Document and enforce some properties for strings created by modules.

When creating multibyte or unibyte strings, we should guarantee the
following invariants:

- When creating empty strings, a NULL data pointer should be allowed.
  This often arises in practice if the string length isn't known in
  advance, and we don't want to unnecessarily trigger undefined
  behavior.  Since functions like memcpy might not accept NULL
  pointers, use the canonical empty string objects in this case.

- Nonzero strings should be guaranteed to be unique and mutable.
  These are the same guarantees expected from Lisp functions such as
  'make-string' or 'unibyte-string'.  On the other hand, empty strings
  might not be unique.

* src/emacs-module.c (module_make_string)
(module_make_unibyte_string): Correctly handle empty strings.

* test/src/emacs-module-resources/mod-test.c (Fmod_test_make_string):
New test function.
(emacs_module_init): Expose it.

* test/src/emacs-module-tests.el (mod-test-make-string/empty)
(mod-test-make-string/nonempty): New unit tests.

* doc/lispref/internals.texi (Module Values): Document properties and
corner cases for strings.

2 files changed, 50 insertions, 0 deletions

diff --git a/test/src/emacs-module-resources/mod-test.c b/test/src/emacs-module-resources/mod-test.c
index f855e9b6da0..30ad352cf8b 100644
--- a/test/src/emacs-module-resources/mod-test.c
+++ b/test/src/emacs-module-resources/mod-test.c
@@ -24,6 +24,7 @@ along with GNU Emacs.  If not, see <https://www.gnu.org/licenses/>.  */
 
 #include <errno.h>
 #include <limits.h>
+#include <stdbool.h>
 #include <stdint.h>
 #include <stdio.h>
 #include <stdlib.h>
@@ -699,6 +700,34 @@ Fmod_test_funcall (emacs_env *env, ptrdiff_t nargs, emacs_value *args,
   return env->funcall (env, args[0], nargs - 1, args + 1);
 }
 
+static emacs_value
+Fmod_test_make_string (emacs_env *env, ptrdiff_t nargs,
+                       emacs_value *args, void *data)
+{
+  assert (nargs == 2);
+  intmax_t length_arg = env->extract_integer (env, args[0]);
+  if (env->non_local_exit_check (env) != emacs_funcall_exit_return)
+    return args[0];
+  if (length_arg < 0 || SIZE_MAX < length_arg)
+    {
+      signal_error (env, "Invalid string length");
+      return args[0];
+    }
+  size_t length = (size_t) length_arg;
+  bool multibyte = env->is_not_nil (env, args[1]);
+  char *buffer = length == 0 ? NULL : malloc (length);
+  if (buffer == NULL && length != 0)
+    {
+      memory_full (env);
+      return args[0];
+    }
+  memset (buffer, 'a', length);
+  emacs_value ret = multibyte ? env->make_string (env, buffer, length)
+                              : env->make_unibyte_string (env, buffer, length);
+  free (buffer);
+  return ret;
+}
+
 /* Lisp utilities for easier readability (simple wrappers).  */
 
 /* Provide FEATURE to Emacs.  */
@@ -790,6 +819,7 @@ emacs_module_init (struct emacs_runtime *ert)
   DEFUN ("mod-test-async-pipe", Fmod_test_async_pipe, 1, 1, NULL, NULL);
   DEFUN ("mod-test-funcall", Fmod_test_funcall, 1, emacs_variadic_function,
          NULL, NULL);
+  DEFUN ("mod-test-make-string", Fmod_test_make_string, 2, 2, NULL, NULL);
 
 #undef DEFUN
 
diff --git a/test/src/emacs-module-tests.el b/test/src/emacs-module-tests.el
index 99d4cafb4af..bf26ffb935c 100644
--- a/test/src/emacs-module-tests.el
+++ b/test/src/emacs-module-tests.el
@@ -30,6 +30,7 @@
 (require 'ert)
 (require 'ert-x)
 (require 'help-fns)
+(require 'subr-x)
 
 (defconst mod-test-emacs
   (expand-file-name invocation-name invocation-directory)
@@ -556,4 +557,23 @@ See Bug#36226."
     (thread-join thread-1)
     (thread-join thread-2)))
 
+(ert-deftest mod-test-make-string/empty ()
+  (dolist (multibyte '(nil t))
+    (ert-info ((format "Multibyte: %s" multibyte))
+      (let ((got (mod-test-make-string 0 multibyte)))
+        (should (stringp got))
+        (should (string-empty-p got))
+        (should (eq (multibyte-string-p got) multibyte))))))
+
+(ert-deftest mod-test-make-string/nonempty ()
+  (dolist (multibyte '(nil t))
+    (ert-info ((format "Multibyte: %s" multibyte))
+      (let ((first (mod-test-make-string 1 multibyte))
+            (second (mod-test-make-string 1 multibyte)))
+        (should (stringp first))
+        (should (eql (length first) 1))
+        (should (eq (multibyte-string-p first) multibyte))
+        (should (string-equal first second))
+        (should-not (eq first second))))))
+
 ;;; emacs-module-tests.el ends here