# Load the jsonlite and httr libraries for handling JSON data and making HTTP requests.
library(jsonlite)
library(httr)

# Define the URL for the Kramerius API.
kramerius5_api <- "https://kramerius.lib.cas.cz/search/api/v5.0/item/"

# Define the UUID of the title for which you want to retrieve OCR text.
UUID_titul <- "uuid:f24658de-c158-43b0-a41b-5d1761c85cb2"

# Construct the URL for retrieving children items of the specified title.
request <- paste0(kramerius5_api, UUID_titul, "/children")

# Send a GET request to the Kramerius API to retrieve the JSON response.
response <- GET(request)

# Parse the JSON response into R data structures.
parsed_json <- fromJSON(content(response, "text"))

# Extract URLs for retrieving OCR text streams.
urls <- paste0(kramerius5_api, parsed_json$pid, "/streams/TEXT_OCR")

# Retrieve OCR text from each URL and store it in a list.
text_list <- lapply(urls, function(url) {
  response <- GET(url)
  content(response, "text", encoding = "UTF-8")
})

# Convert the list of OCR text into a character vector.
text_list <- as.character(text_list)

# Write the OCR text to a text file named "stazeny_text_Vesela_Zide.txt" in the current working directory.
writeLines(text_list, "stazeny_text_Vesela_Zide.txt")