knitr-combine_words

 parse_item_gov_sp  < - function(item_lista) { 

     classes_div  < - item_lista %>% rvest::html_attr("class") %>%  

         stringr::str_split(pattern = " ") %>% purrr::pluck(1) 

     id  < - classes_div[1] %>% stringr::str_remove("post-") 

     categorias  < - classes_div %>% tibble::as_tibble() %>% dplyr::filter(stringr::str_starts(value,  

         "category-")) %>% dplyr::mutate(categorias = stringr::str_remove(value,  

         "category-"), categorias = stringr::str_replace_all(categorias,  

         "-", " ")) %>% dplyr::pull(categorias) %>% knitr::combine_words(sep = ", ",  

         and = "") %>% as.character() 

     tags  < - classes_div %>% tibble::as_tibble() %>% dplyr::filter(stringr::str_starts(value,  

         "tag-")) %>% dplyr::mutate(tags = stringr::str_remove(value,  

         "tag-"), tags = stringr::str_replace_all(tags, "-", " ")) %>%  

         dplyr::pull(tags) %>% knitr::combine_words(sep = ", ",  

         and = "") %>% as.character() 

     if (length(tags) == 0) { 

         tags  < - NA 

     } 

     classe_infos  < - item_lista %>% rvest::html_element(xpath = ".//*[@class=\"col-md-5 category-infos\"]") 

     if (is.na(classe_infos[1])) { 

         classe_infos_antigas  < - item_lista %>% rvest::html_element(xpath = ".//*[@class=\"no-thumbnail category-infos\"]") 

         data_bruta  < - classe_infos_antigas %>% rvest::html_element("span") %>%  

             rvest::html_text() 

         data  < - data_bruta %>% tibble::as_tibble() %>% tidyr::separate(col = value,  

             into = c("data", "horario"), sep = "-") %>% dplyr::mutate(dplyr::across(tidyselect::everything(),  

             stringr::str_trim), data = lubridate::dmy(data)) 

         classe_titulo  < - classe_infos_antigas %>% rvest::html_element("h3") 

         url_noticia  < - classe_infos_antigas %>% rvest::html_element("a") %>%  

             rvest::html_attr("href") 

         titulo  < - classe_infos_antigas %>% rvest::html_element("h3") %>%  

             rvest::html_text() 

         chamada  < - classe_infos_antigas %>% rvest::html_element("p") %>%  

             rvest::html_text() 

     } 

     else { 

         data_bruta  < - classe_infos %>% rvest::html_element("span") %>%  

             rvest::html_text() 

         data  < - data_bruta %>% tibble::as_tibble() %>% tidyr::separate(col = value,  

             into = c("data", "horario"), sep = "-") %>% dplyr::mutate(dplyr::across(tidyselect::everything(),  

             stringr::str_trim), data = lubridate::dmy(data)) 

         classe_titulo  < - classe_infos %>% rvest::html_element("h3") 

         url_noticia  < - classe_titulo %>% rvest::html_element("a") %>%  

             rvest::html_attr("href") 

         titulo  < - classe_titulo %>% rvest::html_text() 

         chamada  < - classe_infos %>% rvest::html_element("p") %>%  

             rvest::html_text() 

     } 

     classe_thumbnail  < - item_lista %>% rvest::html_element(xpath = ".//*[@class=\"col-md-3 category-thumbnail\"]") 

     if (is.na(classe_thumbnail[1])) { 

         img_url  < - NA 

         img_alt  < - NA 

         url_noticia_img  < - NA 

     } 

     else { 

         url_noticia_img  < - classe_thumbnail %>% rvest::html_element("a") %>%  

             rvest::html_attr("href") 

         img_url  < - classe_thumbnail %>% rvest::html_element("img") %>%  

             rvest::html_attr("src") 

         img_alt  < - classe_thumbnail %>% rvest::html_element("img") %>%  

             rvest::html_attr("alt") 

     } 

     tibble::tibble(id, data, url_noticia, titulo, chamada, categorias,  

         tags, img_url, img_alt, url_noticia_img) 

 }