How to save multiples match in one column? rvest, R and stringr

Question

This question is a sequence to the problem stackoverflow

I have these two example html: url1.html ; url2.html

The url3.html is another example with more IPC

In URL2.html there is no information (51) and in URL1.html there is.

I'm using this code in R:

library(rvest)
library(tidyverse)
library(stringr)

x<-data.frame(
    URL=c(1:2),
    page=c(paste(readLines("url1.html"), collapse="\n"),
                 paste(readLines("url2.html"), collapse="\n"))
) 

for (i in 1:nrow(x)){
    html<-x$page[i]%>% unclass() %>% unlist()
    read_html(html,encoding = "ISO-8859-1") %>% 
        rvest::html_elements(xpath = '//*[@id="principal"]/table[2]') %>%
        html_nodes(xpath='//div[@id="classificacao0"]') %>%  
        html_text(trim=T)%>%  
        str_replace_all(.,"[\\n\\r\\t]+", "")%>%
        stringr::str_trim( ) -> tmp
    
    if(length(tmp) == 0) tmp <- "ND"
    x$ipc_0[i] <- tmp %>% str_replace_all(.,"\\s+", " ") %>% str_replace_all(.," \\)", "\\)")
}

for (i in 1:nrow(htm_temp)){
    html<-x$page[i]%>% unclass() %>% unlist()
    read_html(html,encoding = "ISO-8859-1") %>% 
        rvest::html_elements(xpath = '//*[@id="principal"]/table[2]') %>%
        html_nodes(xpath='//div[@id="classificacao1"]') %>%  
        html_text(trim=T)%>%  
        str_replace_all(.,"[\\n\\r\\t]+", "")%>%
        stringr::str_trim( ) -> tmp
    
    if(length(tmp) == 0) tmp <- "ND"
    x$ipc_1[i] <- tmp %>% str_replace_all(.,"\\s+", " ") %>% str_replace_all(.," \\)", "\\)")
}

Result: partially correct

Desired result:create a new dataframe with the following structure.

URL	IPC
1	B62B 1/16 (1968.09)...
1	B62B 1/00 (1968.09)...
2	ND

Problem: There are url`s that have the code (51) and others that do not. When you have the code (51) the structure can contain "n" id with the following structure xpath='//div[@id="classificacao0"]. the Rating Id can contain values from 0 to "n". How to optimize this code to capture the necessary information without having to do a lot of for (variable in vector) for each "n"?

Any idea how to solve this problem?