\"]", xmlValue)
[1] "1154 S Clark St\nChicago, IL 60605\n(312) 212-6300"
>

so just replace the br tags with something else or use your original code and

> xpathSApply(f, "//div[@class=\"sl_results_popup_address\"]/text()", xmlValue)
[1] "1154 S Clark St"   "Chicago, IL 60605" "[phone]"   
>

if you want to keep the tags

dum.fun<-function(x){if(xmlName(x)=="br"){"<br/>"}else{xmlValue(x)}}
xChild<-xpathSApply(f, "//div[@class=\"sl_results_popup_address\"]",xmlChildren)
lapply(xChild,dum.fun)
> unlist(lapply(xChild,dum.fun))
[1] "1154 S Clark St"   "<br/>"             "Chicago, IL 60605"
[4] "<br/>"             "[phone]" 
>