Write an analysis as a Markdown or HTML report

Renders an anltab() result as a report: the change table with its intervals and adjusted p-values, the rates, the trends, the drift screen and the column profile, each with one sentence saying how to read it. The Markdown is plain enough to paste into a notebook or a pull request; the HTML is a single self-contained file with no external resources, suitable for e-mailing to a reviewer.

Usage

rptanl(
  x,
  path = NULL,
  format = NULL,
  title = "Analysis of a published table",
  digits = 1L
)

Arguments

x

A bricklayer_analysis from anltab().

path

Where to write, a file path. With NULL (the default) the text is returned invisibly and nothing is written.

format

"markdown" or "html". With NULL (the default), a path ending in .html or .htm gives HTML and anything else, or no path, gives Markdown.

title

Report title, a string; the top heading and, for HTML, the page title.

digits

Number of decimals for percentages and rates, default 1.

Value

The report text (a length-1 character string, Markdown or a full HTML document), invisibly.

Details

Sections appear in this order: a header line (value, period, groups, rows, period range and the time of the analysis), “Change between periods”, “Rates per ...” (when there were rates), “Trend over the series”, “Drift against the prior capsule” (when a prior was given) and “What is in the table” (with “Missing values” when any are missing). Tables use pipe syntax; numbers are rounded to digits decimals, whole numbers of 10,000 and above get thousands separators, logicals are written yes/no and NA is blank. When the values are percentages the change is reported in percentage points. The HTML is produced by a small built-in converter (headings, paragraphs, tables and inline code, with the text escaped). The file is written as UTF-8 and overwritten if it exists.

Errors: x not from anltab(); an invalid format.

Examples

path <- system.file("extdata", "otis_a01_individuals.csv",
                    package = "rmoriebricklayer")
a <- anltab(read.csv(path), value = "individuals",
                   prd = "year", by = c("table", "group"))
md <- rptanl(a)
cat(substr(md, 1, 400))
#> # Analysis of a published table
#> 
#> `individuals` by `year`, grouped by `table` and `group`; 15 rows; 3 period(s) from 2023 to 2025; generated 2026-10-30 17:26.
#> 
#> ## Change between periods
#> 
#> Percent change with a 95% exact conditional-binomial interval; p-values are exact and adjusted over the 10 comparisons by BH. A row is significant when the adjusted p-value is below 0.05.
#> 
#> | table | group | year | 

# HTML, chosen from the file extension
out <- tempfile(fileext = ".html")
rptanl(a, out, title = "OTIS A01: individuals in custody", digits = 2)
readLines(out, n = 3)
#> [1] "<!DOCTYPE html>"                                 
#> [2] "<html lang=\"en\"><head><meta charset=\"utf-8\">"
#> [3] "<title>OTIS A01: individuals in custody</title>" 

# only an anltab() result can be reported
try(rptanl(read.csv(path)))
#> Error : `x` must come from anltab()
unlink(out)