Upgrading to v3

Bundled datasets removed

What changed: The cup_series, xfinity_series, and truck_series datasets are no longer bundled with the package. All data is now served from Cloudflare R2.

Before (v2.x):

data("cup_series")
data("xfinity_series")
data("truck_series")

After (v3.0.0):

cup <- load_series("cup")
nxs <- load_series("nxs")
truck <- load_series("truck")

Why: Bundling large datasets in the package made installation slow and increased package size. Serving data from R2 with local caching provides faster updates and smaller package downloads.

Series identifier: “xfinity” renamed to “nxs”

What changed: The second-tier series identifier changed from "xfinity" to "nxs".

Before (v2.x):

xfinity <- load_series("xfinity")
get_driver_info("bell", series = "xfinity")

After (v3.0.0):

nxs <- load_series("nxs")
get_driver_info("bell", series = "nxs")

Why: The second-tier NASCAR series has changed title sponsors four times: Busch (1984-2007), Nationwide (2008-2014), Xfinity (2015-2025), and O’Reilly Auto Parts (2026-present). “NXS” is NASCAR’s own sponsor-neutral abbreviation for the series, so it will never go stale with future sponsor changes.

Note: The Series column in returned data now shows "NXS" instead of "Xfinity". Update any code that filters on Series == "Xfinity".

arrow package now required

What changed: The arrow package moved from Suggests to Imports. It is now a required dependency.

Why: All data is stored as Parquet files on R2. The arrow package is required to read this data efficiently.

Action required: None. The arrow package will be installed automatically when you install nascaR.data v3.0.0.

New caching system

What changed: v3.0.0 introduces a two-tier caching system (memory + disk).

How it works:

1. First call: Downloads data from R2 and caches in memory and on disk
2. Subsequent calls in same session: Instant retrieval from memory cache
3. Subsequent calls in new sessions: Fast retrieval from disk cache

Disk cache location: Uses CRAN-approved tools::R_user_dir() path:

# View cache location
tools::R_user_dir("nascaR.data", which = "cache")

Managing the cache:

# Force fresh download
cup <- load_series("cup", refresh = TRUE)

# Clear all cached data (memory + disk)
clear_cache()

Why: Caching dramatically improves performance. After the first download, data loads instantly without network requests.

find_*() functions removed

What changed: find_driver(), find_team(), and find_manufacturer() have been removed.

Before (v2.x):

find_driver("bell")
get_driver_info("Christopher Bell")

After (v3.0.0):

# Just use get_*_info() directly
get_driver_info("bell")

Why: The get_*_info() functions already include fuzzy matching and return actual data. The find_*() functions were redundant.

Summary of required changes

Data unchanged

The underlying race data is unchanged. Only the delivery mechanism and series identifier have been updated. All historical results remain available:

• Cup Series: 1949-present
• NXS: 1982-present
• Truck Series: 1995-present