modstruct: Basic implementation of native struct alignment and types.

pfalcon · pfalcon · commit 0c43cf9154e3 · 2014-04-11T03:59:16.000+03:00
diff --git a/py/binary.c b/py/binary.c
@@ -78,33 +78,62 @@ mp_obj_t mp_binary_get_val_array(char typecode, void *p, int index) {
 }
 
 #define is_signed(typecode) (typecode > 'Z')
-mp_obj_t mp_binary_get_val_unaligned(char typecode, byte **ptr) {
-    char type = '<';
+mp_obj_t mp_binary_get_val(char struct_type, char val_type, byte **ptr) {
     byte *p = *ptr;
-    uint size = 0, align = 0;
-    switch (type) {
+    uint size = 0;
+    switch (struct_type) {
         case '<': case '>':
-            switch (typecode) {
+            switch (val_type) {
                 case 'b': case 'B':
                     size = 1; break;
                 case 'h': case 'H':
                     size = 2; break;
                 case 'i': case 'I':
                     size = 4; break;
+                case 'l': case 'L':
+                    size = 4; break;
+            }
+            break;
+        case '@': {
+            // TODO:
+            // The simplest heuristic for alignment is to align by value
+            // size, but that doesn't work for "bigger than int" types,
+            // for example, long long may very well have long alignment
+            // So, we introduce separate alignment handling, but having
+            // formal support for that is different from actually supporting
+            // particular (or any) ABI.
+            uint align = 0;
+            switch (val_type) {
+                case 'b': case 'B':
+                    align = size = 1; break;
+                case 'h': case 'H':
+                    align = size = sizeof(short); break;
+                case 'i': case 'I':
+                    align = size = sizeof(int); break;
+                case 'l': case 'L':
+                    align = size = sizeof(long); break;
             }
+            // Make pointer aligned
+            p = (byte*)(((machine_uint_t)p + align - 1) & ~(align - 1));
+            #if MP_ENDIANNESS_LITTLE
+            struct_type = '<';
+            #else
+            struct_type = '>';
+            #endif
             break;
+        }
     }
 
     int delta;
-    if (type == '<') {
+    if (struct_type == '<') {
         delta = -1;
         p += size - 1;
     } else {
         delta = 1;
     }
 
     machine_int_t val = 0;
-    if (is_signed(typecode) && *p & 0x80) {
+    if (is_signed(val_type) && *p & 0x80) {
         val = -1;
     }
     for (uint i = 0; i < size; i++) {
@@ -113,8 +142,8 @@ mp_obj_t mp_binary_get_val_unaligned(char typecode, byte **ptr) {
         p += delta;
     }
 
-    *ptr += size + align;
-    if (is_signed(typecode)) {
+    *ptr += size;
+    if (is_signed(val_type)) {
         return mp_obj_new_int(val);
     } else {
         return mp_obj_new_int_from_uint(val);
diff --git a/py/binary.h b/py/binary.h
@@ -3,6 +3,6 @@
 #define BYTEARRAY_TYPECODE 0
 
 int mp_binary_get_size(char typecode);
-mp_obj_t mp_binary_get_val(char typecode, void *p, int index);
-mp_obj_t mp_binary_get_val_unaligned(char typecode, byte **ptr);
-void mp_binary_set_val(char typecode, void *p, int index, mp_obj_t val_in);
+mp_obj_t mp_binary_get_val_array(char typecode, void *p, int index);
+mp_obj_t mp_binary_get_val(char struct_type, char val_type, byte **ptr);
+void mp_binary_set_val_array(char typecode, void *p, int index, mp_obj_t val_in);
diff --git a/py/modstruct.c b/py/modstruct.c
@@ -37,7 +37,7 @@ STATIC uint calcsize_items(const char *fmt) {
 STATIC mp_obj_t struct_calcsize(mp_obj_t fmt_in) {
     const char *fmt = mp_obj_str_get_str(fmt_in);
     char fmt_type = get_fmt_type(&fmt);
-    assert(fmt_type == '<' || fmt_type == '>'); (void)fmt_type;
+    (void)fmt_type;
     machine_uint_t size;
     for (size = 0; *fmt; fmt++) {
         int sz = mp_binary_get_size(*fmt);
@@ -53,15 +53,14 @@ STATIC mp_obj_t struct_unpack(mp_obj_t fmt_in, mp_obj_t data_in) {
     // TODO: "The buffer must contain exactly the amount of data required by the format (len(bytes) must equal calcsize(fmt))."
     const char *fmt = mp_obj_str_get_str(fmt_in);
     char fmt_type = get_fmt_type(&fmt);
-    assert(fmt_type == '<' || fmt_type == '>'); (void)fmt_type;
     uint size = calcsize_items(fmt);
     mp_obj_tuple_t *res = mp_obj_new_tuple(size, NULL);
     buffer_info_t bufinfo;
     mp_get_buffer_raise(data_in, &bufinfo);
     byte *p = bufinfo.buf;
 
     for (uint i = 0; i < size; i++) {
-        mp_obj_t item = mp_binary_get_val_unaligned(*fmt++, &p);
+        mp_obj_t item = mp_binary_get_val(fmt_type, *fmt++, &p);
         res->items[i] = item;
     }
     return res;
diff --git a/tests/basics/struct1.py b/tests/basics/struct1.py
@@ -3,3 +3,6 @@
 print(struct.unpack("<bI", b"\x80\0\0\x01\0"))
 print(struct.calcsize(">bI"))
 print(struct.unpack(">bI", b"\x80\0\0\x01\0"))
+
+# 32-bit little-endian specific
+#print(struct.unpack("bI", b"\x80\xaa\x55\xaa\0\0\x01\0"))
