Complete buffering and encoding code.

git-svn-id: http://svn.pyyaml.org/libyaml/trunk@182 18f92427-320e-0410-9341-c67f048884a3

3 files changed, 365 insertions, 137 deletions

diff --git a/include/yaml/yaml.h b/include/yaml/yaml.h
index 2f32e52..cfdeacf 100644
--- a/include/yaml/yaml.h
+++ b/include/yaml/yaml.h
@@ -300,6 +300,9 @@ typedef struct {
     /** The byte about which the problem occured. */
     size_t problem_offset;
 
+    /** The problematic value (@c -1 is none). */
+    int problem_value;
+
     /**
      * @}
      */
diff --git a/src/reader.c b/src/reader.c
index 5ee8c43..ac11323 100644
--- a/src/reader.c
+++ b/src/reader.c
@@ -12,11 +12,13 @@
  */
 
 int
-yaml_parser_set_reader_error(yaml_parser_t *parser, const char *problem)
+yaml_parser_set_reader_error(yaml_parser_t *parser, const char *problem,
+        size_t offset, int value)
 {
     parser->error = YAML_READER_ERROR;
     parser->problem = problem;
-    parser->problem_offset = parser->offset;
+    parser->problem_offset = offset;
+    parser->problem_value = value;
 
     return 0;
 }
@@ -32,9 +34,9 @@ yaml_parser_set_reader_error(yaml_parser_t *parser, const char *problem)
 int
 yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
 {
-    /* If the EOF flag is set, do nothing. */
+    /* If the EOF flag is set and the raw buffer is empty, do nothing. */
 
-    if (parser->eof)
+    if (parser->eof && !parser->raw_unread)
         return 1;
 
     /* Return if the buffer contains enough characters. */
@@ -71,18 +73,14 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
 
         if (!yaml_parser_update_raw_buffer(parser)) return 0;
 
-        /* If the raw buffer is empty, it is EOF. */
-
-        if (!parser->raw_unread) return 1;
-
         /* Decode the raw buffer. */
 
         while (parser->raw_unread)
         {
             unsigned int value, value2;
             int incomplete = 0;
-            unsigned char utf8_octet;
-            unsigned int utf8_length;
+            unsigned char octet;
+            unsigned int width;
             int k, low, high;
 
             /* Decode the next character. */
@@ -113,25 +111,26 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
 
                     /* Determine the length of the UTF-8 sequence. */
 
-                    utf8_octet = parser->raw_pointer[0];
-                    utf8_length = (
-                            (utf8_octet & 0x80) == 0x00 ? 1 :
-                            (utf8_octet & 0xE0) == 0xC0 ? 2 :
-                            (utf8_octet & 0xF0) == 0xE0 ? 3 :
-                            (utf8_octet & 0xF8) == 0xF0 ? 4 : 0);
+                    octet = parser->raw_pointer[0];
+                    width = (octet & 0x80) == 0x00 ? 1 :
+                            (octet & 0xE0) == 0xC0 ? 2 :
+                            (octet & 0xF0) == 0xE0 ? 3 :
+                            (octet & 0xF8) == 0xF0 ? 4 : 0;
 
                     /* Check if the leading octet is valid. */
 
-                    if (!utf8_length)
+                    if (!width)
                         return yaml_parser_set_reader_error(parser,
-                                "Invalid leading UTF-8 octet");
+                                "Invalid leading UTF-8 octet",
+                                parser->offset, octet);
 
                     /* Check if the raw buffer contains an incomplete character. */
 
-                    if (utf8_length > parser->raw_unread) {
+                    if (width > parser->raw_unread) {
                         if (parser->eof) {
                             return yaml_parser_set_reader_error(parser,
-                                    "Incomplete UTF-8 octet sequence");
+                                    "Incomplete UTF-8 octet sequence",
+                                    parser->offset, -1);
                         }
                         incomplete = 1;
                         break;
@@ -139,47 +138,45 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
 
                     /* Decode the leading octet. */
 
-                    value = (
-                        (utf8_octet & 0x80) == 0x00 ? utf8_octet & 0x7F :
-                        (utf8_octet & 0xE0) == 0xC0 ? utf8_octet & 0x1F :
-                        (utf8_octet & 0xF0) == 0xE0 ? utf8_octet & 0x0F :
-                        (utf8_octet & 0xF8) == 0xF0 ? utf8_octet & 0x07 : 0);
+                    value = (octet & 0x80) == 0x00 ? octet & 0x7F :
+                            (octet & 0xE0) == 0xC0 ? octet & 0x1F :
+                            (octet & 0xF0) == 0xE0 ? octet & 0x0F :
+                            (octet & 0xF8) == 0xF0 ? octet & 0x07 : 0;
 
                     /* Check and decode the trailing octets. */
 
-                    for (k = 1; k < utf8_length; k ++)
+                    for (k = 1; k < width; k ++)
                     {
-                        utf8_octet = parser->raw_pointer[k];
+                        octet = parser->raw_pointer[k];
 
                         /* Check if the octet is valid. */
 
-                        if ((utf8_octet & 0xC0) != 0x80)
+                        if ((octet & 0xC0) != 0x80)
                             return yaml_parser_set_reader_error(parser,
-                                    "Invalid trailing UTF-8 octet");
+                                    "Invalid trailing UTF-8 octet",
+                                    parser->offset+k, octet);
 
                         /* Decode the octet. */
 
-                        value = (value << 6) + (utf8_octet & 0x3F);
+                        value = (value << 6) + (octet & 0x3F);
                     }
 
                     /* Check the length of the sequence against the value. */
 
-                    if (!((utf8_length == 1) ||
-                            (utf8_length == 2 && value >= 0x80) ||
-                            (utf8_length == 3 && value >= 0x800) ||
-                            (utf8_length == 4 && value >= 0x10000)))
+                    if (!((width == 1) ||
+                            (width == 2 && value >= 0x80) ||
+                            (width == 3 && value >= 0x800) ||
+                            (width == 4 && value >= 0x10000)))
                         return yaml_parser_set_reader_error(parser,
-                                "Invalid length of a UTF-8 sequence");
+                                "Invalid length of a UTF-8 sequence",
+                                parser->offset, -1);
 
                     /* Check the range of the value. */
 
                     if ((value >= 0xD800 && value <= 0xDFFF) || value > 0x10FFFF)
                         return yaml_parser_set_reader_error(parser,
-                                "Invalid Unicode character");
-
-                    parser->raw_pointer += utf8_length;
-                    parser->raw_unread -= utf8_length;
-                    parser->offset += utf8_length;
+                                "Invalid Unicode character",
+                                parser->offset, value);
 
                     break;
                 
@@ -220,7 +217,8 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
                     if (parser->raw_unread < 2) {
                         if (parser->eof) {
                             return yaml_parser_set_reader_error(parser,
-                                    "Incomplete UTF-16 character");
+                                    "Incomplete UTF-16 character",
+                                    parser->offset, -1);
                         }
                         incomplete = 1;
                         break;
@@ -235,18 +233,22 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
 
                     if ((value & 0xFC00) == 0xDC00)
                         return yaml_parser_set_reader_error(parser,
-                                "Unexpected low surrogate area");
+                                "Unexpected low surrogate area",
+                                parser->offset, value);
 
                     /* Check for a high surrogate area. */
 
                     if ((value & 0xFC00) == 0xD800) {
 
+                        width = 4;
+
                         /* Check for incomplete surrogate pair. */
 
                         if (parser->raw_unread < 4) {
                             if (parser->eof) {
                                 return yaml_parser_set_reader_error(parser,
-                                        "Incomplete UTF-16 surrogate pair");
+                                        "Incomplete UTF-16 surrogate pair",
+                                        parser->offset, -1);
                             }
                             incomplete = 1;
                             break;
@@ -261,21 +263,16 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
 
                         if ((value2 & 0xFC00) != 0xDC00)
                             return yaml_parser_set_reader_error(parser,
-                                    "Expected low surrogate area");
+                                    "Expected low surrogate area",
+                                    parser->offset+2, value2);
 
                         /* Generate the value of the surrogate pair. */
 
                         value = 0x10000 + ((value & 0x3FF) << 10) + (value2 & 0x3FF);
-
-                        parser->raw_pointer += 4;
-                        parser->raw_unread -= 4;
-                        parser->offset += 4;
                     }
 
                     else {
-                        parser->raw_pointer += 2;
-                        parser->raw_unread -= 2;
-                        parser->offset += 4;
+                        width = 2;
                     }
 
                     break;
@@ -298,7 +295,14 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
                         || (value >= 0xE000 && value <= 0xFFFD)
                         || (value >= 0x10000 && value <= 0x10FFFF)))
                 return yaml_parser_set_reader_error(parser,
-                        "Control characters are not allowed");
+                        "Control characters are not allowed",
+                        parser->offset, value);
+
+            /* Move the raw pointers. */
+
+            parser->raw_pointer += width;
+            parser->raw_unread -= width;
+            parser->offset += width;
 
             /* Finally put the character into the buffer. */
 
@@ -309,22 +313,33 @@ yaml_parser_update_buffer(yaml_parser_t *parser, size_t length)
             /* 0000 0080-0000 07FF -> 110xxxxx 10xxxxxx */
             else if (value <= 0x7FF) {
                 *(parser->buffer_end++) = 0xC0 + (value >> 6);
-                *(parser->buffer_end++) = 0x80 + value & 0x3F;
+                *(parser->buffer_end++) = 0x80 + (value & 0x3F);
             }
             /* 0000 0800-0000 FFFF -> 1110xxxx 10xxxxxx 10xxxxxx */
             else if (value <= 0xFFFF) {
                 *(parser->buffer_end++) = 0xE0 + (value >> 12);
-                *(parser->buffer_end++) = 0x80 + (value >> 6) & 0x3F;
-                *(parser->buffer_end++) = 0x80 + value & 0x3F;
+                *(parser->buffer_end++) = 0x80 + ((value >> 6) & 0x3F);
+                *(parser->buffer_end++) = 0x80 + (value & 0x3F);
             }
             /* 0001 0000-0010 FFFF -> 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx */
             else {
                 *(parser->buffer_end++) = 0xF0 + (value >> 18);
-                *(parser->buffer_end++) = 0x80 + (value >> 12) & 0x3F;
-                *(parser->buffer_end++) = 0x80 + (value >> 6) & 0x3F;
-                *(parser->buffer_end++) = 0x80 + value & 0x3F;
+                *(parser->buffer_end++) = 0x80 + ((value >> 12) & 0x3F);
+                *(parser->buffer_end++) = 0x80 + ((value >> 6) & 0x3F);
+                *(parser->buffer_end++) = 0x80 + (value & 0x3F);
             }
+
+            parser->unread ++;
+        }
+
+        /* On EOF, put NUL into the buffer and return. */
+
+        if (parser->eof) {
+            *(parser->buffer_end++) = '\0';
+            parser->unread ++;
+            return 1;
         }
+
     }
 
     return 1;
@@ -357,18 +372,21 @@ yaml_parser_determine_encoding(yaml_parser_t *parser)
         parser->encoding = YAML_UTF16LE_ENCODING;
         parser->raw_pointer += 2;
         parser->raw_unread -= 2;
+        parser->offset += 2;
     }
     else if (parser->raw_unread >= 2
             && !memcmp(parser->raw_pointer, BOM_UTF16BE, 2)) {
         parser->encoding = YAML_UTF16BE_ENCODING;
         parser->raw_pointer += 2;
         parser->raw_unread -= 2;
+        parser->offset += 2;
     }
     else if (parser->raw_unread >= 3
             && !memcmp(parser->raw_pointer, BOM_UTF8, 3)) {
         parser->encoding = YAML_UTF8_ENCODING;
         parser->raw_pointer += 3;
         parser->raw_unread -= 3;
+        parser->offset += 3;
     }
     else {
         parser->encoding = YAML_UTF8_ENCODING;
@@ -407,7 +425,8 @@ yaml_parser_update_raw_buffer(yaml_parser_t *parser)
                 parser->raw_buffer + parser->raw_unread,
                 YAML_RAW_BUFFER_SIZE - parser->raw_unread,
                 &size_read)) {
-        return yaml_parser_set_reader_error(parser, "Input error");
+        return yaml_parser_set_reader_error(parser, "Input error",
+                parser->offset, -1);
     }
     parser->raw_unread += size_read;
     if (!size_read) {
diff --git a/tests/test-reader.c b/tests/test-reader.c
index b1a7426..53e4e7a 100644
--- a/tests/test-reader.c
+++ b/tests/test-reader.c
@@ -16,84 +16,96 @@ typedef struct {
 } test_case;
 
 test_case utf8_sequences[] = {
-        /* {"title", "test 1|test 2|...|test N!", (0 or 1)}, */
-
-        {"a simple test", "'test' is '\xd0\xbf\xd1\x80\xd0\xbe\xd0\xb2\xd0\xb5\xd1\x80\xd0\xba\xd0\xb0' in Russian!", 1},
-        {"an empty line", "!", 1},
-
-        {"u-0 is a control character", "\x00!", 0},
-        {"u-80 is a control character", "\xc2\x80!", 0},
-        {"u-800 is valid", "\xe0\xa0\x80!", 1},
-        {"u-10000 is valid", "\xf0\x90\x80\x80!", 1},
-        {"5 bytes sequences are not allowed", "\xf8\x88\x80\x80\x80!", 0},
-        {"6 bytes sequences are not allowed", "\xfc\x84\x80\x80\x80\x80!", 0},
-
-        {"u-7f is a control character", "\x7f!", 0},
-        {"u-7FF is valid", "\xdf\xbf!", 1},
-        {"u-FFFF is a control character", "\xef\xbf\xbf!", 0},
-        {"u-1FFFFF is too large", "\xf7\xbf\xbf\xbf!", 0},
-        {"u-3FFFFFF is 5 bytes", "\xfb\xbf\xbf\xbf\xbf!", 0},
-        {"u-7FFFFFFF is 6 bytes", "\xfd\xbf\xbf\xbf\xbf\xbf!", 0},
-
-        {"u-D7FF", "\xed\x9f\xbf!", 1},
-        {"u-E000", "\xee\x80\x80!", 1},
-        {"u-FFFD", "\xef\xbf\xbd!", 1},
-        {"u-10FFFF", "\xf4\x8f\xbf\xbf!", 1},
-        {"u-110000", "\xf4\x90\x80\x80!", 0},
-
-        {"first continuation byte", "\x80!", 0},
-        {"last continuation byte", "\xbf!", 0},
-
-        {"2 continuation bytes", "\x80\xbf!", 0},
-        {"3 continuation bytes", "\x80\xbf\x80!", 0},
-        {"4 continuation bytes", "\x80\xbf\x80\xbf!", 0},
-        {"5 continuation bytes", "\x80\xbf\x80\xbf\x80!", 0},
-        {"6 continuation bytes", "\x80\xbf\x80\xbf\x80\xbf!", 0},
-        {"7 continuation bytes", "\x80\xbf\x80\xbf\x80\xbf\x80!", 0},
-
-        {"sequence of all 64 possible continuation bytes",
-         "\x80|\x81|\x82|\x83|\x84|\x85|\x86|\x87|\x88|\x89|\x8a|\x8b|\x8c|\x8d|\x8e|\x8f|"
-         "\x90|\x91|\x92|\x93|\x94|\x95|\x96|\x97|\x98|\x99|\x9a|\x9b|\x9c|\x9d|\x9e|\x9f|"
-         "\xa0|\xa1|\xa2|\xa3|\xa4|\xa5|\xa6|\xa7|\xa8|\xa9|\xaa|\xab|\xac|\xad|\xae|\xaf|"
-         "\xb0|\xb1|\xb2|\xb3|\xb4|\xb5|\xb6|\xb7|\xb8|\xb9|\xba|\xbb|\xbc|\xbd|\xbe|\xbf!", 0},
-        {"32 first bytes of 2-byte sequences {0xc0-0xdf}",
-         "\xc0 |\xc1 |\xc2 |\xc3 |\xc4 |\xc5 |\xc6 |\xc7 |\xc8 |\xc9 |\xca |\xcb |\xcc |\xcd |\xce |\xcf |"
-         "\xd0 |\xd1 |\xd2 |\xd3 |\xd4 |\xd5 |\xd6 |\xd7 |\xd8 |\xd9 |\xda |\xdb |\xdc |\xdd |\xde |\xdf !", 0},
-        {"16 first bytes of 3-byte sequences {0xe0-0xef}",
-         "\xe0 |\xe1 |\xe2 |\xe3 |\xe4 |\xe5 |\xe6 |\xe7 |\xe8 |\xe9 |\xea |\xeb |\xec |\xed |\xee |\xef !", 0},
-        {"8 first bytes of 4-byte sequences {0xf0-0xf7}", "\xf0 |\xf1 |\xf2 |\xf3 |\xf4 |\xf5 |\xf6 |\xf7 !", 0},
-        {"4 first bytes of 5-byte sequences {0xf8-0xfb}", "\xf8 |\xf9 |\xfa |\xfb !", 0},
-        {"2 first bytes of 6-byte sequences {0xfc-0xfd}", "\xfc |\xfd !", 0},
-
-        {"sequences with last byte missing {u-0}",
-         "\xc0|\xe0\x80|\xf0\x80\x80|\xf8\x80\x80\x80|\xfc\x80\x80\x80\x80!", 0},
-        {"sequences with last byte missing {u-...FF}",
-         "\xdf|\xef\xbf|\xf7\xbf\xbf|\xfb\xbf\xbf\xbf|\xfd\xbf\xbf\xbf\xbf!", 0},
-
-        {"impossible bytes", "\xfe|\xff|\xfe\xfe\xff\xff!", 0},
-
-        {"overlong sequences {u-2f}",
-         "\xc0\xaf|\xe0\x80\xaf|\xf0\x80\x80\xaf|\xf8\x80\x80\x80\xaf|\xfc\x80\x80\x80\x80\xaf!", 0},
-
-        {"maximum overlong sequences",
-         "\xc1\xbf|\xe0\x9f\xbf|\xf0\x8f\xbf\xbf|\xf8\x87\xbf\xbf\xbf|\xfc\x83\xbf\xbf\xbf\xbf!", 0},
-
-        {"overlong representation of the NUL character",
-         "\xc0\x80|\xe0\x80\x80|\xf0\x80\x80\x80|\xf8\x80\x80\x80\x80|\xfc\x80\x80\x80\x80\x80!", 0},
-
-        {"single UTF-16 surrogates",
-         "\xed\xa0\x80|\xed\xad\xbf|\xed\xae\x80|\xed\xaf\xbf|\xed\xb0\x80|\xed\xbe\x80|\xed\xbf\xbf!", 0},
-
-        {"paired UTF-16 surrogates",
-         "\xed\xa0\x80\xed\xb0\x80|\xed\xa0\x80\xed\xbf\xbf|\xed\xad\xbf\xed\xb0\x80|"
-         "\xed\xad\xbf\xed\xbf\xbf|\xed\xae\x80\xed\xb0\x80|\xed\xae\x80\xed\xbf\xbf|"
-         "\xed\xaf\xbf\xed\xb0\x80|\xed\xaf\xbf\xed\xbf\xbf!", 0},
-
-        {"other illegal code positions", "\xef\xbf\xbe|\xef\xbf\xbf!", 0},
-
-        {NULL, NULL, 0}
+    /* {"title", "test 1|test 2|...|test N!", (0 or 1)}, */
+
+    {"a simple test", "'test' is '\xd0\xbf\xd1\x80\xd0\xbe\xd0\xb2\xd0\xb5\xd1\x80\xd0\xba\xd0\xb0' in Russian!", 1},
+    {"an empty line", "!", 1},
+
+    {"u-0 is a control character", "\x00!", 0},
+    {"u-80 is a control character", "\xc2\x80!", 0},
+    {"u-800 is valid", "\xe0\xa0\x80!", 1},
+    {"u-10000 is valid", "\xf0\x90\x80\x80!", 1},
+    {"5 bytes sequences are not allowed", "\xf8\x88\x80\x80\x80!", 0},
+    {"6 bytes sequences are not allowed", "\xfc\x84\x80\x80\x80\x80!", 0},
+
+    {"u-7f is a control character", "\x7f!", 0},
+    {"u-7FF is valid", "\xdf\xbf!", 1},
+    {"u-FFFF is a control character", "\xef\xbf\xbf!", 0},
+    {"u-1FFFFF is too large", "\xf7\xbf\xbf\xbf!", 0},
+    {"u-3FFFFFF is 5 bytes", "\xfb\xbf\xbf\xbf\xbf!", 0},
+    {"u-7FFFFFFF is 6 bytes", "\xfd\xbf\xbf\xbf\xbf\xbf!", 0},
+
+    {"u-D7FF", "\xed\x9f\xbf!", 1},
+    {"u-E000", "\xee\x80\x80!", 1},
+    {"u-FFFD", "\xef\xbf\xbd!", 1},
+    {"u-10FFFF", "\xf4\x8f\xbf\xbf!", 1},
+    {"u-110000", "\xf4\x90\x80\x80!", 0},
+
+    {"first continuation byte", "\x80!", 0},
+    {"last continuation byte", "\xbf!", 0},
+
+    {"2 continuation bytes", "\x80\xbf!", 0},
+    {"3 continuation bytes", "\x80\xbf\x80!", 0},
+    {"4 continuation bytes", "\x80\xbf\x80\xbf!", 0},
+    {"5 continuation bytes", "\x80\xbf\x80\xbf\x80!", 0},
+    {"6 continuation bytes", "\x80\xbf\x80\xbf\x80\xbf!", 0},
+    {"7 continuation bytes", "\x80\xbf\x80\xbf\x80\xbf\x80!", 0},
+
+    {"sequence of all 64 possible continuation bytes",
+     "\x80|\x81|\x82|\x83|\x84|\x85|\x86|\x87|\x88|\x89|\x8a|\x8b|\x8c|\x8d|\x8e|\x8f|"
+     "\x90|\x91|\x92|\x93|\x94|\x95|\x96|\x97|\x98|\x99|\x9a|\x9b|\x9c|\x9d|\x9e|\x9f|"
+     "\xa0|\xa1|\xa2|\xa3|\xa4|\xa5|\xa6|\xa7|\xa8|\xa9|\xaa|\xab|\xac|\xad|\xae|\xaf|"
+     "\xb0|\xb1|\xb2|\xb3|\xb4|\xb5|\xb6|\xb7|\xb8|\xb9|\xba|\xbb|\xbc|\xbd|\xbe|\xbf!", 0},
+    {"32 first bytes of 2-byte sequences {0xc0-0xdf}",
+     "\xc0 |\xc1 |\xc2 |\xc3 |\xc4 |\xc5 |\xc6 |\xc7 |\xc8 |\xc9 |\xca |\xcb |\xcc |\xcd |\xce |\xcf |"
+     "\xd0 |\xd1 |\xd2 |\xd3 |\xd4 |\xd5 |\xd6 |\xd7 |\xd8 |\xd9 |\xda |\xdb |\xdc |\xdd |\xde |\xdf !", 0},
+    {"16 first bytes of 3-byte sequences {0xe0-0xef}",
+     "\xe0 |\xe1 |\xe2 |\xe3 |\xe4 |\xe5 |\xe6 |\xe7 |\xe8 |\xe9 |\xea |\xeb |\xec |\xed |\xee |\xef !", 0},
+    {"8 first bytes of 4-byte sequences {0xf0-0xf7}", "\xf0 |\xf1 |\xf2 |\xf3 |\xf4 |\xf5 |\xf6 |\xf7 !", 0},
+    {"4 first bytes of 5-byte sequences {0xf8-0xfb}", "\xf8 |\xf9 |\xfa |\xfb !", 0},
+    {"2 first bytes of 6-byte sequences {0xfc-0xfd}", "\xfc |\xfd !", 0},
+
+    {"sequences with last byte missing {u-0}",
+     "\xc0|\xe0\x80|\xf0\x80\x80|\xf8\x80\x80\x80|\xfc\x80\x80\x80\x80!", 0},
+    {"sequences with last byte missing {u-...FF}",
+     "\xdf|\xef\xbf|\xf7\xbf\xbf|\xfb\xbf\xbf\xbf|\xfd\xbf\xbf\xbf\xbf!", 0},
+
+    {"impossible bytes", "\xfe|\xff|\xfe\xfe\xff\xff!", 0},
+
+    {"overlong sequences {u-2f}",
+     "\xc0\xaf|\xe0\x80\xaf|\xf0\x80\x80\xaf|\xf8\x80\x80\x80\xaf|\xfc\x80\x80\x80\x80\xaf!", 0},
+
+    {"maximum overlong sequences",
+     "\xc1\xbf|\xe0\x9f\xbf|\xf0\x8f\xbf\xbf|\xf8\x87\xbf\xbf\xbf|\xfc\x83\xbf\xbf\xbf\xbf!", 0},
+
+    {"overlong representation of the NUL character",
+     "\xc0\x80|\xe0\x80\x80|\xf0\x80\x80\x80|\xf8\x80\x80\x80\x80|\xfc\x80\x80\x80\x80\x80!", 0},
+
+    {"single UTF-16 surrogates",
+     "\xed\xa0\x80|\xed\xad\xbf|\xed\xae\x80|\xed\xaf\xbf|\xed\xb0\x80|\xed\xbe\x80|\xed\xbf\xbf!", 0},
+
+    {"paired UTF-16 surrogates",
+     "\xed\xa0\x80\xed\xb0\x80|\xed\xa0\x80\xed\xbf\xbf|\xed\xad\xbf\xed\xb0\x80|"
+     "\xed\xad\xbf\xed\xbf\xbf|\xed\xae\x80\xed\xb0\x80|\xed\xae\x80\xed\xbf\xbf|"
+     "\xed\xaf\xbf\xed\xb0\x80|\xed\xaf\xbf\xed\xbf\xbf!", 0},
+
+    {"other illegal code positions", "\xef\xbf\xbe|\xef\xbf\xbf!", 0},
+
+    {NULL, NULL, 0}
+};
+
+test_case boms[] = {
+    
+    /* {"title", "test!", lenth}, */
+    
+    {"no bom (utf-8)", "Hi is \xd0\x9f\xd1\x80\xd0\xb8\xd0\xb2\xd0\xb5\xd1\x82!", 13},
+    {"bom (utf-8)", "\xef\xbb\xbfHi is \xd0\x9f\xd1\x80\xd0\xb8\xd0\xb2\xd0\xb5\xd1\x82!", 13},
+    {"bom (utf-16-le)", "\xff\xfeH\x00i\x00 \x00i\x00s\x00 \x00\x1f\x04@\x04""8\x04""2\x04""5\x04""B\x04!", 13},
+    {"bom (utf-16-be)", "\xfe\xff\x00H\x00i\x00 \x00i\x00s\x00 \x04\x1f\x04@\x04""8\x04""2\x04""5\x04""B!", 13}
 };
 
+char *bom_original = "Hi is \xd0\x9f\xd1\x80\xd0\xb8\xd0\xb2\xd0\xb5\xd1\x82";
+
 int check_utf8_sequences(void)
 {
     yaml_parser_t *parser;
@@ -124,7 +136,14 @@ int check_utf8_sequences(void)
                 printf("(no error)\n");
             }
             else if (parser->error == YAML_READER_ERROR) {
-                printf("(reader error: %s at %d)\n", parser->problem, parser->problem_offset);
+                if (parser->problem_value != -1) {
+                    printf("(reader error: %s: #%X at %d)\n",
+                            parser->problem, parser->problem_value, parser->problem_offset);
+                }
+                else {
+                    printf("(reader error: %s at %d)\n",
+                            parser->problem, parser->problem_offset);
+                }
             }
             if (*end == '!') break;
             start = ++end;
@@ -136,9 +155,196 @@ int check_utf8_sequences(void)
     return failed;
 }
 
+int check_boms(void)
+{
+    yaml_parser_t *parser;
+    int failed = 0;
+    int k;
+    printf("checking boms...\n");
+    for (k = 0; boms[k].test; k++) {
+        char *title = boms[k].title;
+        int check = boms[k].result;
+        int result;
+        char *start = boms[k].test;
+        char *end = start;
+        while (*end != '!') end++;
+        printf("\t%s: ", title);
+        parser = yaml_parser_new();
+        assert(parser);
+        yaml_parser_set_input_string(parser, (unsigned char *)start, end-start);
+        result = yaml_parser_update_buffer(parser, end-start);
+        if (!result) {
+            printf("- (reader error: %s at %d)\n", parser->problem, parser->problem_offset);
+            failed++;
+        }
+        else {
+            if (parser->unread != check) {
+                printf("- (length=%d while expected length=%d)\n", parser->unread, check);
+                failed++;
+            }
+            else if (memcmp(parser->buffer, bom_original, check) != 0) {
+                printf("- (value '%s' does not equal to the original value '%s')\n", parser->buffer, bom_original);
+                failed++;
+            }
+            else {
+                printf("+\n");
+            }
+        }
+        yaml_parser_delete(parser);
+    }
+    printf("checking boms: %d fail(s)\n", failed);
+    return failed;
+}
+
+#define LONG    100000
+
+int check_long_utf8(void)
+{
+    yaml_parser_t *parser;
+    int k = 0;
+    int j;
+    int failed = 0;
+    unsigned char ch0, ch1;
+    unsigned char *buffer = malloc(3+LONG*2);
+    assert(buffer);
+    printf("checking a long utf8 sequence...\n");
+    buffer[k++] = '\xef';
+    buffer[k++] = '\xbb';
+    buffer[k++] = '\xbf';
+    for (j = 0; j < LONG; j ++) {
+        if (j % 2) {
+            buffer[k++] = '\xd0';
+            buffer[k++] = '\x90';
+        }
+        else {
+            buffer[k++] = '\xd0';
+            buffer[k++] = '\xaf';
+        }
+    }
+    parser = yaml_parser_new();
+    assert(parser);
+    yaml_parser_set_input_string(parser, buffer, 3+LONG*2);
+    for (k = 0; k < LONG; k++) {
+        if (!parser->unread) {
+            if (!yaml_parser_update_buffer(parser, 1)) {
+                printf("\treader error: %s at %d\n", parser->problem, parser->problem_offset);
+                failed = 1;
+                break;
+            }
+        }
+        if (!parser->unread) {
+            printf("\tnot enough characters at %d\n", k);
+            failed = 1;
+            break;
+        }
+        if (k % 2) {
+            ch0 = '\xd0';
+            ch1 = '\x90';
+        }
+        else {
+            ch0 = '\xd0';
+            ch1 = '\xaf';
+        }
+        if (parser->pointer[0] != ch0 || parser->pointer[1] != ch1) {
+            printf("\tincorrect UTF-8 sequence: %X %X instead of %X %X\n",
+                    (int)parser->pointer[0], (int)parser->pointer[1],
+                    (int)ch0, (int)ch1);
+            failed = 1;
+            break;
+        }
+        parser->pointer += 2;
+        parser->unread -= 1;
+    }
+    if (!failed) {
+        if (!yaml_parser_update_buffer(parser, 1)) {
+            printf("\treader error: %s at %d\n", parser->problem, parser->problem_offset);
+            failed = 1;
+        }
+        else if (parser->pointer[0] != '\0') {
+            printf("\texpected NUL, found %X (eof=%d, unread=%d)\n", (int)parser->pointer[0], parser->eof, parser->unread);
+            failed = 1;
+        }
+    }
+    yaml_parser_delete(parser);
+    free(buffer);
+    printf("checking a long utf8 sequence: %d fail(s)\n", failed);
+    return failed;
+}
+
+int check_long_utf16(void)
+{
+    yaml_parser_t *parser;
+    int k = 0;
+    int j;
+    int failed = 0;
+    unsigned char ch0, ch1;
+    unsigned char *buffer = malloc(2+LONG*2);
+    assert(buffer);
+    printf("checking a long utf16 sequence...\n");
+    buffer[k++] = '\xff';
+    buffer[k++] = '\xfe';
+    for (j = 0; j < LONG; j ++) {
+        if (j % 2) {
+            buffer[k++] = '\x10';
+            buffer[k++] = '\x04';
+        }
+        else {
+            buffer[k++] = '/';
+            buffer[k++] = '\x04';
+        }
+    }
+    parser = yaml_parser_new();
+    assert(parser);
+    yaml_parser_set_input_string(parser, buffer, 2+LONG*2);
+    for (k = 0; k < LONG; k++) {
+        if (!parser->unread) {
+            if (!yaml_parser_update_buffer(parser, 1)) {
+                printf("\treader error: %s at %d\n", parser->problem, parser->problem_offset);
+                failed = 1;
+                break;
+            }
+        }
+        if (!parser->unread) {
+            printf("\tnot enough characters at %d\n", k);
+            failed = 1;
+            break;
+        }
+        if (k % 2) {
+            ch0 = '\xd0';
+            ch1 = '\x90';
+        }
+        else {
+            ch0 = '\xd0';
+            ch1 = '\xaf';
+        }
+        if (parser->pointer[0] != ch0 || parser->pointer[1] != ch1) {
+            printf("\tincorrect UTF-8 sequence: %X %X instead of %X %X\n",
+                    (int)parser->pointer[0], (int)parser->pointer[1],
+                    (int)ch0, (int)ch1);
+            failed = 1;
+            break;
+        }
+        parser->pointer += 2;
+        parser->unread -= 1;
+    }
+    if (!failed) {
+        if (!yaml_parser_update_buffer(parser, 1)) {
+            printf("\treader error: %s at %d\n", parser->problem, parser->problem_offset);
+            failed = 1;
+        }
+        else if (parser->pointer[0] != '\0') {
+            printf("\texpected NUL, found %X (eof=%d, unread=%d)\n", (int)parser->pointer[0], parser->eof, parser->unread);
+            failed = 1;
+        }
+    }
+    yaml_parser_delete(parser);
+    free(buffer);
+    printf("checking a long utf16 sequence: %d fail(s)\n", failed);
+    return failed;
+}
 
 int
 main(void)
 {
-    return check_utf8_sequences();
+    return check_utf8_sequences() + check_boms() + check_long_utf8() + check_long_utf16();
 }