perf(build-newer-r): skip packages that already have a binary in bulk

Decide the already-built candidates from one listing per slot and one ABI-verdict query instead of a DB connection and several S3 requests per package; anything undecidable still goes through bincraft.

local/build-newer-r.R

@@ -27,6 +27,94 @@ message(sprintf(
 if ("--dry-run" %in% args) {
   print(pkgs, row.names = FALSE)
 } else {
+  # Most candidates were built on an earlier run. Asking bincraft about each one
+  # costs a DB connection plus several S3 requests, about 4s a package, so the
+  # loop spent well over an hour per run re-confirming existing binaries.
+  # Decide those in bulk first; bincraft still checks whatever is left.
+  local_machine <- Sys.info()[["machine"]]
+  arch <- if (grepl("arm64|aarch64", local_machine)) "arm64" else "amd64"
+  slot_dir <- sprintf(
+    "devxy-rpkgs-binaries/%s/%s/latest/src/contrib",
+    arch,
+    bincraft::set_codename(NULL)
+  )
+  s3fs::s3_file_system(
+    aws_access_key_id = Sys.getenv("B2_S3_ACCESS_KEY"),
+    aws_secret_access_key = Sys.getenv("B2_S3_SECRET_KEY"),
+    endpoint = "https://s3.eu-central-003.backblazeb2.com",
+    region_name = "eu-central-003",
+    refresh = TRUE
+  )
+  # NULL when the listing fails, so that route falls back to bincraft.
+  list_etags <- function(dir) {
+    info <- tryCatch(s3fs::s3_dir_info(dir), error = function(e) {
+      message(sprintf("Could not list %s: %s", dir, conditionMessage(e)))
+      NULL
+    })
+    if (is.null(info)) {
+      return(NULL)
+    }
+    stats::setNames(
+      gsub('^"|"$', "", as.character(info$etag)),
+      basename(as.character(info$uri))
+    )
+  }
+  # The verdicts bincraft::classify_r_minor_sensitive() would read from its
+  # cache, fetched in one query rather than one connection per package.
+  read_verdicts <- function(packages) {
+    tryCatch(
+      {
+        con <- DBI::dbConnect(
+          RPostgres::Postgres(),
+          dbname = "build_metadata",
+          host = "r-binaries.devxy.io",
+          port = 15432,
+          user = "rpkgs",
+          password = Sys.getenv("PGPASS"),
+          sslmode = "verify-full"
+        )
+        on.exit(DBI::dbDisconnect(con), add = TRUE)
+        rows <- DBI::dbGetQuery(
+          con,
+          paste(
+            "SELECT package, version, r_minor_sensitive FROM abi_classification",
+            "WHERE classifier_sig = $1 AND package = ANY($2::text[])"
+          ),
+          params = list(
+            bincraft:::abi_classifier_signature(),
+            paste0("{", paste(packages, collapse = ","), "}")
+          )
+        )
+        stats::setNames(
+          as.logical(rows$r_minor_sensitive),
+          paste(rows$package, rows$version, sep = "_")
+        )
+      },
+      error = function(e) {
+        message(sprintf("Could not read ABI verdicts: %s", conditionMessage(e)))
+        logical()
+      }
+    )
+  }
+  sensitive <- read_verdicts(pkgs$Package)
+  before <- nrow(pkgs)
+  pkgs <- newer_r_outstanding(
+    pkgs,
+    sensitive,
+    list_etags(slot_dir),
+    list_etags(file.path(
+      slot_dir,
+      sub("^(\\d+\\.\\d+).*", "\\1", target_r)
+    )),
+    bincraft:::cran_source_md5()
+  )
+  message(sprintf(
+    "%d of %d already carry a binary at their routed path; %d left for bincraft",
+    before - nrow(pkgs),
+    before,
+    nrow(pkgs)
+  ))
+
   options(crayon.enabled = TRUE, Ncpus = 4, future.globals.onReference = NULL)
   for (i in seq_len(nrow(pkgs))) {
     message(sprintf(