zscan 2.0.4 → 2.0.5

data/ext/pack/pack.c

@@ -2,13 +2,14 @@
 
   pack.c -
 
-  $Author$
+  $Author: naruse $
   created at: Thu Feb 10 15:17:05 JST 1994
 
   Copyright (C) 1993-2007 Yukihiro Matsumoto
 
 **********************************************************************/
 
+#include "ruby/encoding.h"
 #include "internal.h"
 #include <sys/types.h>
 #include <ctype.h>
@@ -126,6 +127,758 @@ str_associated(VALUE str)
     return rb_ivar_lookup(str, id_associated, Qfalse);
 }
 
+/*
+ *  call-seq:
+ *     arr.pack( aTemplateString ) -> aBinaryString
+ *     arr.pack( aTemplateString, buffer: aBufferString ) -> aBufferString
+ *
+ *  Packs the contents of <i>arr</i> into a binary sequence according to
+ *  the directives in <i>aTemplateString</i> (see the table below)
+ *  Directives ``A,'' ``a,'' and ``Z'' may be followed by a count,
+ *  which gives the width of the resulting field. The remaining
+ *  directives also may take a count, indicating the number of array
+ *  elements to convert. If the count is an asterisk
+ *  (``<code>*</code>''), all remaining array elements will be
+ *  converted. Any of the directives ``<code>sSiIlL</code>'' may be
+ *  followed by an underscore (``<code>_</code>'') or
+ *  exclamation mark (``<code>!</code>'') to use the underlying
+ *  platform's native size for the specified type; otherwise, they use a
+ *  platform-independent size. Spaces are ignored in the template
+ *  string. See also <code>String#unpack</code>.
+ *
+ *     a = [ "a", "b", "c" ]
+ *     n = [ 65, 66, 67 ]
+ *     a.pack("A3A3A3")   #=> "a  b  c  "
+ *     a.pack("a3a3a3")   #=> "a\000\000b\000\000c\000\000"
+ *     n.pack("ccc")      #=> "ABC"
+ *
+ *  If <i>aBufferString</i> is specified and its capacity is enough,
+ *  +pack+ uses it as the buffer and returns it.
+ *  When the offset is specified by the beginning of <i>aTemplateString</i>,
+ *  the result is filled after the offset.
+ *  If original contents of <i>aBufferString</i> exists and it's longer than
+ *  the offset, the rest of <i>offsetOfBuffer</i> are overwritten by the result.
+ *  If it's shorter, the gap is filled with ``<code>\0</code>''.
+ *
+ *  Note that ``buffer:'' option does not guarantee not to allocate memory
+ *  in +pack+.  If the capacity of <i>aBufferString</i> is not enough,
+ *  +pack+ allocates memory.
+ *
+ *  Directives for +pack+.
+ *
+ *   Integer       | Array   |
+ *   Directive     | Element | Meaning
+ *   ----------------------------------------------------------------------------
+ *   C             | Integer | 8-bit unsigned (unsigned char)
+ *   S             | Integer | 16-bit unsigned, native endian (uint16_t)
+ *   L             | Integer | 32-bit unsigned, native endian (uint32_t)
+ *   Q             | Integer | 64-bit unsigned, native endian (uint64_t)
+ *   J             | Integer | pointer width unsigned, native endian (uintptr_t)
+ *                 |         | (J is available since Ruby 2.3.)
+ *                 |         |
+ *   c             | Integer | 8-bit signed (signed char)
+ *   s             | Integer | 16-bit signed, native endian (int16_t)
+ *   l             | Integer | 32-bit signed, native endian (int32_t)
+ *   q             | Integer | 64-bit signed, native endian (int64_t)
+ *   j             | Integer | pointer width signed, native endian (intptr_t)
+ *                 |         | (j is available since Ruby 2.3.)
+ *                 |         |
+ *   S_ S!         | Integer | unsigned short, native endian
+ *   I I_ I!       | Integer | unsigned int, native endian
+ *   L_ L!         | Integer | unsigned long, native endian
+ *   Q_ Q!         | Integer | unsigned long long, native endian (ArgumentError
+ *                 |         | if the platform has no long long type.)
+ *                 |         | (Q_ and Q! is available since Ruby 2.1.)
+ *   J!            | Integer | uintptr_t, native endian (same with J)
+ *                 |         | (J! is available since Ruby 2.3.)
+ *                 |         |
+ *   s_ s!         | Integer | signed short, native endian
+ *   i i_ i!       | Integer | signed int, native endian
+ *   l_ l!         | Integer | signed long, native endian
+ *   q_ q!         | Integer | signed long long, native endian (ArgumentError
+ *                 |         | if the platform has no long long type.)
+ *                 |         | (q_ and q! is available since Ruby 2.1.)
+ *   j!            | Integer | intptr_t, native endian (same with j)
+ *                 |         | (j! is available since Ruby 2.3.)
+ *                 |         |
+ *   S> s> S!> s!> | Integer | same as the directives without ">" except
+ *   L> l> L!> l!> |         | big endian
+ *   I!> i!>       |         | (available since Ruby 1.9.3)
+ *   Q> q> Q!> q!> |         | "S>" is same as "n"
+ *   J> j> J!> j!> |         | "L>" is same as "N"
+ *                 |         |
+ *   S< s< S!< s!< | Integer | same as the directives without "<" except
+ *   L< l< L!< l!< |         | little endian
+ *   I!< i!<       |         | (available since Ruby 1.9.3)
+ *   Q< q< Q!< q!< |         | "S<" is same as "v"
+ *   J< j< J!< j!< |         | "L<" is same as "V"
+ *                 |         |
+ *   n             | Integer | 16-bit unsigned, network (big-endian) byte order
+ *   N             | Integer | 32-bit unsigned, network (big-endian) byte order
+ *   v             | Integer | 16-bit unsigned, VAX (little-endian) byte order
+ *   V             | Integer | 32-bit unsigned, VAX (little-endian) byte order
+ *                 |         |
+ *   U             | Integer | UTF-8 character
+ *   w             | Integer | BER-compressed integer
+ *
+ *   Float        | Array   |
+ *   Directive    | Element | Meaning
+ *   ---------------------------------------------------------------------------
+ *   D d          | Float   | double-precision, native format
+ *   F f          | Float   | single-precision, native format
+ *   E            | Float   | double-precision, little-endian byte order
+ *   e            | Float   | single-precision, little-endian byte order
+ *   G            | Float   | double-precision, network (big-endian) byte order
+ *   g            | Float   | single-precision, network (big-endian) byte order
+ *
+ *   String       | Array   |
+ *   Directive    | Element | Meaning
+ *   ---------------------------------------------------------------------------
+ *   A            | String  | arbitrary binary string (space padded, count is width)
+ *   a            | String  | arbitrary binary string (null padded, count is width)
+ *   Z            | String  | same as ``a'', except that null is added with *
+ *   B            | String  | bit string (MSB first)
+ *   b            | String  | bit string (LSB first)
+ *   H            | String  | hex string (high nibble first)
+ *   h            | String  | hex string (low nibble first)
+ *   u            | String  | UU-encoded string
+ *   M            | String  | quoted printable, MIME encoding (see also RFC2045)
+ *                |         | (text mode but input must use LF and output LF)
+ *   m            | String  | base64 encoded string (see RFC 2045, count is width)
+ *                |         | (if count is 0, no line feed are added, see RFC 4648)
+ *   P            | String  | pointer to a structure (fixed-length string)
+ *   p            | String  | pointer to a null-terminated string
+ *
+ *   Misc.        | Array   |
+ *   Directive    | Element | Meaning
+ *   ---------------------------------------------------------------------------
+ *   @            | ---     | moves to absolute position
+ *   X            | ---     | back up a byte
+ *   x            | ---     | null byte
+ */
+
+static VALUE
+pack_pack(int argc, VALUE *argv, VALUE ary)
+{
+    static const char nul10[] = "\0\0\0\0\0\0\0\0\0\0";
+    static const char spc10[] = "          ";
+    const char *p, *pend;
+    VALUE fmt, opt = Qnil, res, from, associates = 0, buffer = 0;
+    char type;
+    long len, idx, plen;
+    const char *ptr;
+    int enc_info = 1;		/* 0 - BINARY, 1 - US-ASCII, 2 - UTF-8 */
+#ifdef NATINT_PACK
+    int natint;		/* native integer */
+#endif
+    int integer_size, bigendian_p;
+
+    rb_scan_args(argc, argv, "10:", &fmt, &opt);
+
+    StringValue(fmt);
+    p = RSTRING_PTR(fmt);
+    pend = p + RSTRING_LEN(fmt);
+    if (!NIL_P(opt)) {
+	static ID keyword_ids[1];
+	if (!keyword_ids[0])
+	    CONST_ID(keyword_ids[0], "buffer");
+
+	rb_get_kwargs(opt, keyword_ids, 0, 1, &buffer);
+
+	if (buffer != Qundef && !RB_TYPE_P(buffer, T_STRING))
+	    rb_raise(rb_eTypeError, "buffer must be String, not %s", rb_obj_classname(buffer));
+    }
+    if (buffer)
+	res = buffer;
+    else
+	res = rb_str_buf_new(0);
+
+    idx = 0;
+
+#define TOO_FEW (rb_raise(rb_eArgError, toofew), 0)
+#define MORE_ITEM (idx < RARRAY_LEN(ary))
+#define THISFROM (MORE_ITEM ? RARRAY_AREF(ary, idx) : TOO_FEW)
+#define NEXTFROM (MORE_ITEM ? RARRAY_AREF(ary, idx++) : TOO_FEW)
+
+    while (p < pend) {
+	int explicit_endian = 0;
+	if (RSTRING_PTR(fmt) + RSTRING_LEN(fmt) != pend) {
+	    rb_raise(rb_eRuntimeError, "format string modified");
+	}
+	type = *p++;		/* get data type */
+#ifdef NATINT_PACK
+	natint = 0;
+#endif
+
+	if (ISSPACE(type)) continue;
+	if (type == '#') {
+	    while ((p < pend) && (*p != '\n')) {
+		p++;
+	    }
+	    continue;
+	}
+
+	{
+          modifiers:
+	    switch (*p) {
+	      case '_':
+	      case '!':
+		if (strchr(natstr, type)) {
+#ifdef NATINT_PACK
+		    natint = 1;
+#endif
+		    p++;
+		}
+		else {
+		    rb_raise(rb_eArgError, "'%c' allowed only after types %s", *p, natstr);
+		}
+		goto modifiers;
+
+	      case '<':
+	      case '>':
+		if (!strchr(endstr, type)) {
+		    rb_raise(rb_eArgError, "'%c' allowed only after types %s", *p, endstr);
+		}
+		if (explicit_endian) {
+		    rb_raise(rb_eRangeError, "Can't use both '<' and '>'");
+		}
+		explicit_endian = *p++;
+		goto modifiers;
+	    }
+	}
+
+	if (*p == '*') {	/* set data length */
+	    len = strchr("@Xxu", type) ? 0
+                : strchr("PMm", type) ? 1
+                : RARRAY_LEN(ary) - idx;
+	    p++;
+	}
+	else if (ISDIGIT(*p)) {
+	    errno = 0;
+	    len = STRTOUL(p, (char**)&p, 10);
+	    if (errno) {
+		rb_raise(rb_eRangeError, "pack length too big");
+	    }
+	}
+	else {
+	    len = 1;
+	}
+
+	switch (type) {
+	  case 'U':
+	    /* if encoding is US-ASCII, upgrade to UTF-8 */
+	    if (enc_info == 1) enc_info = 2;
+	    break;
+	  case 'm': case 'M': case 'u':
+	    /* keep US-ASCII (do nothing) */
+	    break;
+	  default:
+	    /* fall back to BINARY */
+	    enc_info = 0;
+	    break;
+	}
+	switch (type) {
+	  case 'A': case 'a': case 'Z':
+	  case 'B': case 'b':
+	  case 'H': case 'h':
+	    from = NEXTFROM;
+	    if (NIL_P(from)) {
+		ptr = "";
+		plen = 0;
+	    }
+	    else {
+		StringValue(from);
+		ptr = RSTRING_PTR(from);
+		plen = RSTRING_LEN(from);
+		OBJ_INFECT(res, from);
+	    }
+
+	    if (p[-1] == '*')
+		len = plen;
+
+	    switch (type) {
+	      case 'a':		/* arbitrary binary string (null padded)  */
+	      case 'A':         /* arbitrary binary string (ASCII space padded) */
+	      case 'Z':         /* null terminated string  */
+		if (plen >= len) {
+		    rb_str_buf_cat(res, ptr, len);
+		    if (p[-1] == '*' && type == 'Z')
+			rb_str_buf_cat(res, nul10, 1);
+		}
+		else {
+		    rb_str_buf_cat(res, ptr, plen);
+		    len -= plen;
+		    while (len >= 10) {
+			rb_str_buf_cat(res, (type == 'A')?spc10:nul10, 10);
+			len -= 10;
+		    }
+		    rb_str_buf_cat(res, (type == 'A')?spc10:nul10, len);
+		}
+		break;
+
+#define castchar(from) (char)((from) & 0xff)
+
+	      case 'b':		/* bit string (ascending) */
+		{
+		    int byte = 0;
+		    long i, j = 0;
+
+		    if (len > plen) {
+			j = (len - plen + 1)/2;
+			len = plen;
+		    }
+		    for (i=0; i++ < len; ptr++) {
+			if (*ptr & 1)
+			    byte |= 128;
+			if (i & 7)
+			    byte >>= 1;
+			else {
+			    char c = castchar(byte);
+			    rb_str_buf_cat(res, &c, 1);
+			    byte = 0;
+			}
+		    }
+		    if (len & 7) {
+			char c;
+			byte >>= 7 - (len & 7);
+			c = castchar(byte);
+			rb_str_buf_cat(res, &c, 1);
+		    }
+		    len = j;
+		    goto grow;
+		}
+		break;
+
+	      case 'B':		/* bit string (descending) */
+		{
+		    int byte = 0;
+		    long i, j = 0;
+
+		    if (len > plen) {
+			j = (len - plen + 1)/2;
+			len = plen;
+		    }
+		    for (i=0; i++ < len; ptr++) {
+			byte |= *ptr & 1;
+			if (i & 7)
+			    byte <<= 1;
+			else {
+			    char c = castchar(byte);
+			    rb_str_buf_cat(res, &c, 1);
+			    byte = 0;
+			}
+		    }
+		    if (len & 7) {
+			char c;
+			byte <<= 7 - (len & 7);
+			c = castchar(byte);
+			rb_str_buf_cat(res, &c, 1);
+		    }
+		    len = j;
+		    goto grow;
+		}
+		break;
+
+	      case 'h':		/* hex string (low nibble first) */
+		{
+		    int byte = 0;
+		    long i, j = 0;
+
+		    if (len > plen) {
+			j = (len + 1) / 2 - (plen + 1) / 2;
+			len = plen;
+		    }
+		    for (i=0; i++ < len; ptr++) {
+			if (ISALPHA(*ptr))
+			    byte |= (((*ptr & 15) + 9) & 15) << 4;
+			else
+			    byte |= (*ptr & 15) << 4;
+			if (i & 1)
+			    byte >>= 4;
+			else {
+			    char c = castchar(byte);
+			    rb_str_buf_cat(res, &c, 1);
+			    byte = 0;
+			}
+		    }
+		    if (len & 1) {
+			char c = castchar(byte);
+			rb_str_buf_cat(res, &c, 1);
+		    }
+		    len = j;
+		    goto grow;
+		}
+		break;
+
+	      case 'H':		/* hex string (high nibble first) */
+		{
+		    int byte = 0;
+		    long i, j = 0;
+
+		    if (len > plen) {
+			j = (len + 1) / 2 - (plen + 1) / 2;
+			len = plen;
+		    }
+		    for (i=0; i++ < len; ptr++) {
+			if (ISALPHA(*ptr))
+			    byte |= ((*ptr & 15) + 9) & 15;
+			else
+			    byte |= *ptr & 15;
+			if (i & 1)
+			    byte <<= 4;
+			else {
+			    char c = castchar(byte);
+			    rb_str_buf_cat(res, &c, 1);
+			    byte = 0;
+			}
+		    }
+		    if (len & 1) {
+			char c = castchar(byte);
+			rb_str_buf_cat(res, &c, 1);
+		    }
+		    len = j;
+		    goto grow;
+		}
+		break;
+	    }
+	    break;
+
+	  case 'c':		/* signed char */
+	  case 'C':		/* unsigned char */
+            integer_size = 1;
+            bigendian_p = BIGENDIAN_P(); /* not effective */
+            goto pack_integer;
+
+	  case 's':		/* s for int16_t, s! for signed short */
+            integer_size = NATINT_LEN(short, 2);
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'S':		/* S for uint16_t, S! for unsigned short */
+            integer_size = NATINT_LEN(short, 2);
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'i':		/* i and i! for signed int */
+            integer_size = (int)sizeof(int);
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'I':		/* I and I! for unsigned int */
+            integer_size = (int)sizeof(int);
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'l':		/* l for int32_t, l! for signed long */
+            integer_size = NATINT_LEN(long, 4);
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'L':		/* L for uint32_t, L! for unsigned long */
+            integer_size = NATINT_LEN(long, 4);
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'q':		/* q for int64_t, q! for signed long long */
+	    integer_size = NATINT_LEN_Q;
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'Q':		/* Q for uint64_t, Q! for unsigned long long */
+	    integer_size = NATINT_LEN_Q;
+            bigendian_p = BIGENDIAN_P();
+            goto pack_integer;
+
+	  case 'j':		/* j for intptr_t */
+	    integer_size = sizeof(intptr_t);
+	    bigendian_p = BIGENDIAN_P();
+	    goto pack_integer;
+
+	  case 'J':		/* J for uintptr_t */
+	    integer_size = sizeof(uintptr_t);
+	    bigendian_p = BIGENDIAN_P();
+	    goto pack_integer;
+
+	  case 'n':		/* 16 bit (2 bytes) integer (network byte-order)  */
+            integer_size = 2;
+            bigendian_p = 1;
+            goto pack_integer;
+
+	  case 'N':		/* 32 bit (4 bytes) integer (network byte-order) */
+            integer_size = 4;
+            bigendian_p = 1;
+            goto pack_integer;
+
+	  case 'v':		/* 16 bit (2 bytes) integer (VAX byte-order) */
+            integer_size = 2;
+            bigendian_p = 0;
+            goto pack_integer;
+
+	  case 'V':		/* 32 bit (4 bytes) integer (VAX byte-order) */
+            integer_size = 4;
+            bigendian_p = 0;
+            goto pack_integer;
+
+          pack_integer:
+	    if (explicit_endian) {
+		bigendian_p = explicit_endian == '>';
+	    }
+            if (integer_size > MAX_INTEGER_PACK_SIZE)
+                rb_bug("unexpected intger size for pack: %d", integer_size);
+            while (len-- > 0) {
+                char intbuf[MAX_INTEGER_PACK_SIZE];
+
+                from = NEXTFROM;
+                rb_integer_pack(from, intbuf, integer_size, 1, 0,
+                    INTEGER_PACK_2COMP |
+                    (bigendian_p ? INTEGER_PACK_BIG_ENDIAN : INTEGER_PACK_LITTLE_ENDIAN));
+                rb_str_buf_cat(res, intbuf, integer_size);
+            }
+	    break;
+
+	  case 'f':		/* single precision float in native format */
+	  case 'F':		/* ditto */
+	    while (len-- > 0) {
+		float f;
+
+		from = NEXTFROM;
+		f = (float)RFLOAT_VALUE(rb_to_float(from));
+		rb_str_buf_cat(res, (char*)&f, sizeof(float));
+	    }
+	    break;
+
+	  case 'e':		/* single precision float in VAX byte-order */
+	    while (len-- > 0) {
+		FLOAT_CONVWITH(tmp);
+
+		from = NEXTFROM;
+		tmp.f = (float)RFLOAT_VALUE(rb_to_float(from));
+		HTOVF(tmp);
+		rb_str_buf_cat(res, tmp.buf, sizeof(float));
+	    }
+	    break;
+
+	  case 'E':		/* double precision float in VAX byte-order */
+	    while (len-- > 0) {
+		DOUBLE_CONVWITH(tmp);
+		from = NEXTFROM;
+		tmp.d = RFLOAT_VALUE(rb_to_float(from));
+		HTOVD(tmp);
+		rb_str_buf_cat(res, tmp.buf, sizeof(double));
+	    }
+	    break;
+
+	  case 'd':		/* double precision float in native format */
+	  case 'D':		/* ditto */
+	    while (len-- > 0) {
+		double d;
+
+		from = NEXTFROM;
+		d = RFLOAT_VALUE(rb_to_float(from));
+		rb_str_buf_cat(res, (char*)&d, sizeof(double));
+	    }
+	    break;
+
+	  case 'g':		/* single precision float in network byte-order */
+	    while (len-- > 0) {
+		FLOAT_CONVWITH(tmp);
+		from = NEXTFROM;
+		tmp.f = (float)RFLOAT_VALUE(rb_to_float(from));
+		HTONF(tmp);
+		rb_str_buf_cat(res, tmp.buf, sizeof(float));
+	    }
+	    break;
+
+	  case 'G':		/* double precision float in network byte-order */
+	    while (len-- > 0) {
+		DOUBLE_CONVWITH(tmp);
+
+		from = NEXTFROM;
+		tmp.d = RFLOAT_VALUE(rb_to_float(from));
+		HTOND(tmp);
+		rb_str_buf_cat(res, tmp.buf, sizeof(double));
+	    }
+	    break;
+
+	  case 'x':		/* null byte */
+	  grow:
+	    while (len >= 10) {
+		rb_str_buf_cat(res, nul10, 10);
+		len -= 10;
+	    }
+	    rb_str_buf_cat(res, nul10, len);
+	    break;
+
+	  case 'X':		/* back up byte */
+	  shrink:
+	    plen = RSTRING_LEN(res);
+	    if (plen < len)
+		rb_raise(rb_eArgError, "X outside of string");
+	    rb_str_set_len(res, plen - len);
+	    break;
+
+	  case '@':		/* null fill to absolute position */
+	    len -= RSTRING_LEN(res);
+	    if (len > 0) goto grow;
+	    len = -len;
+	    if (len > 0) goto shrink;
+	    break;
+
+	  case '%':
+	    rb_raise(rb_eArgError, "%% is not supported");
+	    break;
+
+	  case 'U':		/* Unicode character */
+	    while (len-- > 0) {
+		SIGNED_VALUE l;
+		char buf[8];
+		int le;
+
+		from = NEXTFROM;
+		from = rb_to_int(from);
+		l = NUM2LONG(from);
+		if (l < 0) {
+		    rb_raise(rb_eRangeError, "pack(U): value out of range");
+		}
+		le = rb_uv_to_utf8(buf, l);
+		rb_str_buf_cat(res, (char*)buf, le);
+	    }
+	    break;
+
+	  case 'u':		/* uuencoded string */
+	  case 'm':		/* base64 encoded string */
+	    from = NEXTFROM;
+	    StringValue(from);
+	    ptr = RSTRING_PTR(from);
+	    plen = RSTRING_LEN(from);
+
+	    if (len == 0 && type == 'm') {
+		encodes(res, ptr, plen, type, 0);
+		ptr += plen;
+		break;
+	    }
+	    if (len <= 2)
+		len = 45;
+	    else if (len > 63 && type == 'u')
+		len = 63;
+	    else
+		len = len / 3 * 3;
+	    while (plen > 0) {
+		long todo;
+
+		if (plen > len)
+		    todo = len;
+		else
+		    todo = plen;
+		encodes(res, ptr, todo, type, 1);
+		plen -= todo;
+		ptr += todo;
+	    }
+	    break;
+
+	  case 'M':		/* quoted-printable encoded string */
+	    from = rb_obj_as_string(NEXTFROM);
+	    if (len <= 1)
+		len = 72;
+	    qpencode(res, from, len);
+	    break;
+
+	  case 'P':		/* pointer to packed byte string */
+	    from = THISFROM;
+	    if (!NIL_P(from)) {
+		StringValue(from);
+		if (RSTRING_LEN(from) < len) {
+		    rb_raise(rb_eArgError, "too short buffer for P(%ld for %ld)",
+			     RSTRING_LEN(from), len);
+		}
+	    }
+	    len = 1;
+	    /* FALL THROUGH */
+	  case 'p':		/* pointer to string */
+	    while (len-- > 0) {
+		char *t;
+		from = NEXTFROM;
+		if (NIL_P(from)) {
+		    t = 0;
+		}
+		else {
+		    t = StringValuePtr(from);
+		    rb_obj_taint(from);
+		}
+		if (!associates) {
+		    associates = rb_ary_new();
+		}
+		rb_ary_push(associates, from);
+		rb_str_buf_cat(res, (char*)&t, sizeof(char*));
+	    }
+	    break;
+
+	  case 'w':		/* BER compressed integer  */
+	    while (len-- > 0) {
+		VALUE buf = rb_str_new(0, 0);
+                size_t numbytes;
+                int sign;
+                char *cp;
+
+		from = NEXTFROM;
+                from = rb_to_int(from);
+                numbytes = rb_absint_numwords(from, 7, NULL);
+                if (numbytes == 0)
+                    numbytes = 1;
+                buf = rb_str_new(NULL, numbytes);
+
+                sign = rb_integer_pack(from, RSTRING_PTR(buf), RSTRING_LEN(buf), 1, 1, INTEGER_PACK_BIG_ENDIAN);
+
+                if (sign < 0)
+                    rb_raise(rb_eArgError, "can't compress negative numbers");
+                if (sign == 2)
+                    rb_bug("buffer size problem?");
+
+                cp = RSTRING_PTR(buf);
+                while (1 < numbytes) {
+                  *cp |= 0x80;
+                  cp++;
+                  numbytes--;
+                }
+
+                rb_str_buf_cat(res, RSTRING_PTR(buf), RSTRING_LEN(buf));
+	    }
+	    break;
+
+	  default: {
+	    char unknown[5];
+	    if (ISPRINT(type)) {
+		unknown[0] = type;
+		unknown[1] = '\0';
+	    }
+	    else {
+		snprintf(unknown, sizeof(unknown), "\\x%.2x", type & 0xff);
+	    }
+	    rb_warning("unknown pack directive '%s' in '% "PRIsVALUE"'",
+		       unknown, fmt);
+	    break;
+	  }
+	}
+    }
+
+    if (associates) {
+	str_associate(res, associates);
+    }
+    OBJ_INFECT(res, fmt);
+    switch (enc_info) {
+      case 1:
+	ENCODING_CODERANGE_SET(res, rb_usascii_encindex(), ENC_CODERANGE_7BIT);
+	break;
+      case 2:
+	rb_enc_set_index(res, rb_utf8_encindex());
+	break;
+      default:
+	/* do nothing, keep ASCII-8BIT */
+	break;
+    }
+    return res;
+}
+
 static const char uu_table[] =
 "`!\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_";
 static const char b64_table[] =
@@ -260,10 +1013,11 @@ hex2num(char c)
 	rb_ary_store(ary, RARRAY_LEN(ary)+tmp_len-1, Qnil); \
 } while (0)
 
-/* Workaround for Oracle Solaris Studio 12.4 C compiler optimization bug
+/* Workaround for Oracle Developer Studio (Oracle Solaris Studio)
+ * 12.4/12.5/12.6 C compiler optimization bug
  * with "-xO4" optimization option.
  */
-#if defined(__SUNPRO_C) && __SUNPRO_C == 0x5130
+#if defined(__SUNPRO_C) && 0x5130 <= __SUNPRO_C && __SUNPRO_C <= 0x5150
 # define AVOID_CC_BUG volatile
 #else
 # define AVOID_CC_BUG
@@ -283,12 +1037,11 @@ infected_str_new(const char *ptr, long len, VALUE str)
 #define UNPACK_BLOCK 1
 #define UNPACK_1 2
 
-#define castchar(from) (char)((from) & 0xff)
-
-VALUE zscan_internal_unpack(VALUE str, VALUE fmt, long* parsed_len)
+static VALUE
+pack_unpack_internal(VALUE str, VALUE fmt, int mode)
 {
 #define hexdigits ruby_hexdigits
-    char *init_s, *s, *send;
+    char *s, *send;
     char *p, *pend;
     VALUE ary;
     char type;
@@ -299,20 +1052,22 @@ VALUE zscan_internal_unpack(VALUE str, VALUE fmt, long* parsed_len)
     int natint;			/* native integer */
 #endif
     int signed_p, integer_size, bigendian_p;
-    int mode = (rb_block_given_p() ? UNPACK_BLOCK : UNPACK_ARRAY);
 #define UNPACK_PUSH(item) do {\
 	VALUE item_val = (item);\
 	if ((mode) == UNPACK_BLOCK) {\
 	    rb_yield(item_val);\
 	}\
-	else {\
+	else if ((mode) == UNPACK_ARRAY) {\
 	    rb_ary_push(ary, item_val);\
 	}\
+	else /* if ((mode) == UNPACK_1) { */ {\
+	    return item_val; \
+	}\
     } while (0)
 
     StringValue(str);
     StringValue(fmt);
-    init_s = s = RSTRING_PTR(str);
+    s = RSTRING_PTR(str);
     send = s + RSTRING_LEN(str);
     p = RSTRING_PTR(fmt);
     pend = p + RSTRING_LEN(fmt);
@@ -374,7 +1129,7 @@ VALUE zscan_internal_unpack(VALUE str, VALUE fmt, long* parsed_len)
 	else if (ISDIGIT(*p)) {
 	    errno = 0;
 	    len = STRTOUL(p, (char**)&p, 10);
-	    if (errno) {
+	    if (len < 0 || errno) {
 		rb_raise(rb_eRangeError, "pack length too big");
 	    }
 	}
@@ -845,6 +1600,7 @@ VALUE zscan_internal_unpack(VALUE str, VALUE fmt, long* parsed_len)
 	    {
 		VALUE buf = infected_str_new(0, send - s, str);
 		char *ptr = RSTRING_PTR(buf), *ss = s;
+		int csum = 0;
 		int c1, c2;
 
 		while (s < send) {
@@ -856,18 +1612,19 @@ VALUE zscan_internal_unpack(VALUE str, VALUE fmt, long* parsed_len)
 			    if ((c1 = hex2num(*s)) == -1) break;
 			    if (++s == send) break;
 			    if ((c2 = hex2num(*s)) == -1) break;
-			    *ptr++ = castchar(c1 << 4 | c2);
+			    csum |= *ptr++ = castchar(c1 << 4 | c2);
 			}
 		    }
 		    else {
-			*ptr++ = *s;
+			csum |= *ptr++ = *s;
 		    }
 		    s++;
 		    ss = s;
 		}
 		rb_str_set_len(buf, ptr - RSTRING_PTR(buf));
 		rb_str_buf_cat(buf, ss, send-ss);
-		ENCODING_CODERANGE_SET(buf, rb_ascii8bit_encindex(), ENC_CODERANGE_VALID);
+		csum = ISASCII(csum) ? ENC_CODERANGE_7BIT : ENC_CODERANGE_VALID;
+		ENCODING_CODERANGE_SET(buf, rb_ascii8bit_encindex(), csum);
 		UNPACK_PUSH(buf);
 	    }
 	    break;
@@ -990,10 +1747,151 @@ VALUE zscan_internal_unpack(VALUE str, VALUE fmt, long* parsed_len)
 	}
     }
 
-    *parsed_len = s - init_s;
     return ary;
 }
 
+/*
+ *  call-seq:
+ *     str.unpack(format)    ->  anArray
+ *
+ *  Decodes <i>str</i> (which may contain binary data) according to the
+ *  format string, returning an array of each value extracted. The
+ *  format string consists of a sequence of single-character directives,
+ *  summarized in the table at the end of this entry.
+ *  Each directive may be followed
+ *  by a number, indicating the number of times to repeat with this
+ *  directive. An asterisk (``<code>*</code>'') will use up all
+ *  remaining elements. The directives <code>sSiIlL</code> may each be
+ *  followed by an underscore (``<code>_</code>'') or
+ *  exclamation mark (``<code>!</code>'') to use the underlying
+ *  platform's native size for the specified type; otherwise, it uses a
+ *  platform-independent consistent size. Spaces are ignored in the
+ *  format string. See also <code>String#unpack1</code>,  <code>Array#pack</code>.
+ *
+ *     "abc \0\0abc \0\0".unpack('A6Z6')   #=> ["abc", "abc "]
+ *     "abc \0\0".unpack('a3a3')           #=> ["abc", " \000\000"]
+ *     "abc \0abc \0".unpack('Z*Z*')       #=> ["abc ", "abc "]
+ *     "aa".unpack('b8B8')                 #=> ["10000110", "01100001"]
+ *     "aaa".unpack('h2H2c')               #=> ["16", "61", 97]
+ *     "\xfe\xff\xfe\xff".unpack('sS')     #=> [-2, 65534]
+ *     "now=20is".unpack('M*')             #=> ["now is"]
+ *     "whole".unpack('xax2aX2aX1aX2a')    #=> ["h", "e", "l", "l", "o"]
+ *
+ *  This table summarizes the various formats and the Ruby classes
+ *  returned by each.
+ *
+ *   Integer       |         |
+ *   Directive     | Returns | Meaning
+ *   ------------------------------------------------------------------
+ *   C             | Integer | 8-bit unsigned (unsigned char)
+ *   S             | Integer | 16-bit unsigned, native endian (uint16_t)
+ *   L             | Integer | 32-bit unsigned, native endian (uint32_t)
+ *   Q             | Integer | 64-bit unsigned, native endian (uint64_t)
+ *   J             | Integer | pointer width unsigned, native endian (uintptr_t)
+ *                 |         |
+ *   c             | Integer | 8-bit signed (signed char)
+ *   s             | Integer | 16-bit signed, native endian (int16_t)
+ *   l             | Integer | 32-bit signed, native endian (int32_t)
+ *   q             | Integer | 64-bit signed, native endian (int64_t)
+ *   j             | Integer | pointer width signed, native endian (intptr_t)
+ *                 |         |
+ *   S_ S!         | Integer | unsigned short, native endian
+ *   I I_ I!       | Integer | unsigned int, native endian
+ *   L_ L!         | Integer | unsigned long, native endian
+ *   Q_ Q!         | Integer | unsigned long long, native endian (ArgumentError
+ *                 |         | if the platform has no long long type.)
+ *   J!            | Integer | uintptr_t, native endian (same with J)
+ *                 |         |
+ *   s_ s!         | Integer | signed short, native endian
+ *   i i_ i!       | Integer | signed int, native endian
+ *   l_ l!         | Integer | signed long, native endian
+ *   q_ q!         | Integer | signed long long, native endian (ArgumentError
+ *                 |         | if the platform has no long long type.)
+ *   j!            | Integer | intptr_t, native endian (same with j)
+ *                 |         |
+ *   S> s> S!> s!> | Integer | same as the directives without ">" except
+ *   L> l> L!> l!> |         | big endian
+ *   I!> i!>       |         |
+ *   Q> q> Q!> q!> |         | "S>" is same as "n"
+ *   J> j> J!> j!> |         | "L>" is same as "N"
+ *                 |         |
+ *   S< s< S!< s!< | Integer | same as the directives without "<" except
+ *   L< l< L!< l!< |         | little endian
+ *   I!< i!<       |         |
+ *   Q< q< Q!< q!< |         | "S<" is same as "v"
+ *   J< j< J!< j!< |         | "L<" is same as "V"
+ *                 |         |
+ *   n             | Integer | 16-bit unsigned, network (big-endian) byte order
+ *   N             | Integer | 32-bit unsigned, network (big-endian) byte order
+ *   v             | Integer | 16-bit unsigned, VAX (little-endian) byte order
+ *   V             | Integer | 32-bit unsigned, VAX (little-endian) byte order
+ *                 |         |
+ *   U             | Integer | UTF-8 character
+ *   w             | Integer | BER-compressed integer (see Array.pack)
+ *
+ *   Float        |         |
+ *   Directive    | Returns | Meaning
+ *   -----------------------------------------------------------------
+ *   D d          | Float   | double-precision, native format
+ *   F f          | Float   | single-precision, native format
+ *   E            | Float   | double-precision, little-endian byte order
+ *   e            | Float   | single-precision, little-endian byte order
+ *   G            | Float   | double-precision, network (big-endian) byte order
+ *   g            | Float   | single-precision, network (big-endian) byte order
+ *
+ *   String       |         |
+ *   Directive    | Returns | Meaning
+ *   -----------------------------------------------------------------
+ *   A            | String  | arbitrary binary string (remove trailing nulls and ASCII spaces)
+ *   a            | String  | arbitrary binary string
+ *   Z            | String  | null-terminated string
+ *   B            | String  | bit string (MSB first)
+ *   b            | String  | bit string (LSB first)
+ *   H            | String  | hex string (high nibble first)
+ *   h            | String  | hex string (low nibble first)
+ *   u            | String  | UU-encoded string
+ *   M            | String  | quoted-printable, MIME encoding (see RFC2045)
+ *   m            | String  | base64 encoded string (RFC 2045) (default)
+ *                |         | base64 encoded string (RFC 4648) if followed by 0
+ *   P            | String  | pointer to a structure (fixed-length string)
+ *   p            | String  | pointer to a null-terminated string
+ *
+ *   Misc.        |         |
+ *   Directive    | Returns | Meaning
+ *   -----------------------------------------------------------------
+ *   @            | ---     | skip to the offset given by the length argument
+ *   X            | ---     | skip backward one byte
+ *   x            | ---     | skip forward one byte
+ *
+ *  HISTORY
+ *
+ *  * J, J! j, and j! are available since Ruby 2.3.
+ *  * Q_, Q!, q_, and q! are available since Ruby 2.1.
+ *  * I!<, i!<, I!>, and i!> are available since Ruby 1.9.3.
+ */
+
+static VALUE
+pack_unpack(VALUE str, VALUE fmt)
+{
+    int mode = rb_block_given_p() ? UNPACK_BLOCK : UNPACK_ARRAY;
+    return pack_unpack_internal(str, fmt, mode);
+}
+
+/*
+ *  call-seq:
+ *     str.unpack1(format)    ->  obj
+ *
+ *  Decodes <i>str</i> (which may contain binary data) according to the
+ *  format string, returning the first value extracted.
+ *  See also <code>String#unpack</code>, <code>Array#pack</code>.
+ */
+
+static VALUE
+pack_unpack1(VALUE str, VALUE fmt)
+{
+    return pack_unpack_internal(str, fmt, UNPACK_1);
+}
+
 int
 rb_uv_to_utf8(char buf[6], unsigned long uv)
 {
@@ -1100,3 +1998,13 @@ utf8_to_uv(const char *p, long *lenp)
     }
     return uv;
 }
+
+void
+Init_pack(void)
+{
+    rb_define_method(rb_cArray, "pack", pack_pack, -1);
+    rb_define_method(rb_cString, "unpack", pack_unpack, 1);
+    rb_define_method(rb_cString, "unpack1", pack_unpack1, 1);
+
+    id_associated = rb_make_internal_id();
+}