Fix arena flags

src/arena.h

@@ -5,12 +5,13 @@
 typedef U32 ArenaFlags;
 typedef struct ArenaOptions ArenaOptions;
 enum {
-    ArenaFlag_NoDecommit, // arena_ temp_end/reset/reset_to doesn't decommit memory
+    ArenaFlag_NoDecommit = 1, // operations which shrinks the arena does not decommit memory.
+                              // (arena_trim() can still decommit when called explicitly)
 };
 struct ArenaOptions {
     U32 size;
     U32 commit_size;
-    ArenaFlags flags; // No flags defined yet
+    ArenaFlags flags;
 };
 
 typedef struct MemoryBlock MemoryBlock;

src/string.c

@@ -93,36 +93,50 @@ S64 str_index(Str haystack, Str needle)
     return -1;
 }
 
+
+
+static INLINE S16 char_hex_to_binary(U8 c)
+{
+    if (c >='0' && c <='9')
+        return c - '0';
+    else if (c >= 'A' && c <= 'F')
+        return c - 'A' + 10;
+    else if (c >= 'a' && c <= 'f')
+        return c - 'a' + 10;
+
+    return -1;
+}
+
 static const char hex_chars[] = "0123456789ABCDEF";
 
-Str str_hex_encode(Str source, Str destination)
+Str str_hex_encode(Str binary_data, Str hex_result)
 {
-    Assert(destination.len  >= source.len * 2);
+    Assert(hex_result.len  >= binary_data.len * 2);
-    U32 hex_len = source.len * 2;
+    U32 hex_len = binary_data.len * 2;
-    for (U32 i = 0, t = 0; i < source.len; ++i,t+=2) {
+    for (U32 i = 0, t = 0; i < binary_data.len; ++i,t+=2) {
         U8 high_nibble;
         U8 low_nibble;
 
-        high_nibble = (source.str[i] & 0xF0) >> 4 ;
+        high_nibble = (binary_data.str[i] & 0xF0) >> 4 ;
-        low_nibble =  source.str[i] & 0x0F ;
+        low_nibble =  binary_data.str[i] & 0x0F ;
 
-        destination.str[t]   = hex_chars[high_nibble];
+        hex_result.str[t]   = hex_chars[high_nibble];
-        destination.str[t+1] = hex_chars[low_nibble];
+        hex_result.str[t+1] = hex_chars[low_nibble];
     }
 
-    destination.len = hex_len;
+    hex_result.len = hex_len;
-    return destination;
+    return hex_result;
 }
 
-Str str_hex_decode(Str source, Str destination)
+Str str_hex_decode(Str hex_str, Str result)
 {
-    Assert(destination.len >= source.len/2);
+    Assert(result.len >= hex_str.len/2);
     U32 dest_pos = 0;
     U32 source_pos = 0;
 
-    while (source_pos + 1 < source.len) {
+    while (source_pos + 1 < hex_str.len) {
 
-        if (char_is_ascii_whitespace(source.str[source_pos])) {
+        if (char_is_ascii_whitespace(hex_str.str[source_pos])) {
             source_pos++;
             continue;
         }
@@ -130,20 +144,20 @@ Str str_hex_decode(Str source, Str destination)
         S16 high_nibble;
         S16 low_nibble;
 
-        high_nibble = char_hex_to_binary(source.str[source_pos]);
+        high_nibble = char_hex_to_binary(hex_str.str[source_pos]);
-        low_nibble = char_hex_to_binary(source.str[source_pos + 1]);
+        low_nibble = char_hex_to_binary(hex_str.str[source_pos + 1]);
         if (high_nibble == -1 || low_nibble == -1) {
-            destination.len = 0;
+            result.len = 0;
-            return destination;
+            return result;
         }
 
-        destination.str[dest_pos] = (high_nibble << 4 ) | low_nibble;
+        result.str[dest_pos] = (high_nibble << 4 ) | low_nibble;
         source_pos += 2;
         dest_pos++;
     }
 
-    destination.len = dest_pos;
+    result.len = dest_pos;
-    return destination;
+    return result;
 }
 // split @str on any byte in @separators, place split strings in @vec
 void str_split(Str str, Str separators, StrVec *vec)
@@ -210,6 +224,105 @@ Str str_join(Allocator *allocator, StrVec strings, Str separator)
 }
 
 
+
+/*
+ * Each entry in utf8_lead[] is an 8-bit descriptor:
+ *
+ *     7 6 5 4 3 2 1 0
+ *    +---------+-----+
+ *    |  range  | len |
+ *    +---------+-----+
+ *
+ * len:
+ *     Number of bytes in the UTF-8 sequence.
+ *
+ *     0 = byte cannot start a valid UTF-8 sequence
+ *     1 = ASCII
+ *     2 = two-byte UTF-8 sequence
+ *     3 = three-byte UTF-8 sequence
+ *     4 = four-byte UTF-8 sequence
+ *
+ * range:
+ *     Selects the legal range for byte 1, the first continuation byte.
+ *
+ * This range is enough to encode all of UTF-8's special boundary rules:
+ *
+ *     E0 A0..BF    prevents overlong 3-byte encodings
+ *     ED 80..9F    excludes UTF-16 surrogate code points
+ *     F0 90..BF    prevents overlong 4-byte encodings
+ *     F4 80..8F    limits Unicode to U+10FFFF
+ *
+ * All continuation bytes after byte 1 simply use the normal 80..BF range.
+ */
+
+#define UTF8_LENGTH_BITS 3u
+#define UTF8_LENGTH_MASK ((1u << UTF8_LENGTH_BITS) - 1u)
+
+#define UTF8_DESC(length, range) \
+    ((U8)((length) | ((U8)(range) << UTF8_LENGTH_BITS)))
+
+#define UTF8_DESC_LENGTH(desc) \
+    ((U8)((desc) & UTF8_LENGTH_MASK))
+
+#define UTF8_DESC_RANGE(desc) \
+    ((U8)((desc) >> UTF8_LENGTH_BITS))
+
+
+/*
+ * Classes of allowed values for the first continuation byte.
+ */
+enum Utf8RangeClass {
+    /*
+     * No continuation byte exists.
+     *
+     * Used only by ASCII descriptors.
+     */
+    UTF8_RANGE_NONE = 0,
+
+    /*
+     * Normal continuation-byte range.
+     */
+    UTF8_RANGE_80_BF,
+
+    /*
+     * E0 xx xx:
+     *
+     * E0 80..9F would encode values below U+0800 and therefore be
+     * non-canonical overlong UTF-8.
+     */
+    UTF8_RANGE_A0_BF,
+
+    /*
+     * ED xx xx:
+     *
+     * ED A0..BF would encode U+D800..U+DFFF, which are UTF-16 surrogate
+     * code points and are not Unicode scalar values.
+     */
+    UTF8_RANGE_80_9F,
+
+    /*
+     * F0 xx xx xx:
+     *
+     * F0 80..8F would encode values below U+10000 and therefore be
+     * non-canonical overlong UTF-8.
+     */
+    UTF8_RANGE_90_BF,
+
+    /*
+     * F4 xx xx xx:
+     *
+     * F4 90..BF would encode values above Unicode's maximum scalar value,
+     * U+10FFFF.
+     */
+    UTF8_RANGE_80_8F,
+};
+
+
+typedef struct {
+    U8 lo;
+    U8 hi;
+} Utf8Range;
+
 /*
  * Allowed range of the first continuation byte for each range class.
  *
@@ -225,6 +338,36 @@ static const Utf8Range utf8_ranges[] = {
 };
 
 
+
+/*
+ * Each entry in utf8_lead[] is an 8-bit descriptor:
+ *
+ *     7 6 5 4 3 2 1 0
+ *    +---------+-----+
+ *    |  range  | len |
+ *    +---------+-----+
+ *
+ * len:
+ *     Number of bytes in the UTF-8 sequence.
+ *
+ *     0 = byte cannot start a valid UTF-8 sequence
+ *     1 = ASCII
+ *     2 = two-byte UTF-8 sequence
+ *     3 = three-byte UTF-8 sequence
+ *     4 = four-byte UTF-8 sequence
+ *
+ * range:
+ *     Selects the legal range for byte 1, the first continuation byte.
+ *
+ * This range is enough to encode all of UTF-8's special boundary rules:
+ *
+ *     E0 A0..BF    prevents overlong 3-byte encodings
+ *     ED 80..9F    excludes UTF-16 surrogate code points
+ *     F0 90..BF    prevents overlong 4-byte encodings
+ *     F4 80..8F    limits Unicode to U+10FFFF
+ *
+ * All continuation bytes after byte 1 simply use the normal 80..BF range.
+ */
 /*
  * Classification of every possible first byte.
  *

src/string.h

@@ -9,6 +9,7 @@ struct Str {
 
 Str str_empty(void);
 Str str_slice(Str str, U32 start, U32 len);
+// returns -1 if not found, or index start of first needle
 S64 str_index(Str haystack, Str needle);
 
 Str str_concat(Allocator *alloc, Str first, Str second);
@@ -65,6 +66,17 @@ static INLINE S64 str_find_byte(Str str, U8 byte)
     return -1;
 }
 
+static INLINE Str str_replace_byte(Str str, U8 orig_byte, U8 new_byte)
+{
+    for (U32 i = 0; i < str.len; i++) {
+        if (str.str[i] == orig_byte) {
+            str.str[i] = new_byte;
+        }
+    }
+
+    return str;
+}
+
 static INLINE bool str_starts_with(Str str, Str prefix)
 {
     if (prefix.len > str.len) {
@@ -91,28 +103,28 @@ static INLINE Str str_skip(Str str, U32 len)
     return new_str;
 }
 
-// Destination must be at least source.len * 2. Returns destination.
+// Destination must be at least source.len * 2. Returns hex_result.
-Str str_hex_encode(Str source, Str destination);
+Str str_hex_encode(Str binary_data, Str hex_result);
-// Destination must be at least source.len / 2. Returns destination.
+// Destination must be at least source.len / 2. Returns result.
-Str str_hex_decode(Str source, Str destination);
+Str str_hex_decode(Str hex_str, Str result);
 
-static INLINE Str str_hex_encode_alloc(Allocator *allocator, Str str)
+static INLINE Str str_hex_encode_alloc(Allocator *allocator, Str binary_data)
 {
-    Assert(str.len <= U32_Max / 2);
+    Assert(binary_data.len <= U32_Max / 2);
-    U8 hex_len = str.len * 2;
+    U8 hex_len = binary_data.len * 2;
     U8 *result = allocator_push_array(allocator, U8, hex_len);
     Str destination = str_create(result, hex_len);
 
-    return str_hex_encode(str, destination);
+    return str_hex_encode(binary_data, destination);
 }
 
-static INLINE Str str_hex_decode_alloc(Allocator *allocator, Str str)
+static INLINE Str str_hex_decode_alloc(Allocator *allocator, Str hex_str)
 {
-    U8 binary_len = str.len / 2;
+    U8 binary_len = hex_str.len / 2;
     U8 *result = allocator_push_array(allocator, U8, binary_len);
     Str destination = str_create(result, binary_len);
 
-    return str_hex_decode(str, destination);
+    return str_hex_decode(hex_str, destination);
 }
 // String Vector
 typedef struct StrVec StrVec;
@@ -147,17 +159,6 @@ static INLINE bool char_is_digit(U8 byte)
     return byte >= '0' && byte <= '9';
 }
 
-static INLINE S16 char_hex_to_binary(U8 c)
-{
-    if (c >='0' && c <='9')
-        return c - '0';
-    else if (c >= 'A' && c <= 'F')
-        return c - 'A' + 10;
-    else if (c >= 'a' && c <= 'f')
-        return c - 'a' + 10;
-
-    return -1;
-}
 
 // UTF-8/Unicode
 
@@ -197,111 +198,12 @@ typedef struct {
     /*
      * True only when rune was decoded from valid UTF-8.
      *
-     * This is required because U+FFFD itself may legitimately occur in
+     *  U+FFFD itself may legitimately occur in  valid
-     * valid input, so rune == RUNE_ERROR does not imply an error.
+     *  input, so rune == RUNE_ERROR does not imply an error.
      */
     bool valid;
 } RuneDecode;
 
-
-/*
- * Each entry in utf8_lead[] is an 8-bit descriptor:
- *
- *     7 6 5 4 3 2 1 0
- *    +---------+-----+
- *    |  range  | len |
- *    +---------+-----+
- *
- * len:
- *     Number of bytes in the UTF-8 sequence.
- *
- *     0 = byte cannot start a valid UTF-8 sequence
- *     1 = ASCII
- *     2 = two-byte UTF-8 sequence
- *     3 = three-byte UTF-8 sequence
- *     4 = four-byte UTF-8 sequence
- *
- * range:
- *     Selects the legal range for byte 1, the first continuation byte.
- *
- * This range is enough to encode all of UTF-8's special boundary rules:
- *
- *     E0 A0..BF    prevents overlong 3-byte encodings
- *     ED 80..9F    excludes UTF-16 surrogate code points
- *     F0 90..BF    prevents overlong 4-byte encodings
- *     F4 80..8F    limits Unicode to U+10FFFF
- *
- * All continuation bytes after byte 1 simply use the normal 80..BF range.
- */
-
-#define UTF8_LENGTH_BITS 3u
-#define UTF8_LENGTH_MASK ((1u << UTF8_LENGTH_BITS) - 1u)
-
-#define UTF8_DESC(length, range) \
-    ((U8)((length) | ((U8)(range) << UTF8_LENGTH_BITS)))
-
-#define UTF8_DESC_LENGTH(desc) \
-    ((U8)((desc) & UTF8_LENGTH_MASK))
-
-#define UTF8_DESC_RANGE(desc) \
-    ((U8)((desc) >> UTF8_LENGTH_BITS))
-
-
-/*
- * Classes of allowed values for the first continuation byte.
- */
-enum Utf8RangeClass {
-    /*
-     * No continuation byte exists.
-     *
-     * Used only by ASCII descriptors.
-     */
-    UTF8_RANGE_NONE = 0,
-
-    /*
-     * Normal continuation-byte range.
-     */
-    UTF8_RANGE_80_BF,
-
-    /*
-     * E0 xx xx:
-     *
-     * E0 80..9F would encode values below U+0800 and therefore be
-     * non-canonical overlong UTF-8.
-     */
-    UTF8_RANGE_A0_BF,
-
-    /*
-     * ED xx xx:
-     *
-     * ED A0..BF would encode U+D800..U+DFFF, which are UTF-16 surrogate
-     * code points and are not Unicode scalar values.
-     */
-    UTF8_RANGE_80_9F,
-
-    /*
-     * F0 xx xx xx:
-     *
-     * F0 80..8F would encode values below U+10000 and therefore be
-     * non-canonical overlong UTF-8.
-     */
-    UTF8_RANGE_90_BF,
-
-    /*
-     * F4 xx xx xx:
-     *
-     * F4 90..BF would encode values above Unicode's maximum scalar value,
-     * U+10FFFF.
-     */
-    UTF8_RANGE_80_8F,
-};
-
-
-typedef struct {
-    U8 lo;
-    U8 hi;
-} Utf8Range;
-
 // Helper:
 RuneDecode _str_utf8_decode_full(Str str, U32 offset);
 /*