hpricot 0.8.2 → 0.8.3

data/ext/fast_xs/FastXsService.java

@@ -10,7 +10,6 @@ import org.jruby.RubyModule;
 import org.jruby.runtime.CallbackFactory;
 import org.jruby.runtime.builtin.IRubyObject;
 import org.jruby.runtime.load.BasicLibraryService;
-import org.jruby.util.collections.IntHashMap;
 
 public class FastXsService implements BasicLibraryService {
 
@@ -477,6 +476,100 @@ class Entities {
         int value(String name);
     }
 
+    // Very limited IntHashMap - it only supports get() and put()
+    private static class IntHashMap {
+        private transient Entry table[];
+
+        private transient int count;
+
+
+        private int threshold;
+ 
+        private final float loadFactor;
+ 
+        private static class Entry {
+            final int hash;
+            final int key;
+            Object value;
+            Entry next;
+ 
+            protected Entry(int hash, int key, Object value, Entry next) {
+                this.hash = hash;
+                this.key = key;
+                this.value = value;
+                this.next = next;
+            }
+        }
+ 
+        public IntHashMap() {
+            this.loadFactor = 0.75f;
+            table = new Entry[20];
+            threshold = (int) (20 * 0.75f);
+        }
+
+        public Object get(int key) {
+            Entry tab[] = table;
+            int hash = key;
+            int index = (hash & 0x7FFFFFFF) % tab.length;
+            for (Entry e = tab[index]; e != null; e = e.next) {
+                if (e.hash == hash) {
+                    return e.value;
+                }
+            }
+            return null;
+        }
+ 
+        protected void rehash() {
+            int oldCapacity = table.length;
+            Entry oldMap[] = table;
+ 
+            int newCapacity = oldCapacity * 2 + 1;
+            Entry newMap[] = new Entry[newCapacity];
+ 
+            threshold = (int) (newCapacity * loadFactor);
+            table = newMap;
+ 
+            for (int i = oldCapacity; i-- > 0;) {
+                for (Entry old = oldMap[i]; old != null;) {
+                    Entry e = old;
+                    old = old.next;
+ 
+                    int index = (e.hash & 0x7FFFFFFF) % newCapacity;
+                    e.next = newMap[index];
+                    newMap[index] = e;
+                }
+            }
+        }
+
+        public Object put(int key, Object value) {
+            // Makes sure the key is not already in the hashtable.
+            Entry tab[] = table;
+            int hash = key;
+            int index = (hash & 0x7FFFFFFF) % tab.length;
+            for (Entry e = tab[index]; e != null; e = e.next) {
+                if (e.hash == hash) {
+                    Object old = e.value;
+                    e.value = value;
+                    return old;
+                }
+            }
+ 
+            if (count >= threshold) {
+                // Rehash the table if the threshold is exceeded
+                rehash();
+ 
+                tab = table;
+                index = (hash & 0x7FFFFFFF) % tab.length;
+            }
+ 
+            // Creates the new entry.
+            Entry e = new Entry(hash, key, value, tab[index]);
+            tab[index] = e;
+            count++;
+            return null;
+        }
+    }
+
     static class PrimitiveEntityMap implements EntityMap {
         private Map mapNameToValue = new HashMap();
 

data/ext/fast_xs/fast_xs.c

@@ -1,8 +1,12 @@
-#define VERSION	"0.1"
-
 #include <ruby.h>
 #include <assert.h>
-/* #include <stdio.h> */
+
+#ifdef HAVE_RUBY_ENCODING_H
+#include <ruby/encoding.h>
+# define ASSOCIATE_INDEX(s,enc)  rb_enc_associate_index((s), rb_enc_to_index(enc))
+#else
+# define ASSOCIATE_INDEX(s,enc)
+#endif
 
 #ifndef RARRAY_LEN
 #define RARRAY_LEN(arr)  RARRAY(arr)->len
@@ -72,11 +76,6 @@ static const int cp_1252[] = {
 		n = cp_1252[n - 128]; \
 	} while(0)
 
-#define return_const_len(x) do { \
-	memcpy(buf, x, sizeof(x) - 1); \
-	return (sizeof(x) - 1); \
-} while (0)
-
 static inline size_t bytes_for(int n)
 {
 	if (n < 1000)
@@ -91,18 +90,24 @@ static inline size_t bytes_for(int n)
 	return sizeof("&#9999999;") - 1;
 }
 
-static long escape(char *buf, int n)
+static size_t escape(char *buf, int n)
 {
+
+#define return_const_len(x) do { \
+	memcpy(buf, x, sizeof(x) - 1); \
+	return (sizeof(x) - 1); \
+} while (0)
+
 	/* handle ASCII first */
 	if (likely(n < 128)) {
-		if (likely(n >= 0x20 || n == 0x9 || n == 0xA || n == 0xD)) {
-			if (unlikely(n == 34))
+		if (likely(n >= 0x20 || n == '\t' || n == '\n' || n == '\r')) {
+			if (unlikely(n == '"'))
 				return_const_len("&quot;");
-			if (unlikely(n == 38))
+			if (unlikely(n == '&'))
 				return_const_len("&amp;");
-			if (unlikely(n == 60))
+			if (unlikely(n == '<'))
 				return_const_len("&lt;");
-			if (unlikely(n == 62))
+			if (unlikely(n == '>'))
 				return_const_len("&gt;");
 			buf[0] = (char)n;
 			return 1;
@@ -112,16 +117,18 @@ static long escape(char *buf, int n)
 		return 1;
 	}
 
+#undef return_const_len
+
 	CP_1252_ESCAPE(n);
 
 	if (VALID_VALUE(n)) {
 		/* return snprintf(buf, sizeof("&#1114111;"), "&#%i;", n); */
-		RUBY_EXTERN const char ruby_digitmap[];
-		int rv = 3; /* &#; */
+		static const char digitmap[] = "0123456789";
+		size_t rv = sizeof("&#;") - 1;
 		buf += bytes_for(n);
 		*--buf = ';';
 		do {
-			*--buf = ruby_digitmap[(int)(n % 10)];
+			*--buf = digitmap[(int)(n % 10)];
 			++rv;
 		} while (n /= 10);
 		*--buf = '#';
@@ -132,27 +139,6 @@ static long escape(char *buf, int n)
 	return 1;
 }
 
-#undef return_const_len
-
-static long escaped_len(int n)
-{
-	if (likely(n < 128)) {
-		if (unlikely(n == 34))
-			return (sizeof("&quot;") - 1);
-		if (unlikely(n == 38))
-			return (sizeof("&amp;") - 1);
-		if (unlikely(n == 60 || n == 62))
-			return (sizeof("&gt;") - 1);
-		return 1;
-	}
-
-	CP_1252_ESCAPE(n);
-
-	if (VALID_VALUE(n))
-		return bytes_for(n);
-	return 1;
-}
-
 static VALUE unpack_utf8(VALUE self)
 {
 	return rb_funcall(self, unpack_id, 1, U_fmt);
@@ -163,28 +149,49 @@ static VALUE unpack_uchar(VALUE self)
 	return rb_funcall(self, unpack_id, 1, C_fmt);
 }
 
-VALUE fast_xs(VALUE self)
+/*
+ * escapes strings for XML
+ * The double-quote (") character is translated to "&quot;"
+ */
+static VALUE fast_xs(VALUE self)
 {
 	long i;
-	struct RArray *array;
-	char *s, *c;
-	long s_len = 0;
+	VALUE array;
+	char *c;
+	size_t s_len;
 	VALUE *tmp;
+	VALUE rv;
+
+	array = rb_rescue(unpack_utf8, self, unpack_uchar, self);
+
+	for (tmp = RARRAY_PTR(array), s_len = i = RARRAY_LEN(array);
+	     --i >= 0;
+	     tmp++) {
+		int n = NUM2INT(*tmp);
+		if (likely(n < 128)) {
+			if (unlikely(n == '"'))
+				s_len += (sizeof("&quot;") - 2);
+			if (unlikely(n == '&'))
+				s_len += (sizeof("&amp;") - 2);
+			if (unlikely(n == '>' || n == '<'))
+				s_len += (sizeof("&gt;") - 2);
+			continue;
+		}
 
-	array = RARRAY(rb_rescue(unpack_utf8, self, unpack_uchar, self));
+		CP_1252_ESCAPE(n);
 
-	tmp = RARRAY_PTR(array);
-	for (i = RARRAY_LEN(array); --i >= 0; tmp++)
-		s_len += escaped_len(NUM2INT(*tmp));
+		if (VALID_VALUE(n))
+			s_len += bytes_for(n) - 1;
+	}
 
-	c = s = alloca(s_len + 1);
+	rv = rb_str_new(NULL, s_len);
+        ASSOCIATE_INDEX(rv, rb_default_external_encoding());
+	c = RSTRING_PTR(rv);
 
-	tmp = RARRAY_PTR(array);
-	for (i = RARRAY_LEN(array); --i >= 0; tmp++)
+	for (tmp = RARRAY_PTR(array), i = RARRAY_LEN(array); --i >= 0; tmp++)
 		c += escape(c, NUM2INT(*tmp));
 
-	*c = '\0';
-	return rb_str_new(s, s_len);
+	return rv;
 }
 
 void Init_fast_xs(void)
@@ -193,7 +200,9 @@ void Init_fast_xs(void)
 
 	unpack_id = rb_intern("unpack");
 	U_fmt = rb_str_new("U*", 2);
+        ASSOCIATE_INDEX(U_fmt, rb_ascii8bit_encoding());
 	C_fmt = rb_str_new("C*", 2);
+        ASSOCIATE_INDEX(C_fmt, rb_ascii8bit_encoding());
 	rb_global_variable(&U_fmt);
 	rb_global_variable(&C_fmt);
 

data/ext/hpricot_scan/HpricotCss.java

@@ -1,3 +1,4 @@
+
 // line 1 "hpricot_css.java.rl"
 import java.io.IOException;
 
@@ -81,7 +82,7 @@ public class HpricotCss {
     }
     
 
-// line 85 "HpricotCss.java"
+// line 86 "HpricotCss.java"
 private static byte[] init__hpricot_css_actions_0()
 {
 	return new byte [] {
@@ -279,7 +280,7 @@ private static short[] init__hpricot_css_index_offsets_0()
 private static final short _hpricot_css_index_offsets[] = init__hpricot_css_index_offsets_0();
 
 
-private static byte[] init__hpricot_css_trans_targs_wi_0()
+private static byte[] init__hpricot_css_trans_targs_0()
 {
 	return new byte [] {
 	    1,   89,    1,   87,   90,    3,   90,    4,   90,   90,   90,    5,
@@ -353,14 +354,22 @@ private static byte[] init__hpricot_css_trans_targs_wi_0()
 	   93,  124,    9,   93,   93,   93,   10,   11,   12,   87,   93,    8,
 	   93,    9,   93,   93,   93,   10,   11,   12,   87,   93,    8,   93,
 	  126,    9,   93,   93,   93,   10,   11,   12,   87,   93,    8,   93,
-	    9,   93,   93,   93,   10,   11,   12,   87,    0
+	    9,   93,   93,   93,   10,   11,   12,   87,   87,   87,   87,   87,
+	   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,
+	   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,
+	   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,
+	   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,
+	   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,
+	   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,
+	   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,   87,
+	   87,   87,    0
 	};
 }
 
-private static final byte _hpricot_css_trans_targs_wi[] = init__hpricot_css_trans_targs_wi_0();
+private static final byte _hpricot_css_trans_targs[] = init__hpricot_css_trans_targs_0();
 
 
-private static byte[] init__hpricot_css_trans_actions_wi_0()
+private static byte[] init__hpricot_css_trans_actions_0()
 {
 	return new byte [] {
 	    0,    0,    0,   33,   99,    1,   99,    1,   99,   99,   99,    1,
@@ -434,11 +443,19 @@ private static byte[] init__hpricot_css_trans_actions_wi_0()
 	   79,   79,    0,   79,   79,   79,    0,    0,    0,   58,   79,    0,
 	   79,    0,   79,   79,   79,    0,    0,    0,   58,   79,    0,   79,
 	   79,    0,   79,   79,   79,    0,    0,    0,   58,   79,    0,   79,
-	    0,   79,   79,   79,    0,    0,    0,   58,    0
+	    0,   79,   79,   79,    0,    0,    0,   58,   33,   35,   35,   35,
+	   35,   35,   35,   35,   35,   35,   35,   35,   35,   35,   35,   35,
+	   31,   35,   35,   35,   35,   35,   35,   35,   35,   35,   35,   35,
+	   35,   35,   35,   35,   35,   35,   35,   35,   35,   35,   29,   29,
+	   27,   27,   27,   27,   27,   27,   27,   27,   25,   25,   25,   23,
+	   21,   52,   19,   58,   58,   55,   67,   67,   64,   67,   67,   67,
+	   67,   67,   67,   64,   67,   67,   67,   67,   67,   61,   67,   67,
+	   67,   67,   67,   67,   67,   67,   67,   17,   17,   58,   58,   58,
+	   58,   58,    0
 	};
 }
 
-private static final byte _hpricot_css_trans_actions_wi[] = init__hpricot_css_trans_actions_wi_0();
+private static final byte _hpricot_css_trans_actions[] = init__hpricot_css_trans_actions_0();
 
 
 private static byte[] init__hpricot_css_to_state_actions_0()
@@ -484,17 +501,17 @@ private static final byte _hpricot_css_from_state_actions[] = init__hpricot_css_
 private static short[] init__hpricot_css_eof_trans_0()
 {
 	return new short [] {
-	    0,    1,    0,   11,   11,   11,   11,   11,   11,   11,   11,   11,
-	   11,    0,   11,   11,   11,   11,   11,    0,   39,   11,   11,   11,
-	   11,   11,   11,   11,   11,   11,   11,   11,   11,   11,   11,   11,
-	   11,   11,   11,   11,   11,   11,   63,   63,   67,   67,   67,   67,
-	   67,   67,   67,   67,    0,    0,    0,    0,    0,    0,    0,  105,
-	  105,    0,  105,    0,    0,    0,    0,    0,    0,    0,    0,    0,
+	    0,  861,    0,  898,  898,  898,  898,  898,  898,  898,  898,  898,
+	  898,    0,  898,  898,  898,  898,  898,    0,  877,  898,  898,  898,
+	  898,  898,  898,  898,  898,  898,  898,  898,  898,  898,  898,  898,
+	  898,  898,  898,  898,  898,  898,  900,  900,  908,  908,  908,  908,
+	  908,  908,  908,  908,    0,    0,    0,    0,    0,    0,    0,  911,
+	  911,    0,  911,    0,    0,    0,    0,    0,    0,    0,    0,    0,
 	    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,
-	    0,    0,    0,    0,  150,  151,  152,  156,  157,  157,  163,  167,
-	  167,  174,  167,  167,  167,  167,  167,  167,  174,  167,  167,  167,
-	  167,  167,  193,  167,  167,  167,  167,  167,  167,  167,  167,  167,
-	  201,  201,  157,  157,  157,  157,  157
+	    0,    0,    0,    0,  912,  913,  914,  915,  950,  950,  918,  943,
+	  943,  928,  943,  943,  943,  943,  943,  943,  928,  943,  943,  943,
+	  943,  943,  934,  943,  943,  943,  943,  943,  943,  943,  943,  943,
+	  945,  945,  950,  950,  950,  950,  950
 	};
 }
 
@@ -506,21 +523,23 @@ static final int hpricot_css_error = 0;
 
 static final int hpricot_css_en_main = 87;
 
+
 // line 147 "hpricot_css.java.rl"
 
 
     public IRubyObject scan() {
 
-// line 515 "HpricotCss.java"
+// line 533 "HpricotCss.java"
 	{
 	cs = hpricot_css_start;
 	ts = -1;
 	te = -1;
 	act = 0;
 	}
+
 // line 151 "hpricot_css.java.rl"
 
-// line 524 "HpricotCss.java"
+// line 543 "HpricotCss.java"
 	{
 	int _klen;
 	int _trans = 0;
@@ -546,10 +565,10 @@ case 1:
 	while ( _nacts-- > 0 ) {
 		switch ( _hpricot_css_actions[_acts++] ) {
 	case 6:
-// line 1 "hpricot_css.java.rl"
+// line 1 "NONE"
 	{ts = p;}
 	break;
-// line 553 "HpricotCss.java"
+// line 572 "HpricotCss.java"
 		}
 	}
 
@@ -603,10 +622,10 @@ case 1:
 	} while (false);
 
 case 3:
-	cs = _hpricot_css_trans_targs_wi[_trans];
+	cs = _hpricot_css_trans_targs[_trans];
 
-	if ( _hpricot_css_trans_actions_wi[_trans] != 0 ) {
-		_acts = _hpricot_css_trans_actions_wi[_trans];
+	if ( _hpricot_css_trans_actions[_trans] != 0 ) {
+		_acts = _hpricot_css_trans_actions[_trans];
 		_nacts = (int) _hpricot_css_actions[_acts++];
 		while ( _nacts-- > 0 )
 	{
@@ -641,7 +660,7 @@ case 3:
   }
 	break;
 	case 7:
-// line 1 "hpricot_css.java.rl"
+// line 1 "NONE"
 	{te = p+1;}
 	break;
 	case 8:
@@ -753,7 +772,7 @@ case 3:
 	{{p = ((te))-1;}}
 	break;
 	case 35:
-// line 1 "hpricot_css.java.rl"
+// line 1 "NONE"
 	{	switch( act ) {
 	case 0:
 	{{cs = 0; _goto_targ = 2; if (true) continue _goto;}}
@@ -776,11 +795,10 @@ case 3:
 	case 9:
 	{{p = ((te))-1;} FILTER("PSUEDO"); }
 	break;
-	default: break;
 	}
 	}
 	break;
-// line 784 "HpricotCss.java"
+// line 802 "HpricotCss.java"
 			}
 		}
 	}
@@ -791,14 +809,14 @@ case 2:
 	while ( _nacts-- > 0 ) {
 		switch ( _hpricot_css_actions[_acts++] ) {
 	case 4:
-// line 1 "hpricot_css.java.rl"
+// line 1 "NONE"
 	{ts = -1;}
 	break;
 	case 5:
-// line 1 "hpricot_css.java.rl"
+// line 1 "NONE"
 	{act = 0;}
 	break;
-// line 802 "HpricotCss.java"
+// line 820 "HpricotCss.java"
 		}
 	}
 
@@ -824,6 +842,7 @@ case 5:
 	}
 	break; }
 	}
+
 // line 152 "hpricot_css.java.rl"
 
         return focus;