Merge pull request #95 from ropensci-review-tools/cache

add 'pkgmatch_update_cache' fn for #92

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: pkgmatch
 Title:  Find R Packages Matching Either Descriptions or Other R Packages
-Version: 0.4.3.001
+Version: 0.4.3.002
 Authors@R: c(
     person("Mark", "Padgham", , "[email protected]", role = c("aut", "cre"),
            comment = c(ORCID = "0000-0003-2172-5265")),

NAMESPACE

@@ -13,6 +13,7 @@ export(pkgmatch_load_data)
 export(pkgmatch_similar_fns)
 export(pkgmatch_similar_pkgs)
 export(pkgmatch_treesitter_fn_tags)
+export(pkgmatch_update_cache)
 export(pkgmatch_update_data)
 export(set_ollama_url)
 export(text_is_code)

R/cache.R

@@ -31,8 +31,10 @@
 #' idfs <- pkgmatch_load_data ("idfs")
 #' idfs_fns <- pkgmatch_load_data ("idfs", fns = TRUE)
 #' }
-pkgmatch_load_data <- function (what = "embeddings", corpus = "ropensci",
-                                fns = FALSE, raw = FALSE) {
+pkgmatch_load_data <- function (what = "embeddings",
+                                corpus = "ropensci",
+                                fns = FALSE,
+                                raw = FALSE) {
 
     checkmate::assert_character (what, len = 1L)
     checkmate::assert_character (corpus, len = 1L)
@@ -55,6 +57,43 @@ pkgmatch_cache_update_interval <- function () {
     return (dt)
 }
 
+# nocov start
+
+#' Update all locally-cached `pkgmatch` data to latest versions.
+#'
+#' Locally-cached data used by this package are generally updated every 30
+#' days, or according to `options("pkgmatch_update_frequency")` if that is set.
+#' This function forces all locally-cached data to be updated, regardless of
+#' update frequencies. Remote data are provided on the latest release of GitHub
+#' repository at
+#' \url{https://github.com/ropensci-review-tools/pkgmatch/releases}.
+#'
+#' @examples
+#' \dontrun{
+#' pkgmatch_update_cache()
+#' }
+#' @export
+pkgmatch_update_cache <- function () {
+
+    what <- c ("embeddings", "idfs", "functions", "calls")
+    corpus <- c ("ropensci", "cran")
+    fns <- c (FALSE, TRUE)
+    raw <- c (FALSE, TRUE)
+
+    vals <- expand.grid (what = what, corpus = corpus, fns = fns, raw = raw)
+    vals$fname <- apply (vals, 1, function (i) {
+        get_cache_file_name (what = i [1], corpus = i [2], fns = i [3], raw = i [4])
+    })
+    vals <- vals [-which (duplicated (vals$fname)), ]
+
+    files <- apply (vals, 1, function (i) {
+        pkgmatch_dl_data (what = i [1], corpus = i [2], fns = i [3], raw = i [4])
+    })
+
+    invisible (files)
+}
+# nocov end
+
 load_data_internal <- function (what, corpus, fns, raw) {
     fname <- get_cache_file_name (what, corpus, fns, raw)
 

R/data-update.R

@@ -10,6 +10,12 @@ RELEASE_TAG <- "v0.4.0"
 #' "pkgmatch-results" of R's current temporary directory. This updating may
 #' take a very long time!
 #'
+#' Note that this function is categorically different from
+#' \link{pkgmatch_update_cache}. This function updates the internal data used
+#' by the `pkgmatch` package, and should only ever be run by package
+#' maintainers. The \link{pkgmatch_update_cache} downloads the latest versions
+#' of these data to a local cache for use in this package.
+#'
 #' @param upload If `TRUE`, upload updated results to GitHub release.
 #' @return Local path to directory containing updated results.
 #' @family data

codemeta.json

@@ -8,7 +8,7 @@
   "codeRepository": "https://github.com/ropensci-review-tools/pkgmatch",
   "issueTracker": "https://github.com/ropensci-review-tools/pkgmatch/issues",
   "license": "https://spdx.org/licenses/MIT",
-  "version": "0.4.3.001",
+  "version": "0.4.3.002",
   "programmingLanguage": {
     "@type": "ComputerLanguage",
     "name": "R",