reflow.R

{
library(tidyverse)
library(officer)
library(dplyr)
library(stringr)
library(dbplyr)
library(RSQLite)
library(jsonlite)
}

proj_path <- getwd()
input_path <- 'INPUT'
output_path <- 'OUTPUT'
data_path <- 'DATA'

search_title <- 'Big Data'

find_date_lang_format <- function(sample) {
  langs <- c('pl', 'eng')
  score_pl <- unlist(gregexpr("\\d{1,2}\\s\\p{L}+\\s\\d{4}", sample, perl = TRUE))
  score_eng <- unlist(gregexpr("[A-Za-z]+ \\d{1,2}, \\d{4}", sample, perl = TRUE))
  datelang <- langs[c(score_pl > 0, score_eng > 0)]
  return(datelang)
}

remove_dates <- function(text, datelang) {
  pattern <- switch(datelang,
                    'pl' = "\\d{1,2}\\s\\p{L}+\\s\\d{4}",
                    'eng' = "[A-Za-z]+ \\d{1,2}, \\d{4}",
                    stop("specify 'pl' or 'eng'"))
  text_mod <- str_replace(text, pattern, " Page ")
  text_mod <- str_trim(text_mod)
  return(text_mod)
}

remove_links <- function(text) {
  match <- unlist(gregexpr('HYPERLINK[^"]+"', text, perl = TRUE))
  if (match > 0) {
    text_clean <- gsub('\\s(HYPERLINK)\\s\"\\w.*\"\\s', '', text)
    return(text_clean)
  } else {
    return(text)
  }
}

quote_transform <- function(text, admonitions) {
  if (substr(text, 1, 2) == "##") {
    return(text)
  }
  match <- gregexpr("Page\\s\\d+\\s*(.*?)($|\n(?=Page))", text, perl = TRUE)
  page_nums <- regmatches(text, match)[[1]]
  text <- regmatches(text, match, invert = TRUE)[[1]]
  quote_format <- if (admonitions) "> [!quote] " else "> "
  quotes <- paste0(quote_format, page_nums, "\n> ", text[1])
  return(paste(quotes, collapse = "\n\n"))
}

getGooglePlay <- function(title_docx, admonitions = TRUE) {
  doc <- read_docx(file.path(proj_path, input_path, title_docx))
  text <- docx_summary(doc)
  text <- replace(text,is.na(text),"")
  text <- filter(text, text != '')
  text <- text[2:4]
  
  text$text[text$style_name == "heading 2"] <- paste("###", text$text[text$style_name == "heading 2"], " \n", sep = " ")
  row_num <- which(startsWith(text$text, "##"))[1]
  text <- text[row_num:length(text$text),]
  
  format_d <- find_date_lang_format(text$text[2])
  
  text$text <- unlist(lapply(text$text, remove_dates, datelang = format_d))
  text$text <- unlist(lapply(text$text, remove_links))
  text$text <- unlist(lapply(text$text, quote_transform, admonitions = admonitions))
  md_text <- paste(text$text, collapse = "\n\n")
  
  title_md <- gsub('\\.docx', '.md', title_docx)
  writeLines(md_text, file.path(proj_path, output_path, title_md))
  print("DONE!")
}

# Function to append the front-matter at the beginning of the file
meta_appender <- function(filename, new_text) {
  # Read the contents of the file into a character vector
  lines <- readLines(filename)
  
  # Concatenate the new text and the existing lines into character vector
  lines <- c(new_text, lines)
  
  # Write the modified lines back to the file
  writeLines(lines, filename)
}

# Function to construct the front-matter
fmatter_constructor <- function(meta) {
  # Conditionality on translator
  if (meta$original_language == '') {
    citation <-
      paste0(
        "'",
        meta$author,
        ", ",
        meta$title,
        ", ",
        meta$publisher,
        ", ",
        meta$pub_place,
        " ",
        meta$pub_year,
        "'"
      )
  } else {
    citation <-
      paste0(
        "'",
        meta$author,
        ", ",
        meta$title,
        ", tłum. ",
        meta$translator,
        ", ",
        meta$publisher,
        ", ",
        meta$pub_place,
        " ",
        meta$pub_year,
        "'"
      )
  }
  
  note_meta <- paste0(
    "---\n# NOTE-META\ncreated: ", as.character(today()), "\n",
    "type: 'reference'\n",
    "subtype: 'book'\n",
    "status: 'DRAFT'\n",
    "tags: [Wiedza/Streszczenia, ]\n",
    "alias: \n",
    "#\n"
  )
  object_meta <- paste0(
    "# OBJECT-META\n",
    "title: '", meta$title, "'\n",
    "shorttitle: ", meta$short_title, "\n",
    "original_title: ", meta$original_title, "\n",
    "author: '", meta$author, "'\n",
    "coauthors: ", meta$coauthors, "\n",
    "translator: ", meta$translator, "\n",
    "subject: \n",
    "language: ", meta$language, "\n",
    "original_language: ", meta$original_language, "\n",
    "pages: ", meta$pages, "\n",
    "pub_date: ", meta$pub_date, "\n",
    "pub_place: ", meta$pub_place, "\n",
    "publisher: ", meta$publisher, "\n",
    "ISBN: ", meta$isbn, "\n",
    "OCLC: ", meta$oclc, "\n",
    "#\n"
  )
  sub_meta <- paste0("# SUB-META\n",
                     "read: TRUE\n",
                     "read_year: \n",
                     "citation: ", citation, "\n---\n")
  
  fmatter <- paste0(note_meta, object_meta, sub_meta)
  return(fmatter)
}
create_manifest <- function(){
  meta <- as.data.frame(matrix(NA, ncol = 15, nrow = 1))

names(meta) <-
  c(
    "title",
    "original_title",
    "short_title",
    "publisher",
    "author",
    "coauthors",
    "translator",
    "isbn",
    "oclc",
    "language",
    "original_language",
    "pages",
    "pub_date",
    "pub_year",
    "pub_place"
  )
return(meta)
}
fill_frontmatter <- function(search_title, filename) {
  # ... (The rest of the code is unchanged)
  
  
  # Exec
  fmatter <- fmatter_constructor(meta)
  meta_appender(filename, fmatter)
}





get_callibre <- function(search_title){
con <- dbConnect(SQLite(), dbname = file.path(proj_path,data_path,"metadata.db"))
query <-
  paste0(
    "SELECT books.title, authors.name AS author, books.pubdate, publishers.name AS publisher, languages.lang_code AS lang, custom_column_11.value AS pages, identifiers.val, identifiers.type
FROM books
LEFT JOIN books_authors_link ON books.id = books_authors_link.book
LEFT JOIN authors ON books_authors_link.author = authors.id
LEFT JOIN books_publishers_link ON books.id = books_publishers_link.book
LEFT JOIN publishers ON books_publishers_link.publisher = publishers.id
LEFT JOIN books_languages_link ON books.id = books_languages_link.book
LEFT JOIN languages ON books_languages_link.lang_code = languages.id
LEFT JOIN custom_column_11 ON books.id = custom_column_11.book
LEFT JOIN identifiers ON books.id = identifiers.book
WHERE title LIKE '",
    search_title,
    #"' AND identifiers.type = 'isbn';"
    "';"
  )


meta_calibre <- dbGetQuery(con, query)
meta_calibre<- meta_calibre %>% filter(type == 'isbn')

# Close the connection to the database
dbDisconnect(con)
rm(con)
return(meta_calibre)
}

get_bn <- function(meta){
  #if ISBN missing then use different condition
 # if (isbn_exist) {
    bn_url <-
      paste0("https://data.bn.org.pl/api/institutions/bibs.json?isbnIssn=",
             meta$isbn)
  #} else {
 #   bn_url <-
  #    URLencode(
    #    paste0(
    #      "https://data.bn.org.pl/api/institutions/bibs.json?title=",
      #    meta$title,
      #    "&limit=1"
        #)
     # )
  #}
  json_object_bn <- fromJSON(bn_url)
  
  pos <-
    which(!is.na(json_object_bn[["bibs"]][["marc"]][["fields"]][[1]][["041"]][["ind1"]]))
  if (length(pos) >= 1) {
    meta$original_language <-
      json_object_bn[["bibs"]][["marc"]][["fields"]][[1]][["041"]][["subfields"]][[pos]][["h"]][2]
  } else {
    rm(pos)
    
  }
  
  
  if (!isbn_exist) {
    meta$isbn <- json_object_bn[["bibs"]][["isbnIssn"]]
  }
  
  oc <- json_object_bn[["bibs"]][["marc"]][["fields"]][[1]][["035"]][3]
  uoc <- unlist(oc, recursive = TRUE, use.names = FALSE)
  ocl <- grep("^\\(OCoLC\\)", uoc, value = TRUE)
  if (length(ocl)!=0) {
    meta$oclc <- substr(ocl, start = 8, stop = 100)
  }
  
  
  plc <-
    json_object_bn[["bibs"]][["marc"]][["fields"]][[1]][["260"]][["subfields"]]
  plb <- unlist(plc, recursive = TRUE, use.names = FALSE)
  pub_place <- plb[1]
  meta$pub_place <- substring(pub_place, 1, nchar(pub_place) - 2)
  
  if (!is.na(meta$original_language)) {
    tl <-
      json_object_bn[["bibs"]][["marc"]][["fields"]][[1]][["700"]][["subfields"]]
    tl <- unlist(tl, recursive = TRUE, use.names = FALSE)
    translator <- tl[1]
    translator <- strsplit(translator, " ")[[1]]
    translator <- rev(translator)
    translator <- paste(translator, collapse = " ")
    meta$translator <- substring(translator, 1, nchar(translator) - 1)
  } else {
    meta$translator <- ''
    
  }
  
  
  if (!is.na(meta$original_language)) {
    pos <-
      which(!is.na(json_object_bn[["bibs"]][["marc"]][["fields"]][[1]][["246"]][["ind1"]]))
    or <-
      json_object_bn[["bibs"]][["marc"]][["fields"]][[1]][["246"]][["subfields"]][pos]
    or <- unlist(or, recursive = TRUE, use.names = FALSE)
    or <- or[!is.na(or)]
    #original_title<-paste0(or[2],or[3])
    char <- ","
    og <- or[2]
    
    if (substr(og, nchar(og), nchar(og)) == char) {
      meta$original_title <- og
    } else {
      meta$original_title <- paste0(or[2], or[3])
    }
  } else {
    meta$original_title <- ''
    meta$original_language <- ''
  }
}

get_callibre(search_title)

#-----META------
meta<-create_manifest()
meta_calibre<-get_callibre(search_title)
meta$title <- as.character(meta_calibre$title[1])
meta$short_title <- str_extract(meta$title, "[^.:]+")
meta$publisher <- as.character(meta_calibre$publisher[1])
meta$author <- as.character(meta_calibre$author[1])

#coauthors
if (length(meta_calibre$author) > 1) {
  meta$coauthors <-
    paste0('[', paste(meta_calibre$author[2:length(meta_calibre$author)], collapse =
                        ", "), ']')
} else {
  meta$coauthors <- ''
}
meta$isbn <- as.character(meta_calibre$val)
isbn_exist <- !is.na(meta$isbn)

meta$language <- as.character(meta_calibre$lang[1])
meta$pages <- as.numeric(meta_calibre$pages[1])

pub_date_raw <- as.POSIXct(
  as.character.Date(meta_calibre$pubdate[1]),
  tz = "",
  tryFormats = c(
    "%Y-%m-%d %H:%M:%OS",
    "%Y/%m/%d %H:%M:%OS",
    "%Y-%m-%d %H:%M",
    "%Y/%m/%d %H:%M",
    "%Y-%m-%d",
    "%Y/%m/%d"
  )
)
meta$pub_date <- format(pub_date_raw, "%Y-%m-%d")
meta$pub_year <- format(pub_date_raw, "%Y")

get_bn(meta)