From 2024f2753484f7ec417ef6f9ce56940f3709ca9b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Florian=20M=C3=A4rkl?= <info@florianmaerkl.de>
Date: Thu, 4 Nov 2021 18:44:02 +0100
Subject: [PATCH 1/2] Make SubtreeInlineData work on Big-Endian

---
 lib/src/host.h    | 21 +++++++++++++++++
 lib/src/subtree.h | 60 +++++++++++++++++++++++++++++++++++++----------
 2 files changed, 69 insertions(+), 12 deletions(-)
 create mode 100644 lib/src/host.h

diff --git a/lib/src/host.h b/lib/src/host.h
new file mode 100644
index 00000000..a07e9f89
--- /dev/null
+++ b/lib/src/host.h
@@ -0,0 +1,21 @@
+
+// Determine endian and pointer size based on known defines.
+// TS_BIG_ENDIAN and TS_PTR_SIZE can be set as -D compiler arguments
+// to override this.
+
+#if !defined(TS_BIG_ENDIAN)
+#if (defined(__BYTE_ORDER__) && __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__) \
+  || (defined( __APPLE_CC__) && (defined(__ppc__) || defined(__ppc64__)))
+#define TS_BIG_ENDIAN 1
+#else
+#define TS_BIG_ENDIAN 0
+#endif
+#endif
+
+#if !defined(TS_PTR_SIZE)
+#if UINTPTR_MAX == 0xFFFFFFFF
+#define TS_PTR_SIZE 32
+#else
+#define TS_PTR_SIZE 64
+#endif
+#endif
diff --git a/lib/src/subtree.h b/lib/src/subtree.h
index 97dee5ea..bfbdbc41 100644
--- a/lib/src/subtree.h
+++ b/lib/src/subtree.h
@@ -11,6 +11,7 @@ extern "C" {
 #include "./length.h"
 #include "./array.h"
 #include "./error_costs.h"
+#include "./host.h"
 #include "tree_sitter/api.h"
 #include "tree_sitter/parser.h"
 
@@ -39,21 +40,56 @@ typedef struct {
 //
 // This representation is used for small leaf nodes that are not
 // errors, and were not created by an external scanner.
+// The idea behind the layout of this struct is that the `is_inline`
+// bit will fall exactly into the same location as the least significant
+// bit of the pointer in `Subtree` or `MutableSubtree`, respectively.
+// Because of alignment, for any valid pointer this will be 0, giving
+// us the opportunity to make use of this bit to signify whether to use
+// the pointer or the inline struct.
 typedef struct {
-  bool is_inline : 1;
-  bool visible : 1;
-  bool named : 1;
-  bool extra : 1;
-  bool has_changes : 1;
-  bool is_missing : 1;
+#define SUBTREE_3_BYTES \
+  uint8_t symbol; \
+  uint8_t padding_bytes; \
+  uint8_t size_bytes; \
+
+#define SUBTREE_6_BITS \
+  bool visible : 1; \
+  bool named : 1; \
+  bool extra : 1; \
+  bool has_changes : 1; \
+  bool is_missing : 1; \
   bool is_keyword : 1;
-  uint8_t symbol;
-  uint8_t padding_bytes;
-  uint8_t size_bytes;
-  uint8_t padding_columns;
-  uint8_t padding_rows : 4;
-  uint8_t lookahead_bytes : 4;
+
+#define SUBTREE_4_BYTES \
+  uint8_t padding_columns; \
+  uint8_t padding_rows : 4; \
+  uint8_t lookahead_bytes : 4; \
   uint16_t parse_state;
+
+#if TS_BIG_ENDIAN
+#if TS_PTR_SIZE == 32
+  SUBTREE_3_BYTES
+  SUBTREE_6_BITS
+  bool unused : 1;
+  bool is_inline : 1;
+  SUBTREE_4_BYTES
+#else
+  SUBTREE_4_BYTES
+  SUBTREE_3_BYTES
+  SUBTREE_6_BITS
+  bool unused : 1;
+  bool is_inline : 1;
+#endif
+#else
+  bool is_inline : 1;
+  SUBTREE_6_BITS
+  SUBTREE_3_BYTES
+  SUBTREE_4_BYTES
+#endif
+
+#undef SUBTREE_3_BYTES
+#undef SUBTREE_6_BITS
+#undef SUBTREE_4_BYTES
 } SubtreeInlineData;
 
 // A heap-allocated representation of a subtree.

From d5d99e0bfb552dd2e8b32ef7918c13d208aa8075 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Florian=20M=C3=A4rkl?= <info@florianmaerkl.de>
Date: Fri, 24 Dec 2021 17:07:32 +0100
Subject: [PATCH 2/2] Address feedback

---
 lib/src/subtree.h | 36 +++++++++++++++++++-----------------
 1 file changed, 19 insertions(+), 17 deletions(-)

diff --git a/lib/src/subtree.h b/lib/src/subtree.h
index bfbdbc41..992d31c5 100644
--- a/lib/src/subtree.h
+++ b/lib/src/subtree.h
@@ -49,10 +49,9 @@ typedef struct {
 typedef struct {
 #define SUBTREE_3_BYTES \
   uint8_t symbol; \
-  uint8_t padding_bytes; \
-  uint8_t size_bytes; \
+  uint16_t parse_state;
 
-#define SUBTREE_6_BITS \
+#define SUBTREE_BITS \
   bool visible : 1; \
   bool named : 1; \
   bool extra : 1; \
@@ -60,36 +59,39 @@ typedef struct {
   bool is_missing : 1; \
   bool is_keyword : 1;
 
-#define SUBTREE_4_BYTES \
+#define SUBTREE_SIZE \
   uint8_t padding_columns; \
   uint8_t padding_rows : 4; \
   uint8_t lookahead_bytes : 4; \
-  uint16_t parse_state;
+  uint8_t padding_bytes; \
+  uint8_t size_bytes;
 
 #if TS_BIG_ENDIAN
 #if TS_PTR_SIZE == 32
-  SUBTREE_3_BYTES
-  SUBTREE_6_BITS
+  uint16_t parse_state;
+  uint8_t symbol;
+  SUBTREE_BITS
   bool unused : 1;
   bool is_inline : 1;
-  SUBTREE_4_BYTES
+  SUBTREE_SIZE
 #else
-  SUBTREE_4_BYTES
-  SUBTREE_3_BYTES
-  SUBTREE_6_BITS
+  SUBTREE_SIZE
+  uint16_t parse_state;
+  uint8_t symbol;
+  SUBTREE_BITS
   bool unused : 1;
   bool is_inline : 1;
 #endif
 #else
   bool is_inline : 1;
-  SUBTREE_6_BITS
-  SUBTREE_3_BYTES
-  SUBTREE_4_BYTES
+  SUBTREE_BITS
+  uint8_t symbol;
+  uint16_t parse_state;
+  SUBTREE_SIZE
 #endif
 
-#undef SUBTREE_3_BYTES
-#undef SUBTREE_6_BITS
-#undef SUBTREE_4_BYTES
+#undef SUBTREE_BITS
+#undef SUBTREE_SIZE
 } SubtreeInlineData;
 
 // A heap-allocated representation of a subtree.