How to extract longest string between patterns in HTML [in R] - r

Question

I am extracting text from HTML from a series of articles. However, I am still to get the articles to a format I happy with. More specifically I hope to find the longest string between the occurence of a pattern ("/n").
The code I use now is the following:
library(newsanchor)
library(htm2txt)
library(RCurl)
library(XML)
results <- get_everything(query = "Trump +Trade", language = "en")
test <- results$results_df
test$txt <- NA
for(i in 1:22){
tryCatch({
html <- getURL(test$url[i], followlocation = TRUE)
doc = htmlParse(html, asText=TRUE)
plain.text <- xpathSApply(doc, "//p", xmlValue)
test$txt[i] <- c(paste(plain.text, collapse = "\n"))
}, error=function(e){})
print(i)
}
The result looks something like this
[1] "EDITION\nUS President Donald Trump has made his first meaningful remarks on the Huawei firestorm since his administration blacklisted the Chinese tech giant last week.\nThe president was speaking at a news conference announcing a $US16 billion aid package for farmers caught up in the China trade war when he addressed Huawei, which has been placed on a list that means US firms need permission to do business with the Chinese company.\nTrump started out by saying that Huawei poses a huge security threat to the US. US officials have long floated suspicions that Huawei acts as a conduit for Chinese surveillance.\n“Huawei is something that’s very dangerous. You look at what they have done from a security standpoint, from a military standpoint, it’s very dangerous,” the president told reporters.\n Read more: Here are all the companies that have cut ties with Huawei, dealing the Chinese tech giant a crushing blow\nHe then immediately switched gears to suggest that Huawei could form part of a trade deal with America and China. “So it’s possible that Huawei even would be included in some kind of a trade deal. If we made a deal, I could imagine Huawei being possibly included in some form,” he said.\n\"Huawei is very dangerous,\" Trump says, adding that an exception for the company could be made in a trade deal with China pic.twitter.com/TFlClewBNt\n— TicToc by Bloomberg (#tictoc) May 23, 2019\n\nTrump: “Huawei is something that’s very dangerous. You look at what they have done from a security standpoint, from a military standpoint, it’s very dangerous. So, it’s possible that Huawei even would be included in some kind of a trade deal. If we made a deal, I could imagine Huawei being possibly included in some form of, or some part of a trade deal.”\nJournalist: “How would that look?”\nTrump: “It would look very good for us.”\nJournalist:
I hope to get the most part of the most important part - the actual article. I am not sure how to best do this, but I think it might be to find the longest string between two incidences of ("/n"). Can anyone help doing that, or perhaps even suggest a better method?

Answer 1

Edit: #user101 explained that nchar is vectorized. Here is a more optimal solution:
splitarticle <- unlist(strsplit(i, "\n"))
splitarticle[which.max(nchar(splitarticle))]
Something like this could work, unless I misunderstood what you're trying to do.
splitarticle <- unlist(strsplit(i, "\n"))
lengths <- unlist(lapply(splitarticle, nchar))
splitarticle[match(max(lengths), lengths)]