r2rtf
The clinical-report table engine learned Chinese, then learned to leave RTF entirely
A side-by-side editorial comparison of sofa and tidymodels — release velocity, themes, recent moves, and the top alternatives to consider.
A CouchDB client for R whose recent work is all test infrastructure, not new routes.
sofa wraps the CouchDB HTTP API for R — database and document CRUD, Mango queries and indexes, design documents, replication, and attachments, organized around a Cushion connection object. Feature development effectively stopped after 0.4.0 brought CouchDB v3 compatibility in 2020. The two 2026 releases are a testing and packaging overhaul: 0.4.1 rebuilt the suite around a crul-mocked fake CouchDB so tests no longer need Cloudant or a local Docker instance, reaching 100% coverage, and 0.4.2 cleared CRAN check notes.
The meta-package ships almost nothing, which is exactly what a version-pinning shim should do
The tidymodels package is a loader and version pin for the modeling framework's core set rather than a place where features live. Its entire changelog consists of updated dependency versions, adjustments to how tidymodels_prefer() resolves name conflicts against other packages, and the occasional addition of a package to the core set — workflowsets in 0.1.3, tailor in 1.4.0. The most recent releases moved the package's own code from the magrittr pipe to R's base pipe and patched a bug where some attached packages were omitted.
sofa wraps the CouchDB HTTP API for R — database and document CRUD, Mango queries and indexes, design documents, replication, and attachments, organized around a Cushion connection object. Feature development effectively stopped after 0.4.0 brought CouchDB v3 compatibility in 2020. The two 2026 releases are a testing and packaging overhaul: 0.4.1 rebuilt the suite around a crul-mocked fake CouchDB so tests no longer need Cloudant or a local Docker instance, reaching 100% coverage, and 0.4.2 cleared CRAN check notes.
This is a rOpenSci package being brought back to a maintainable state rather than extended. The 0.4.1 decision to mock the server is the consequential one — it decouples CI from a live CouchDB, which is what made the cross-platform GitHub Actions workflow and full coverage possible at all. Two genuine bugs surfaced in that process, in db_replicate() URL construction against non-Cloudant servers and db_alldocs(disk=) not returning the documented file path, suggesting the untested paths had drifted.
With the test harness now self-contained, the next plausible move is catching up on CouchDB routes added since v3 rather than further infrastructure work, though nothing in these entries commits to it.
The tidymodels package is a loader and version pin for the modeling framework's core set rather than a place where features live. Its entire changelog consists of updated dependency versions, adjustments to how tidymodels_prefer() resolves name conflicts against other packages, and the occasional addition of a package to the core set — workflowsets in 0.1.3, tailor in 1.4.0. The most recent releases moved the package's own code from the magrittr pipe to R's base pipe and patched a bug where some attached packages were omitted.
Release cadence tracks the ecosystem rather than any roadmap of its own: a version bump when member packages release, a tidymodels_prefer() rule when a new conflict appears — DALEX::explains() over dplyr::explains(), recipes::update() over other update() methods. Additions to the core set are the only structurally interesting events, and there have been two in seven releases. Everything else is plumbing that exists so a single library() call attaches a consistent set of versions.
The next release will most likely be another version-set update, with any new core package the only thing worth noting. Feature news for this framework will keep arriving in the member packages, not here.
Other Analytics products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either sofa or tidymodels.
The clinical-report table engine learned Chinese, then learned to leave RTF entirely
New stewardship at openpharma, then two releases adding the methods MCP-Mod was missing
The stubbing library added httr2 support, then spent a year cutting itself free of everything else
crul took mocking back from webmockr and made it a property of the client itself
Six releases, six identical bodies — the feed carries the package abstract instead of release notes
chattr deleted every LLM integration it had written and outsourced the lot to ellmer
See all sofa alternatives → · See all tidymodels alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. sofa is currently shipping more aggressively (velocity 2.5 vs 0.0), with 0 editorial sparks in the last 30 days against 0. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. sofa is currently shipping more aggressively (velocity 2.5 vs 0.0), with 0 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Analytics products to evaluate alongside.
Top sofa alternatives in Analytics are ranked by recent ship velocity. Browse the "sofa alternatives" section above for the current picks, or visit /alternatives/sofa for the full list with editorial commentary on each.
Top tidymodels alternatives in Analytics are ranked by recent ship velocity. Browse the "tidymodels alternatives" section above for the current picks, or visit /alternatives/tidymodels for the full list with editorial commentary on each.