## HTML tags are removed and duplicates are merged
test_that("HTML tags are removed and records are merged", {
  df <- data.frame(
    id = c("A1", "B2"),
    source = c("WOS", "SCOPUS"),
    title = c("Study X", "Study X"),
    author = c("Smith J", "Smith J"),
    year = c("2020", "2020"),
    journal = c("Journal A", "Journal A"),
    doi = c("10.1234/xyz", "10.1234/xyz"),
    volume = c("1", "1"),
    issue = c("1", "1"),
    pages = c("1-5", "1-5"),
    abstract = c(NA, "Abstract text"),
    stringsAsFactors = FALSE
  )

  out <- remove_duplicate(df)

  expect_equal(nrow(out), 1)
  expect_equal(out$abstract, "Abstract text")
})

## Primary source is preserved; provenance is tracked
test_that("primary source is preserved and provenance is tracked", {

  df <- data.frame(
    id = c("A1", "B2"),
    source = c("WOS", "SCOPUS"),
    title = c("Study X", "Study X"),
    author = c("Smith J", "Smith J"),
    year = c("2020", "2020"),
    journal = c("Journal A", "Journal A"),
    doi = c("10.1234/xyz", "10.1234/xyz"),
    volume = c("1", "1"),
    issue = c("1", "1"),
    pages = c("1-5", "1-5"),
    stringsAsFactors = FALSE
  )

  out <- remove_duplicate(df)

  expect_equal(nrow(out), 1)
  expect_equal(out$source, "WOS")
  expect_equal(out$source_provenance, "WOS;SCOPUS")
})

## Existing values are not overwritten during merge
test_that("existing values are not overwritten when merging", {
  df <- data.frame(
    id = c("A1", "B2"),
    source = c("WOS", "SCOPUS"),
    title = c("Study Y", "Study Y"),
    author = c("Smith J", "Smith J"),
    year = c("2022", "2022"),
    journal = c("Journal B", "Journal B"),
    doi = c("10.1234/abc", "10.1234/abc"),
    volume = c("2", "2"),
    issue = c("1", "1"),
    pages = c("10-20", "10-20"),
    stringsAsFactors = FALSE
  )

  out <- remove_duplicate(df)
  expect_equal(out$doi, "10.1234/abc")
})