🛡️ Sentinel: [HIGH] Fix readline integer coercion DoS vulnerability

.github/workflows/r.yml

@@ -38,6 +38,8 @@ jobs:
       - name: Set up R package dependencies
         uses: r-lib/actions/setup-r-dependencies@d3c5be51b12e724e68f33216ca3c148b66d5f0b6
         with:
+          # Refresh stale qs2/RcppParallel TBB ABI artifacts as one cache generation.
+          cache-version: '2'
           extra-packages: any::rcmdcheck
           needs: check
 

.jules/sentinel.md

@@ -2,3 +2,7 @@
 **Vulnerability:** Unvalidated inputs passed to `if()` statements can cause process crashes (`condition has length > 1`) or unexpected coercion vulnerabilities.
 **Learning:** In R, optional boolean parameters that default to `NULL` should be validated using explicit runtime type validation (e.g., `if (!is.null(flag) && (!is.logical(flag) || length(flag) != 1 || is.na(flag)))`).
 **Prevention:** Always implement explicit runtime type validation for optional boolean parameters.
+## 2024-07-16 - Fix readline integer coercion DoS vulnerability
+**Vulnerability:** Weak regex `^[0-9]+$` for integer validation allows extremely large numeric strings, which coerce to `NA` via `as.integer()` and crash the R process when evaluated in `if` conditions.
+**Learning:** When reading integer inputs via `readline()` in R, avoid weak regex validation. Large numbers coerce to `NA` breaking control flow logic.
+**Prevention:** Use strictly bounded exact-match regex like `^[12]$` when prompting for specific integer options to prevent unexpected coercions and DoS vulnerabilities.

DESCRIPTION

@@ -10,7 +10,9 @@ Description: Automates fixed item parameter linking for test linking under
     the item response theory paradigm using mirt package estimates.
 License: GPL-3 | file LICENSE
 Imports: mirt, methods
-Suggests: testthat (>= 3.0.0)
+Suggests:
+    mockery,
+    testthat (>= 3.0.0)
 Encoding: UTF-8
 Config/testthat/edition: 3
 Config/roxygen2/version: 8.0.0

R/aFIPC.R

@@ -141,7 +141,7 @@ autoFIPC <-
       }
       for (attempt in seq_len(3)) {
         n <- readline(prompt = "Is it correct? (1: Yes 2: No) : ")
-        if (grepl("^[0-9]+$", n)) {
+        if (grepl("^[12]$", n)) {
           return(as.integer(n))
         }
       }
@@ -171,7 +171,7 @@ autoFIPC <-
               readline(
                 prompt = "Do you want to use default BILOG-MG priors for oldform Data? (1: Yes 2: No) : "
               )
-            if (grepl("^[0-9]+$", n)) {
+            if (grepl("^[12]$", n)) {
               return(as.integer(n))
             }
           }
@@ -390,7 +390,7 @@ autoFIPC <-
               readline(
                 prompt = "Do you want to use default BILOG-MG priors for newform Data? (1: Yes 2: No) : "
               )
-            if (grepl("^[0-9]+$", n)) {
+            if (grepl("^[12]$", n)) {
               return(as.integer(n))
             }
           }

tests/testthat/test-sentinel-validation.R

@@ -35,3 +35,89 @@ test_that("autoFIPC validates boolean flags for newformBILOGprior, oldformBILOGp
     "Security Error: confirmCommonItems must be a single non-NA logical value or NULL"
   )
 })
+
+test_that("interactive readline input bounded regex validation prevents DoS", {
+  mockery::stub(aFIPC::autoFIPC, 'interactive', TRUE)
+  my_readline <- function(...) '99999999999999999999'
+  mockery::stub(aFIPC::autoFIPC, 'readline', my_readline)
+
+  # This should error out from 'Too many invalid common item confirmation attempts'
+  # instead of failing due to NA coercion in if condition
+  expect_error(
+    aFIPC::autoFIPC(
+      newformXData = data.frame(A=1),
+      oldformYData = data.frame(A=2),
+      newformCommonItemNames = c('A'),
+      oldformCommonItemNames = c('A')
+    ),
+    "Too many invalid common item confirmation attempts"
+  )
+})
+
+
+test_that("oldform BILOG prompt rejects repeated overlong integer input", {
+  mockery::stub(aFIPC::autoFIPC, "interactive", TRUE)
+  calls <- new.env(parent = emptyenv())
+  calls$count <- 0L
+  overlong_readline <- function(...) {
+    calls$count <- calls$count + 1L
+    "99999999999999999999"
+  }
+  mockery::stub(aFIPC::autoFIPC, "readline", overlong_readline)
+
+  expect_error(
+    aFIPC::autoFIPC(
+      newformXData = data.frame(A = 1),
+      oldformYData = data.frame(A = 2),
+      newformCommonItemNames = "A",
+      oldformCommonItemNames = "A",
+      newformBILOGprior = FALSE,
+      confirmCommonItems = TRUE
+    ),
+    "Too many invalid oldform BILOG prior attempts"
+  )
+  expect_equal(calls$count, 3L)
+})
+
+test_that("newform BILOG prompt rejects repeated overlong integer input", {
+  skip_if_not_installed("mirt")
+  set.seed(20260815)
+  old_data <- as.data.frame(mirt::simdata(
+    a = matrix(rep(1, 4), ncol = 1),
+    d = c(-1, -0.3, 0.3, 1),
+    itemtype = rep("2PL", 4),
+    N = 200
+  ))
+  names(old_data) <- LETTERS[1:4]
+  old_model <- mirt::mirt(
+    old_data,
+    1,
+    itemtype = "2PL",
+    SE = FALSE,
+    verbose = FALSE
+  )
+
+  mockery::stub(aFIPC::autoFIPC, "interactive", TRUE)
+  calls <- new.env(parent = emptyenv())
+  calls$count <- 0L
+  overlong_readline <- function(...) {
+    calls$count <- calls$count + 1L
+    "99999999999999999999"
+  }
+  mockery::stub(aFIPC::autoFIPC, "readline", overlong_readline)
+
+  expect_error(
+    aFIPC::autoFIPC(
+      newformXData = old_data,
+      oldformYData = old_model,
+      newformCommonItemNames = "A",
+      oldformCommonItemNames = "A",
+      itemtype = "3PL",
+      oldformBILOGprior = FALSE,
+      confirmCommonItems = TRUE,
+      checkIPD = FALSE
+    ),
+    "Too many invalid newform BILOG prior attempts"
+  )
+  expect_equal(calls$count, 3L)
+})

tests/testthat/test-workflow-cache.R

@@ -0,0 +1,29 @@
+test_that("R CMD check uses an active reviewed dependency cache generation", {
+  repository_root <- Sys.getenv(
+    "GITHUB_WORKSPACE",
+    unset = testthat::test_path("..", "..")
+  )
+  workflow_path <- file.path(repository_root, ".github", "workflows", "r.yml")
+  workflow_lines <- readLines(workflow_path, warn = FALSE)
+  dependency_step <- grep(
+    "r-lib/actions/setup-r-dependencies@",
+    workflow_lines,
+    fixed = TRUE
+  )
+  expect_length(dependency_step, 1L)
+  dependency_block <- workflow_lines[
+    dependency_step:min(dependency_step + 8L, length(workflow_lines))
+  ]
+  active_cache_version_pattern <- paste0(
+    "^[[:space:]]*cache-version:[[:space:]]*",
+    "['\\\"]2['\\\"][[:space:]]*(#.*)?$"
+  )
+
+  expect_false(
+    grepl(active_cache_version_pattern, "# cache-version: '2'", perl = TRUE)
+  )
+  expect_true(
+    any(grepl(active_cache_version_pattern, dependency_block, perl = TRUE)),
+    info = "The TBB ABI cache refresh must remain an active workflow input"
+  )
+})