#### The complete picture: all Bible books

The table below summarises every Bible book that appears in the data — Folger marks,
Shaheen references, Stritmatter additions and matches — sortable by any column.
Books are in canonical Bible order by default. Sort by **Shaheen refs** to see immediately
where Shakespeare's attention fell; sort by **Folger marks** to see where the annotator's
did. The two lists do not coincide.



```{r}
#| echo: false
#| message: false
#| warning: false
library(DT)
library(dplyr)
library(stringr)
library(ggplot2)
library(tidyr)
library(purrr)

cb <- readr::read_csv("cb16.csv", show_col_types = FALSE)
```


# ---------------------------------------------------------------
# CHART 2: scatter — marks vs shaheen refs, labelled
# ---------------------------------------------------------------
scatter_data <- book_summary %>%
  filter(marks > 0 | shaheen > 10) %>%
  mutate(label = ifelse(marks > 5 | shaheen > 50, booklink_clean, ""))

ggplot(scatter_data, aes(x=shaheen, y=marks, label=label)) +
  geom_point(aes(colour=testament), size=2.5, alpha=0.8) +
  geom_text(nudge_y=1.5, size=2.8, colour="#444", family="sans") +
  scale_colour_manual(values=c("OT"="#6b3a2a","NT"="#4a6741","Ap"="#888888"),
                      labels=c("Old Testament","New Testament","Apocrypha"),
                      na.value="#aaaaaa") +
  labs(x="Shaheen references", y="Folger marks",
       colour=NULL,
       title="Marks vs references: the scatter tells the story",
       subtitle="Books in the top-right would support the Oxfordian claim. None are there.") +
  theme_minimal(base_family="sans") +
  theme(
    plot.title    = element_text(size=11, face="bold", colour="#3d0e0e"),
    plot.subtitle = element_text(size=8,  colour="#666"),
    legend.position = "top"
  )


# ---------------------------------------------------------------
# BUILD BOOK-LEVEL SUMMARY
# Marks keyed on 'book', references on 'booklink'
# ---------------------------------------------------------------
cb_summary <- cb %>%
  mutate(
    booklink_clean = str_replace_all(trimws(booklink), "\\+", " "),
    book_clean     = trimws(book)
  )

mark_counts <- cb_summary %>%
  filter(!sourcetype %in% c("ShakRef","StritAdd","StritDirect","StritIndirect",""),
         !is.na(sourcetype)) %>%
  group_by(book_clean) %>%
  summarise(marks = n(), .groups="drop") %>%
  rename(booklink_clean = book_clean)

ref_counts <- cb_summary %>%
  filter(!is.na(booklink_clean), booklink_clean != "") %>%
  group_by(booklink_clean) %>%
  summarise(
    shaheen   = sum(sourcetype == "ShakRef",    na.rm=TRUE),
    s_add     = sum(sourcetype == "StritAdd",   na.rm=TRUE),
    s_dir     = sum(sourcetype == "StritDirect",na.rm=TRUE),
    s_ind     = sum(sourcetype == "StritIndirect",na.rm=TRUE),
    matches   = sum(diagnosticmatch == "FBM",   na.rm=TRUE),
    testament = first(na.omit(testament)),
    book_order= min(book.order.key, na.rm=TRUE),
    .groups="drop"
  )

book_summary <- ref_counts %>%
  left_join(mark_counts, by="booklink_clean") %>%
  mutate(marks = coalesce(marks, 0L)) %>%
  filter(shaheen > 0 | s_add > 0 | s_dir > 0 | s_ind > 0) %>%
  arrange(book_order) %>%
  mutate(
    total_refs = shaheen + s_add + s_dir + s_ind,
    odds_n     = ifelse(marks > 0, round(shaheen / marks), NA_integer_),
    odds_sort  = ifelse(is.na(odds_n), 99999L, odds_n),
    odds_display = case_when(
      marks == 0 ~ "—",
      odds_n == 0 ~ "1:0",
      TRUE ~ paste0("1:", odds_n)
    ),
    testament = case_when(
      testament == "Old" ~ "OT",
      testament == "New" ~ "NT",
      TRUE ~ testament
    )
  )

# ---------------------------------------------------------------
# TABLE
# ---------------------------------------------------------------
datatable(
  book_summary %>%
    select(booklink_clean, testament, marks, shaheen,
           s_add, s_dir, s_ind, matches, odds_display, odds_sort),
  colnames = c("Book", "T", "Folger marks", "Shaheen refs",
               "Strit. add.", "Strit. direct", "Strit. indirect",
               "Matches", "Odds", "odds_sort"),
  rownames  = FALSE,
  escape    = FALSE,
  filter    = 'top',
  options   = list(
    paging       = TRUE,
    pageLength   = 25,
    lengthMenu   = list(c(25, 50, -1), c("25","50","All")),
    scrollX      = TRUE,
    autoWidth    = FALSE,
    dom          = 'lfrtip',
    order        = list(list(2, 'desc')),  # default: sort by Shaheen refs desc
    columnDefs   = list(
      list(width='160px', targets=0),
      list(width='35px',  targets=1, className='dt-center'),
      list(width='90px',  targets=2, className='dt-center'),
      list(width='90px',  targets=3, className='dt-center'),
      list(width='80px',  targets=4, className='dt-center'),
      list(width='80px',  targets=5, className='dt-center'),
      list(width='80px',  targets=6, className='dt-center'),
      list(width='70px',  targets=7, className='dt-center'),
      list(width='70px',  targets=8, className='dt-center'),
      list(visible=FALSE, targets=9)   # hide numeric odds sort column
    ),
    initComplete = JS(
      "function(settings, json) {",
      "  $(this.api().table().container()).css({",
      "    'font-family': 'Fira Sans Condensed, sans-serif',",
      "    'font-size': '0.82rem'",
      "  });",
      "}"
    )
  )
) %>%
  formatStyle('booklink_clean',
    fontWeight = 'bold', color = '#3d0e0e') %>%
  formatStyle('marks',
    background = styleColorBar(range(book_summary$marks), '#c8b09080'),
    backgroundSize = '98% 60%',
    backgroundRepeat = 'no-repeat',
    backgroundPosition = 'center') %>%
  formatStyle('shaheen',
    background = styleColorBar(range(book_summary$shaheen), '#4a674180'),
    backgroundSize = '98% 60%',
    backgroundRepeat = 'no-repeat',
    backgroundPosition = 'center') %>%
  formatStyle('odds_display',
    fontWeight = 'bold', color = '#800000', textAlign = 'center')


#| echo: false
#| message: false
#| fig-width: 9
#| fig-height: 6
#| fig-cap: "Folger marks vs Shaheen references by Bible book (top 25 by total references). The mismatch between where Shakespeare looked and where the annotator marked is the core of the argument."

# ---------------------------------------------------------------
# CHART: marks vs refs, top 25 books by shaheen refs
# ---------------------------------------------------------------
# rebuild book_summary in case it didn't persist from chunk 1
book_summary <- ref_counts %>%
  left_join(mark_counts, by="booklink_clean") %>%
  mutate(marks = coalesce(marks, 0L)) %>%
  filter(shaheen > 0 | s_add > 0 | s_dir > 0 | s_ind > 0) %>%
  arrange(book_order) %>%
  mutate(
    total_refs   = shaheen + s_add + s_dir + s_ind,
    odds_n       = ifelse(marks > 0, round(shaheen / marks), NA_integer_),
    odds_sort    = ifelse(is.na(odds_n), 99999L, odds_n),
    odds_display = case_when(
      marks == 0 ~ "—",
      odds_n == 0 ~ "1:0",
      TRUE ~ paste0("1:", odds_n)
    ),
    testament = case_when(
      testament == "Old" ~ "OT",
      testament == "New" ~ "NT",
      TRUE ~ testament
    )
  )

top25 <- book_summary %>%
  arrange(desc(shaheen)) %>%
  slice_head(n=25) %>%
  mutate(book = factor(booklink_clean,
                       levels=rev(booklink_clean)))

# Reshape for grouped bars
plot_data <- bind_rows(
  top25 %>% transmute(book, value=marks,   type="Folger marks"),
  top25 %>% transmute(book, value=shaheen, type="Shaheen refs")
)

ggplot(plot_data, aes(x=book, y=value, fill=type)) +
  geom_col(position=position_dodge(width=0.7), width=0.6) +
  scale_fill_manual(values=c("Folger marks"="#6b3a2a",
                              "Shaheen refs"="#4a6741")) +
  coord_flip() +
  labs(x=NULL, y="Count", fill=NULL,
       title="Where Shakespeare looked vs where the annotator marked",
       subtitle="Top 25 Bible books by Shaheen reference count") +
  theme_minimal(base_family="sans") +
  theme(
    plot.title    = element_text(size=11, face="bold", colour="#3d0e0e"),
    plot.subtitle = element_text(size=8,  colour="#666"),
    legend.position = "top",
    legend.text   = element_text(size=9),
    axis.text.y   = element_text(size=8),
    panel.grid.major.y = element_blank(),
    panel.grid.minor   = element_blank()
  )

#| echo: false
#| message: false
#| fig-width: 7
#| fig-height: 5
#| fig-cap: "Books with Folger marks but few or no Shakespeare references — the annotator marked these; Shakespeare largely ignored them."

# ---------------------------------------------------------------
# CHART 2: scatter — marks vs shaheen refs, interactive
# ---------------------------------------------------------------
library(plotly)

scatter_data <- book_summary %>%
  filter(marks > 0 | shaheen > 10) %>%
  mutate(
    testament_label = case_when(
      testament == "OT" ~ "Old Testament",
      testament == "NT" ~ "New Testament",
      testament == "Ap" ~ "Apocrypha",
      TRUE ~ "Unknown"
    ),
    hover_text = paste0(
      "<b>", booklink_clean, "</b><br>",
      "Shaheen refs: ", shaheen, "<br>",
      "Folger marks: ", marks, "<br>",
      "Strit. additions: ", s_add, "<br>",
      "Matches: ", matches, "<br>",
      testament_label
    )
  )

p <- ggplot(scatter_data,
       aes(x=shaheen, y=marks,
           colour=testament,
           text=hover_text)) +
  geom_point(size=2.5, alpha=0.85) +
  scale_colour_manual(
    values = c("OT"="#6b3a2a","NT"="#4a6741","Ap"="#888888"),
    labels = c("Old Testament","New Testament","Apocrypha"),
    na.value = "#aaaaaa"
  ) +
  labs(
    x      = "Shaheen references",
    y      = "Folger marks",
    colour = NULL,
    title  = "Marks vs references: the scatter tells the story",
    subtitle = "Books in the top-right would support the Oxfordian claim. None are there."
  ) +
  theme_minimal(base_family="sans") +
  theme(
    plot.title    = element_text(size=11, face="bold", colour="#3d0e0e"),
    plot.subtitle = element_text(size=8,  colour="#666"),
    legend.position = "top"
  )

ggplotly(p, tooltip="text") %>%
  layout(
    hoverlabel = list(
      bgcolor    = "#1a1a1a",
      font       = list(family="Fira Sans Condensed", size=12, color="#f5f1e8"),
      bordercolor = "#3d0e0e"
    )
  )
```