aftables: Create Spreadsheet Publications Following Best Practice

Description

Generate spreadsheet publications that follow best practice guidance from the UK government's Analysis Function, available at https://analysisfunction.civilservice.gov.uk/policy-store/releasing-statistics-in-spreadsheets/, with a focus on accessibility. See also the 'Python' package 'gptables'.

Author(s)

Maintainer: Olivia Box Power Olivia.BoxPower@dhsc.gov.uk [contributor]

Authors:

• Matt Dray mwdray@gmail.com

Other contributors:

• Tim Taylor [contributor]

• Matt Kerlogue [contributor]

• Crown Copyright [copyright holder]

See Also

Useful links:

• https://best-practice-and-impact.github.io/aftables/

• https://github.com/best-practice-and-impact/aftables

• Report bugs at https://github.com/best-practice-and-impact/aftables/issues

Coerce To An 'aftable' Object

Description

Functions to check if an object is an aftable, or coerce it if possible.

Usage

as_aftable(x)

is_aftable(x)

Arguments

Value

as_aftable returns an object of class aftable if possible. is_aftable returns TRUE if the object has class aftable, otherwise FALSE.

Examples

as_aftable(demo_df)
is_aftable(demo_aftable)

Insert Demo 'create_aftable' Template

Description

Insert at the cursor a template for create_aftable from the 'aftable' package, pre-filled with demo data.

Usage

at_template_aftable()

Value

Empty list. Function is used for side effect.

Insert Full Demo 'aftables' Template Workflow

Description

Insert at the cursor (a) demo templates for cover, contents and notes tables, and (b) a call to create_aftable pre-filled with demo data.

Usage

at_template_workflow()

Value

Empty list. Function is used for side effect.

Create An 'aftable' Object

Description

Create a new aftable-class object, which is a special data.frame that contains all the information needed in your output spreadsheet. In turn, the object created by this function can be used to populate an 'openxlsx' Workbook-class object with the function generate_workbook.

Usage

create_aftable(
  tab_titles,
  sheet_types = c("cover", "contents", "notes", "tables"),
  sheet_titles,
  blank_cells = NA_character_,
  sources = NA_character_,
  custom_rows = list(NA_character_),
  tables
)

Arguments

Details

How to supply data to the 'tables' argument

Formats for the elements collected as a list and passed to the 'tables' argument, depending on the sheet type.

• Sheet type 'cover': either (a) a list where each element name is a section header and each element's content is a character vector whose elements will make up separate rows of that section (recommended), or (b) a data.frame with one row per subsection, with one column for section titles and one column for corresponding for that section's body text. For example, you may have a section with the title 'Contact details' that contains an email address and telephone number. You can use line breaks (i.e. '\n') to separate text into paragraphs.

• Sheet type 'contents': one row per sheet, two columns suggested at least (named 'Tab title' and 'Worksheet title').

• Sheet type 'notes': one row per note, two columns suggested (named 'Note number', 'Note text'), where notes are in the form '[note 1]'.

• Sheet type 'tables': a tidy, rectangular data.frame containing the data to be published. It's the user's responsibility to add notes in the form '[note 1]' to column headers, or in a special 'Notes' row.

How to supply hyperlinks

You can provide text in Markdown link syntax (e.g. '[GOV.UK](https://www.gov.uk)', adding 'mailto:' before an email address) and the containing cell will be rendered as a hyperlink in the output spreadsheet. Note that whole cells will become hyperlinks; there is no support for selected words in a sentence to be rendered as a hyperlink.

Hyperlinks can be supplied in the character strings to three arguments:

• To the 'tables' argument for sheet type 'cover' only. It's recommended to supply the cover information as a list rather than a data.frame, which will allow you to make specific rows within a section (e.g. 'contact us') into hyperlinks.

• To the 'custom_rows' argument for sheets of type 'contents, 'notes' and 'tables'.

• To the 'source' argument for sheets of type 'table' only.

Value

An object with classes 'aftable', 'tbl' and 'data.frame'.

Examples

# Prepare some demo tables of information

set.seed(1066)

cover_list <- list(
  "Section 1" = c("First row of Section 1.", "Second row of Section 1."),
  "Section 2" = "The only row of Section 2.",
  "Section 3" = c(
    "[Website](https://best-practice-and-impact.github.io/aftables/)",
    "[Email address](mailto:fake.address@aftables.com)"
  )
)

contents_df <- data.frame(
  "Sheet name" = c("Notes", "Table_1", "Table_2"),
  "Sheet title" = c(
    "Notes used in this workbook",
    "First Example Sheet",
    "Second Example Sheet"
  ),
  check.names = FALSE
)

notes_df <- data.frame(
  "Note number" = paste0("[note ", 1:2, "]"),
  "Note text" = c("First note.", "Second note."),
  check.names = FALSE
)


table_1_df <- data.frame(
  Category = LETTERS[1:10],
  "Numeric [note 1]" = 1:10,
  "Numeric suppressed" = c(1:4, "[c]", 6:9, "[x]"),
  "Numeric thousands" = abs(round(rnorm(10), 4) * 1e5),
  "Numeric decimal" = abs(round(rnorm(10), 5)),
  "Long name that means that the column width needs to be widened" = 1:10,
  Notes = c("[note 1]", rep(NA_character_, 4), "[note 2]",
            rep(NA_character_, 4)),
  check.names = FALSE
)

table_2_df <- data.frame(Category = LETTERS[1:10], Numeric = 1:10)

# Create 'aftables' object

x <- aftables::create_aftable(
  tab_titles = c("Cover", "Contents", "Notes", "Table_1", "Table_2"),
  sheet_types = c("cover", "contents", "notes", "tables", "tables"),
  sheet_titles = c(
    "The 'aftables' Demo Workbook",
    "Table of contents",
    "Notes",
    "Table 1: First Example Sheet",
    "Table 2: Second Example Sheet"
  ),
  blank_cells = c(
    rep(NA_character_, 3),
    "Blank cells indicate that there's no note in that row.",
    NA_character_
  ),
  custom_rows = list(
    NA_character_,
    NA_character_,
    "A custom row.",
    c(
      paste0("First custom row [with a hyperlink.]",
             "(https://best-practice-and-impact.github.io/aftables/)"),
      "Second custom row."
    ),
    "A custom row."
  ),
  sources = c(
    rep(NA_character_, 3),
    paste0("[The Source Material, 2024.]",#
           "(https://best-practice-and-impact.github.io/aftables/)"),
    "The Source Material, 2024."
  ),
  tables = list(cover_list, contents_df, notes_df, table_1_df, table_2_df)
)

# Test that 'aftable' is one of the object's classes
is_aftable(x)

# Look at the structure of the object
str(x, max.level = 2)

A Demo 'aftables' Object

Description

A pre-created 'aftables' object ready to be converted to an 'openxlsx' Workbook-class object with generate_workbook.

Usage

demo_aftable

Format

A data.frame with 6 rows and 7 columns:

tab_title

Character. Text to appear on each sheet's tab.

sheet_type

Character. The content type for each sheet: 'cover', 'contents', 'notes', or 'tables'.

sheet_title

Character. The title that will appear in cell A1 (top-left) of each sheet.

blank_cells

Character. An explanation for any blank cells in the table.

custom_rows

List-column of character vectors. Additional arbitrary pre-table information provided by the user.

source

Character. The origin of the data, if relevant.

table

List-column of data.frames (apart from the cover, which is a list) containing the statistical tables.

A Demo 'data.frame' Object

Description

A pre-created data.frame ready to be converted to an aftables-class object with as_aftable and then an 'openxlsx' Workbook-class object with generate_workbook.

Usage

demo_df

Format

A data.frame with 6 rows and 7 columns:

tab_title

Character. Text to appear on each sheet's tab.

sheet_type

Character. The content type for each sheet: 'cover', 'contents', 'notes', or 'tables'.

sheet_title

Character. The title that will appear in cell A1 (top-left) of each sheet.

blank_cells

Character. An explanation for any blank cells in the table.

custom_rows

List-column of character vectors. Additional arbitrary pre-table information provided by the user.

source

Character. The origin of the data, if relevant.

table

List-column of data.frames (apart from the cover, which is a list) containing the statistical tables.

A Demo 'Workbook' Object

Description

A pre-created 'openxlsx' Workbook'-class object generated from an aftables-class object with generate_workbook.

Usage

demo_workbook

Format

An 'openxlsx' Workbook-class object with 5 sheets.

Generate A Workbook Object From An 'aftable'

Description

Populate an 'openxlsx' Workbook-class object with content from an aftable-class object. In turn, the output can be passed to saveWorkbook from 'openxlsx'

Usage

generate_workbook(aftable)

Arguments

Value

A Workbook-class object.

Examples

# Convert an aftable to a Workbook-class object
x <- generate_workbook(demo_aftable)
class(x)

# As above, using a compliant data.frame and the base pipe
y <- demo_df |>
  as_aftable() |>
  generate_workbook()

Summarise An 'aftable' Object

Description

A concise result summary of an aftable-class object to see information about the sheet content. Shows a numbered list of sheets with each tab title, sheet type and table dimensions.

Usage

## S3 method for class 'aftable'
summary(object, ...)

Arguments

Value

object unaltered.

Examples

# Print a concise summary of the aftable-class object
summary(demo_aftable)

# Alternatively, look at the structure
str(demo_aftable, max.level = 2)

Provide A Succinct Summary Of An 'aftable' Object

Description

A brief text description of an aftable-class object.

Usage

## S3 method for class 'aftable'
tbl_sum(x, ...)

Arguments

Value

Named character vector.

Examples

# Print with description
print(demo_aftable)

# Print description only (package 'tibble' must be installed)
tibble::tbl_sum(demo_aftable)