Changelog • olympicAthletes

olympicAthletes 0.5.9

Replace the \dontrun{} block in the olympic_athletes example with @examplesIf, so the dplyr/tidyr example runs whenever those Suggests packages are installed (per CRAN policy on runnable examples).
citation("olympicAthletes") now returns all three entries the README promises: the package plus its two upstream data sources (the rgriff23 ‘Olympic_history’ dataset and Olympedia).
Refresh the README and vignette dataset listings to cover all 14 shipped datasets (volleyball_athletes, curling_athletes, and season_counts were missing).
Fix the PDF manual build on CRAN’s win-builder (R-devel): the editions and olympic_athletes docs cited native-script host-city names (Cyrillic, CJK, and Hangul examples such as the native forms of Moskva, Tokyo, Beijing, and PyeongChang) that pdflatex cannot typeset. The docs now list only the romanized forms; Latin-1 diacritics (e.g. “Athína”, “München”) are retained.

CRAN preparation. No data changes — every dataset is retained. Brought the source tarball back under CRAN’s 5 MB limit (5.13 MB -> 4.80 MB) by resizing the oversized man/figures/logo.png (885x1024, 512 KB -> 480 px, 184 KB); the full-resolution master is kept at the repo-root hex.png.
Added Language: en-US and an inst/WORDLIST so spelling::spell_check_package() runs clean (the flagged words were all proper nouns and British spellings, not typos).
Wrapped the Olympedia URL in DESCRIPTION with angle brackets and added a cran-comments.md for submission.
R CMD check --as-cran now reports 0 errors, 0 warnings, and only the routine new-submission / CC-BY-license NOTE.

Add season_counts (2 rows): athlete-event row counts per Olympic season, pre-counted so plotting exercises can contrast geom_bar() (counts raw rows) with geom_col() (plots pre-counted values) without base-R counting code.

Removed olympic_athletes_2000_2026 (1.4 MB — the second-largest file in the package) to bring the built package back under CRAN’s 5 MB source-package guideline. It was a pure subset of olympic_athletes (year >= 2000); recreate it any time with olympic_athletes |> filter(year >= 2000). The ModernDive exercises that used it now use team_sport_athletes, athletics_athletes, or an explicit filter().
Removed olympic_athletes_1960 (referenced nowhere in the book, exercises, or teaching materials); recreate with olympic_athletes |> filter(year == 1960, season == "Summer").

Add three exercise-support subsets so ModernDive Chapter 2 exercises (which predate dplyr) never need base-R subsetting workarounds: usa_summer_medals (30 rows — one per Summer Games the USA attended; note the boycotted 1980 Games have no row), art_competitions_athletes (3,578 rows — the 1912-1948 Art Competitions entrants), and team_sport_athletes (10,912 rows — Basketball, Volleyball, and Curling combined, for cross-group distribution comparisons).

Renamed recent_olympic_athletes to olympic_athletes_2000_2026 for a clearer, time-explicit name. The data is unchanged (year >= 2000, i.e. the 2000–2026 Games).

Add volleyball_athletes (4,792 rows) and curling_athletes (961 rows) convenience subsets, parallel to the existing basketball_athletes / gymnastics_athletes, for ModernDive Chapter 5+ regression examples.

Recovered missing noc/team (country) values: rows where the country was unknown but the same athlete (id) appears in another Games with a known country are now backfilled (unambiguous cases only). Reduces NA noc from 10,094 to 3,907 rows; the rest are athletes who never appear with any known country, or who competed for more than one NOC, and are deliberately left NA.

New dataset paris_2024_top_medals: the ten countries that won the most total medals at the Paris 2024 Summer Games, in long format (one row per country-and-medal-type: country, noc, medal, count). A small, tidy dataset built for barplot examples (stacked, dodged, proportional) without filtering the ~90 medal-winning nations in medal_table.

New single-Games convenience datasets: olympic_athletes_2024 (Paris 2024 Summer Games) and olympic_athletes_1960 (Rome 1960 Summer Games). The 2024 subset is small enough to show over-plotting in a height vs weight scatterplot without the full dataset’s scale; paired with the near-complete 1960 subset it supports generational comparisons of athlete physiques. Both have columns identical to olympic_athletes.

New convenience datasets: athletics_athletes, gymnastics_athletes, and basketball_athletes (single-sport subsets of olympic_athletes), plus recent_olympic_athletes (Games since 2000, year >= 2000). These let teaching examples focus on one sport or recent Games without a filter(). Columns are identical to olympic_athletes.
olympic_athletes bio coverage substantially improved for the five olympedia-scraped editions (2018-2026). Team-event and multi-athlete-row athletes (Ice Hockey, Curling, Bobsleigh, Football, Hockey, Rowing, relays, doubles, pairs, …) previously had no age and mostly no height/weight; these bios have now been fetched from olympedia. age is now ~100% complete across all five editions, and height/ weight coverage rose accordingly (e.g. 2018 Winter height ~81% -> ~96%, Paris 2024 ~24% -> ~34%). The refresh is purely additive — row count, athlete roster, and medal counts are unchanged.
editions: Tokyo 2020 participants corrected 11,319 -> 11,318 to match olympedia’s revised count.

Breaking: all olympic_athletes columns are now lowercase snake_case (id, name, sex, age, height, weight, team, noc, games, year, season, sport, event, medal), matching the convention already used by medal_table and editions. The upstream rgriff23 TitleCase headers (ID, Name, NOC, …) are no longer preserved.
Host-city columns now come in both local-language and English forms. In olympic_athletes, the City column has been replaced by city_local_latin and city_english. In editions, city has been replaced by city_local_latin and city_english. city_local_latin always uses the Latin alphabet — Latin-script names keep their native diacritics ("Athína", "München", "Montréal", "Ciudad de México") and non-Latin-script names use a standard romanization ("Moskva", "Tokyo", "Beijing", "PyeongChang") rather than the city’s native script.

medal_table now covers every Olympic edition from Athens 1896 through Milano-Cortina 2026 (1,929 rows, up from 273). Two new columns: edition_id (Olympedia’s internal edition number) and notes.
editions now covers every Olympic edition from Athens 1896 through Milano-Cortina 2026, including cancelled Games (62 rows, up from 5).

Initial release.

olympic_athletes — 314,749 athlete-event rows spanning Athens 1896 to Milano-Cortina 2026, extending the rgriff23 Olympic_history dataset (1896-2016) with five additional editions scraped from olympedia.org.
medal_table — 273-row verified medal table for the five new editions (2018, 2020, 2022, 2024, 2026), with one row per (Games, NOC).
editions — 5-row metadata table covering opening/closing dates, participants, NOC count, medal events, and disciplines for each new edition.