oeis.tools

R interface to the Online Encyclopedia of Integer Sequences (OEIS). Fetches sequence metadata and b-files, parses them into S3 objects (Sequence, BFile), and plots terms with ggplot2. Terms are stored as gmp::bigz values, so precision is not lost for sequences whose terms exceed double-precision range.

This is the R port of oeis-tools, the Python package by the same author. The public API (function and class names) matches the Python package where R's type system and idioms allow it.

Installation

Install the released version from CRAN:

install.packages("oeis.tools")

Or the development version from GitHub:

# install.packages("pak")
pak::pak("oeistools/oeis-tools-R")

Usage

Sequence metadata

library(oeis.tools)

fib <- Sequence("A000045")

fib$name
# [1] "Fibonacci numbers: F(n) = F(n-1) + F(n-2) with F(0) = 0 and F(1) = 1."

fib$keyword
# [1] "core" "nonn" "nice" "easy" "look" "hear" "changed"

fib$author
# [1] "N. J. A. Sloane"

fib$data
# gmp bigz vector of the terms given in the OEIS entry

Sequence also exposes data_raw, offset, comment, reference, formula, example, maple, mathematica, program, xref, references, revision, m_id, n_id, time, created, link (OEIS link field reformatted to Markdown), and bfile (an embedded BFile).

B-files

B-files contain more terms than the main entry (often thousands).

primes <- BFile("A000040")
get_bfile_data(primes)[1:10]
get_bfile_indices(primes)[1:10]

plot_data(primes, n = 1000, color = "royalblue")

plot_data() accepts plot_style = "line" | "joined" | "scatter", p to layer onto an existing ggplot object, and return_plot = TRUE to get the object back instead of (or in addition to) printing it. When a term exceeds double-precision range, the plot falls back to a signed log10(|value|) axis rather than truncating or erroring.

create_bfile("A999999", data = c(1, 1, 2, 3, 5, 8), offset = 0, output_path = tempdir())
# [1] "/tmp/RtmpXXXX/b999999.txt"

Cross-references and keywords

get_xref_ids(fib)
# [1] "A000032" "A000071" "A000108" ...

oeis_keyword_description("core")
# [1] "A fundamental sequence."

Graph image

png_bytes <- get_graph_png(fib)  # cached after first call
get_graph_image(fib)             # displays inline under IRkernel; else returns raw bytes

BibTeX citation

cat(get_bibtex(fib))
# @misc{A000045,
#   author       = {N. J. A. Sloane},
#   title        = {A000045: Fibonacci numbers: F(n) = F(n-1) + F(n-2) with F(0) = 0 and F(1) = 1.},
#   howpublished = {The {O}n-{L}ine {E}ncyclopedia of {I}nteger {S}equences},
#   year         = {1991},
#   month        = apr,
#   day          = {30},
#   date         = {1991-04-30},
#   url          = {https://oeis.org/A000045}
# }

API summary

Function	Applies to	Description
Sequence(oeis_id)	—	Fetch and parse an OEIS entry from the JSON API
BFile(oeis_id)	—	Fetch and parse a b-file
create_bfile(oeis_id, data, offset, output_path)	—	Write a b-file to disk
check_id(oeis_id)	—	Validate an OEIS id (^A\d{6}$)
oeis_url(oeis_id, fmt)	—	Build an OEIS URL (json, text, bfile, graph)
oeis_bfile(oeis_id)	—	Build a b-file filename
oeis_keyword_description(tag)	—	Look up an OEIS keyword tag description
extract_oeis_ids(text)	—	Extract A\d{6} ids from text
get_bfile_data, get_bfile_indices, get_filename, get_url	BFile	Accessors
get_bfile_info, get_xref_ids, get_data_values, get_keyword_description, get_graph_png, get_graph_image, get_bibtex	Sequence	Derived metadata
plot_data(x, ...)	BFile	ggplot2 line/scatter plot
plot(x, ...)	Sequence	Delegates to plot_data on the embedded BFile

Requirements

• R (>= 4.1.0)
• httr2
• jsonlite
• ggplot2
• gmp

Citation

citation("oeis.tools")

A CITATION.cff file is also provided (GitHub's "Cite this repository" button).