Issue scraping a collapsible table using rvest - r

Question

I am trying to scrape information from multiple collapsible tables from a website called APIS.
An example of what I'm trying to collect is here http://www.apis.ac.uk/select-feature?site=1001814&SiteType=SSSI&submit=Next
Ideally I'd like to be able to have the drop down heading followed by the information underneath, though when using rvest I cant seem to get it to select the correct section from the html.
I'm reasonably new to R, this is what I have from watching some videos about scraping:
link = "http://www.apis.ac.uk/select-feature?site=1001814&SiteType=SSSI&submit=Next"
page = read_html(link)
name = page %>% html_nodes(".tab-tables :nth-child(1)") %>% html_text()
the "name" value displays "Character (empty)"
It may be because I'm new to this and there's a really obvious answer but any help would be appreciated

Answer 1

The data for each tab comes from additional requests you can find in the browser network tab when pressing F5 to refresh the page. For example, the nutrients info comes from:
http://www.apis.ac.uk/sites/default/files/AJAX/srcl_2019/apis_tab_nnut.php?ajax=true&site=1001814&BH=&populateBH=true
Which you can think of more generally as:
scheme='http'
netloc='www.apis.ac.uk'
path='/sites/default/files/AJAX/srcl_2019/apis_tab_nnut.php'
params=''
query='ajax=true&site=1001814&BH=&populateBH=true'
fragment=''
So, you would make your request to those urls you see in the network tab.
If you want to dynamically determine these urls, then make a request, as you did, to the landing page, then regex out from the response text the path (see above) of the urls. This can be done using the following pattern url: "(\\/sites\\/default\\/files\\/.*?)".
You then need to add the protocol + domain (scheme and netloc) to the returned matches based on landing page protocol and domain.
There are some additional query string parameters, which come after the ?, which can also be dynamically retrieved, if reconstructing the urls from the response text. You can see these within the page source:
You probably want to extract each of those data param specs for the Ajax requests e.g. with data:\\s\\((.*?)\\), then have a custom function which turns the matches into the required query string suffix to add to the previously retrieved urls.
Something like the following:
library(rvest)
library(magrittr)
library(stringr)
get_query_string <- function(match, site_code) {
string <- paste0(
"?",
gsub("siteCode", site_code, gsub('["{}]', "", gsub(",\\s+", "&", gsub(":\\s+", "=", match))))
)
return(string)
}
link <- "http://www.apis.ac.uk/select-feature?site=1001814&SiteType=SSSI&submit=Next"
page <- read_html(link) %>% toString()
links <- paste0("http://www.apis.ac.uk", stringr::str_match_all(page, 'url: "(\\/sites\\/default\\/files\\/.*?)"')[[1]][, 2])
params <- stringr::str_match_all(page, "data:\\s\\((.*?)\\),")[[1]][, 2]
site_code <- stringr::str_match_all(page, 'var siteCode = "(.*?)"')[[1]][, 2]
params <- lapply(params, get_query_string, site_code)
urls <- paste0(links, params)