Spread Large Table Over Smaller Tables

Takes a large table and spreads it over smaller tables to paginate it. It will preserve common row information on the left and separate columns according to maximum specifications. The final tables will have widths less than or equal to both max_col and max_width, and heights less than or equal to both max_row and max_height.

Usage

span_table(
  table_body = NULL,
  row_common = NULL,
  table_body_head = NULL,
  row_common_head = NULL,
  header_format = "text",
  obnd = NULL,
  max_row = 20,
  max_col = 10,
  max_height = 7,
  max_width = 6.5,
  table_alignment = "center",
  inner_border = officer::fp_border(color = "black", width = 0.3),
  outer_border = officer::fp_border(color = "black", width = 2),
  set_header_inner_border_v = TRUE,
  set_header_inner_border_h = TRUE,
  set_header_outer_border = TRUE,
  set_body_inner_border_v = TRUE,
  set_body_inner_border_h = FALSE,
  set_body_outer_border = TRUE,
  notes_detect = NULL
)

Arguments

table_body: Data frame with the body of the large table.
row_common: Data frame with the common rows.
table_body_head: Data frame or matrix with headers for the table body.
row_common_head: Data frame or matrix with headers for the common rows.
header_format: Format of the header either "text" (default) or "md" for markdown.
obnd: Optional onbrand object used to format markdown. The default NULL value will use default formatting.
max_row: Maximum number of rows in output tables (A value of NULL will set max_row to the number of rows in the table).
max_col: Maximum number of columns in output tables (A value of NULL will set max_col to number of columns in the table).
max_height: Maximum height of the final table in inches (A value of NULL will use 100 inches).
max_width: Maximum width of the final table in inches (A value of NULL will use 100 inches).
table_alignment: Character string specifying the alignment #'of the table (body and headers). Can be "center" (default), "left", "right", or "justify"
inner_border: Border object for inner border lines defined using officer::fp_border()
outer_border: Border object for outer border lines defined using officer::fp_border()
set_header_inner_border_v: Boolean value to enable or disable inner vertical borders for headers
set_header_inner_border_h: Boolean value to enable or disable inner horizontal borders for headers
set_header_outer_border: Boolean value to enable or disable outer border for headers
set_body_inner_border_v: Boolean value to enable or disable inner vertical borders for the body
set_body_inner_border_h: Boolean value to enable or disable inner horizontal borders for the body
set_body_outer_border: Boolean value to enable or disable outer border borders for the body
notes_detect: Vector of strings to detect in output tables (example c("NC", "BLQ")).

Value

list with the following elements

isgood: Boolean indicating the exit status of the function.
one_body: Full table with no headers.
one_table: Full table with headers.
msgs: Vector of text messages describing any errors that were found.
tables: Named list of tables. Each list element is of the output. format from build_span().

Details

The way the data frames relate to each other are mapped out below. The dimensions of the different data frames are identified below (nrow x ncol)

#|-------------------------------------| ---
#|                 |                   |  ^
#|                 |                   |  |
#| row_common_head |  table_body_head  |  | m
#|      m x n      |       m x c       |  |
#|                 |                   |  v
#|-------------------------------------| ---
#|                 |                   |  ^
#|                 |                   |  |
#|    row_common   |    table_body     |  | r
#|      r x n      |      r x c        |  |
#|                 |                   |  |
#|                 |                   |  v
#|-------------------------------------| ---
#
#|<--------------->|<----------------->|
#        n                   c

Examples

library(formods)
library(readxl)
library(dplyr)
library(tidyr)

# First we need to create some data. 
# This will read in a test dataset:
DS = readxl::read_excel(
 path  = system.file(package="formods", 
                     "test_data", 
                     "TEST_DATA.xlsx"),
 sheet = "DATA")

# This will filter the dataset down and modify the formatting
DS = dplyr::filter(DS, EVID == 0)                          |>
  dplyr::filter(ID <= 30)                                  |>
  dplyr::select(ID, TIME, DV, CMT)                         |>
  dplyr::mutate(CMT  = ifelse(.data[["CMT"]] == "C_ng_ml", 
                             "Test Article", 
                             .data[["CMT"]]))              |>
  dplyr::select(ID, TIME, DV, CMT)                         |>
  dplyr::mutate(CMT = ifelse(.data[["CMT"]] == "BM_ng_ml", 
                            "Biomarker", 
                            .data[["CMT"]]))               |>
  dplyr::rename(Analyte = "CMT")                           |>
  dplyr::mutate(DV = ifelse(.data[["DV"]] == 0, "BQL", .data[["DV"]]))

# This represents the large table we want to split up into smaller tables 
wide_df = tidyr::pivot_wider(DS,
  values_from = "DV",
  names_from  = "ID")               |>
  dplyr::arrange(Analyte, TIME)     |>
  dplyr::mutate(TIME = as.character(.data[["TIME"]]))    

# The first two columns represent the rows
# that are common across the tables:
row_common = wide_df[,1:2]

# The remaining columns represent the body of the table:
table_body = wide_df[,3:ncol(wide_df)]

# Next we create matrices that contain the header 
# information for each component above:
row_common_head = matrix(
  data  = c("Time", "Analyte",
            "(hr)", "(ng/ml)"),
  ncol  = 2,
  byrow = TRUE)

table_body_head = matrix(
  data  = c(rep("Subject ID", times=ncol(table_body)), 
            names(table_body)),
  ncol  = ncol(table_body),
  byrow = TRUE)


# This builds all of the tables:
span_res = span_table(table_body      = table_body,
                      row_common      = row_common,
                      table_body_head = table_body_head,
                      row_common_head = row_common_head,
                      notes_detect    = c("BQL"))

# This will show you the first one:
span_res$tables[["Table 1"]]$ft
#> a flextable object.
#> col_keys: `TIME`, `Analyte`, `1`, `2`, `3`, `4`, `5`, `6` 
#> header has 2 row(s) 
#> body has 20 row(s) 
#> original dataset sample: 
#>   TIME   Analyte    1    2     3    4    5     6
#> 1    0 Biomarker 1250 1030   831 1170 1140   868
#> 2 0.25 Biomarker 1255 1032 832.2 1172 1145 869.4
#> 3  0.5 Biomarker 1259 1034 833.4 1174 1149 870.9
#> 4    1 Biomarker 1269 1037 835.8 1178 1158 873.7
#> 5    2 Biomarker 1287 1045 840.5 1186 1176 879.4

# This will build a single table with the rows and 
# column ranges specified
bs_res =   build_span(table_body      = table_body,
                      row_common      = row_common,
                      table_body_head = table_body_head,
                      row_common_head = row_common_head,
                      col_sel         = c(1:10),
                      row_sel         = c(1:10),
                      notes_detect    = c("BQL"))

# This will show you the result:
bs_res$ft
#> a flextable object.
#> col_keys: `TIME`, `Analyte`, `1`, `2`, `3`, `4`, `5`, `6`, `7`, `8`, `9`, `10` 
#> header has 2 row(s) 
#> body has 10 row(s) 
#> original dataset sample: 
#>   TIME   Analyte    1    2     3    4    5     6   7    8    9    10
#> 1    0 Biomarker 1250 1030   831 1170 1140   868 643 1050 1130   744
#> 2 0.25 Biomarker 1255 1032 832.2 1172 1145 869.4 644 1051 1134 745.9
#> 3  0.5 Biomarker 1259 1034 833.4 1174 1149 870.9 645 1052 1138 747.8
#> 4    1 Biomarker 1269 1037 835.8 1178 1158 873.7 647 1055 1146 751.7
#> 5    2 Biomarker 1287 1045 840.5 1186 1176 879.4 651 1059 1162 759.2

Usage

Arguments

Value

Details

See also

Examples