master 368d2531be: Fix 'treesit-max-buffer-size' and its use

[Eli Zaretskii]

branch: master
commit 368d2531bec84c2e75b2fd2f32bb59fc2e7ee520
Author: Eli Zaretskii <eliz@gnu.org>
Commit: Eli Zaretskii <eliz@gnu.org>

    Fix 'treesit-max-buffer-size' and its use
    
    * lisp/treesit.el (treesit-max-buffer-size): Avoid overflow in
    computing buffer-size limit.  Account for 32-but systems built
    "--with-wide-int".  Extend doc string.
    (treesit-ready-p): Compare the limit with the size of the buffer
    in bytes, not in characters.
    
    * src/treesit.c (treesit_check_buffer_size): Measure buffer size
    in bytes.
---
 lisp/treesit.el | 16 +++++++++++-----
 src/treesit.c   |  6 +++---
 2 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/lisp/treesit.el b/lisp/treesit.el
index 10c6b09322..75e859f13f 100644
--- a/lisp/treesit.el
+++ b/lisp/treesit.el
@@ -98,10 +98,16 @@ indent, imenu, etc."
 (defcustom treesit-max-buffer-size
   (let ((mb (* 1024 1024)))
     ;; 40MB for 64-bit systems, 15 for 32-bit.
-    (if (> most-positive-fixnum (* 4 1024 mb))
-        (* 40 mb)
-      (* 15 mb)))
-  "Maximum buffer size for enabling tree-sitter parsing (in bytes)."
+    (if (or (< most-positive-fixnum (* 2.0 1024 mb))
+            ;; 32-bit system with wide ints.
+            (string-match-p "--with-wide-int" system-configuration-options))
+        (* 15 mb)
+      (* 40 mb)))
+  "Maximum buffer size (in bytes) for enabling tree-sitter parsing.
+
+A typical tree-sitter parser needs 10 times as much memory as the
+buffer it parses.  Also, the tree-sitter library has a hard limit
+of max unsigned 32-bit value for byte offsets into buffer text."
   :type 'integer
   :version "29.1")
 
@@ -1583,7 +1589,7 @@ instead of emitting a warning."
       (when (not (treesit-available-p))
         (setq msg "tree-sitter library is not compiled with Emacs")
         (throw 'term nil))
-      (when (> (buffer-size) treesit-max-buffer-size)
+      (when (> (position-bytes (1- (point-max))) treesit-max-buffer-size)
         (setq msg "buffer larger than `treesit-max-buffer-size'")
         (throw 'term nil))
       (dolist (lang language-list)
diff --git a/src/treesit.c b/src/treesit.c
index 835b07c931..4a7c401956 100644
--- a/src/treesit.c
+++ b/src/treesit.c
@@ -840,11 +840,11 @@ treesit_ensure_position_synced (Lisp_Object parser)
 static void
 treesit_check_buffer_size (struct buffer *buffer)
 {
-  ptrdiff_t buffer_size = (BUF_Z (buffer) - BUF_BEG (buffer));
-  if (buffer_size > UINT32_MAX)
+  ptrdiff_t buffer_size_bytes = (BUF_Z_BYTE (buffer) - BUF_BEG_BYTE (buffer));
+  if (buffer_size_bytes > UINT32_MAX)
     xsignal2 (Qtreesit_buffer_too_large,
              build_pure_c_string ("Buffer size cannot be larger than 4GB"),
-             make_fixnum (buffer_size));
+             make_fixnum (buffer_size_bytes));
 }
 
 static Lisp_Object treesit_make_ranges (const TSRange *, uint32_t, struct 
buffer *);