ox 2.2.0 → 2.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: a72bd41c796380d832d6fe08b7b7e77aeca0bbe4
-  data.tar.gz: 5aa7c7ef218a7a97cda25bd1586e133f168e91af
+  metadata.gz: bec620893f1af26eaf853fec54ac033da9516689
+  data.tar.gz: 26fd103eb9d31e42ce45bf5371b51740d91d7b21
 SHA512:
-  metadata.gz: 715a10def5ec9ab769de09a7ac0ed6e7cbd56b48cb75a8e5f347a3143cfb3128df8224bf2a7b3d7d598fd297ef294f3120a5de3ad153e4a175d970005bb5f9a6
-  data.tar.gz: d6edb43235036f091a8150b998a5ce4115c19317203cc853d60f3e34f77b3971896c4bd87fae16b74ce9e3ce6f950664eec7fb42232bd88c12329e93f887bbc0
+  metadata.gz: 64abf65cab8bd13621831043039a8e4e8ccf7e1bf839bf061a443fb5e9cd7e5d7d6a1334cc36ab5caa9bb71034c03ddcd0f40b0f1d9630920a73be8828230b7c
+  data.tar.gz: 78332e2f9ef6892242bb8ed4a84a4353a122e8980f660575c719520a83aca5be0d8c6f376b1dd9628b168c7ddc347403c59afd783a175d7fbed6f76be407a2b5

data/README.md

@@ -34,25 +34,19 @@ A fast XML parser and Object marshaller as a Ruby gem.
 
 ## Release Notes
 
-### Current Release 2.2.0
+### Current Release 2.2.1
 
- - Added the SAX convert_special option to the default options.
-
- - Added the SAX smart option to the default options.
-
- - Other SAX options are now taken from the defaults if not specified.
+ - Added support to handle script elements in html.
 
-### Release 2.1.8
+ - Added support for position from start for the sax parser.
 
- - Fixed a bug that caused all input to be read before parsing with the sax
-   parser and an IO.pipe.
+### Release 2.2.0
 
-### Release 2.1.7
+ - Added the SAX convert_special option to the default options.
 
- - Empty elements such as <foo></foo> are now called back with empty text.
+ - Added the SAX smart option to the default options.
 
- - Fixed GC problem that occurs with the new GC in Ruby 2.2 that garbage
-   collects Symbols.
+ - Other SAX options are now taken from the defaults if not specified.
 
 ## Description
 

data/ext/ox/ox.c

@@ -29,6 +29,7 @@ ID	ox_at_column_id;
 ID	ox_at_content_id;
 ID	ox_at_id;
 ID	ox_at_line_id;
+ID	ox_at_pos_id;
 ID	ox_at_value_id;
 ID	ox_attr_id;
 ID	ox_attr_value_id;
@@ -910,6 +911,7 @@ void Init_ox() {
     ox_at_content_id = rb_intern("@content");
     ox_at_id = rb_intern("at");
     ox_at_line_id = rb_intern("@line");
+    ox_at_pos_id = rb_intern("@pos");
     ox_at_value_id = rb_intern("@value");
     ox_attr_id = rb_intern("attr");
     ox_attr_value_id = rb_intern("attr_value");

data/ext/ox/ox.h

@@ -172,6 +172,7 @@ extern ID	ox_at_column_id;
 extern ID	ox_at_content_id;
 extern ID	ox_at_id;
 extern ID	ox_at_line_id;
+extern ID	ox_at_pos_id;
 extern ID	ox_at_value_id;
 extern ID	ox_attr_id;
 extern ID	ox_attr_value_id;

data/ext/ox/sax.c

@@ -49,11 +49,12 @@ static char		read_comment(SaxDrive dr);
 static char		read_element_start(SaxDrive dr);
 static char		read_element_end(SaxDrive dr);
 static char		read_text(SaxDrive dr);
+static char		read_jump(SaxDrive dr, const char *pat);
 static char		read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req);
 static char		read_name_token(SaxDrive dr);
 static char		read_quoted_value(SaxDrive dr);
 
-static void		end_element_cb(SaxDrive dr, VALUE name, int line, int col);
+static void		end_element_cb(SaxDrive dr, VALUE name, int pos, int line, int col);
 
 static void		hint_clear_empty(SaxDrive dr);
 static Nv		hint_try_close(SaxDrive dr, const char *name);
@@ -155,6 +156,7 @@ ox_sax_parse(VALUE handler, VALUE io, SaxOptions options) {
     printf("    has_start_element = %s\n", dr.has.start_element ? "true" : "false");
     printf("    has_end_element = %s\n", dr.has.end_element ? "true" : "false");
     printf("    has_error = %s\n", dr.has.error ? "true" : "false");
+    printf("    has_pos = %s\n", dr.has.pos ? "true" : "false");
     printf("    has_line = %s\n", dr.has.line ? "true" : "false");
     printf("    has_column = %s\n", dr.has.column ? "true" : "false");
 #endif
@@ -217,13 +219,19 @@ ox_sax_drive_cleanup(SaxDrive dr) {
 }
 
 static void
-ox_sax_drive_error_at(SaxDrive dr, const char *msg, int line, int col) {
+ox_sax_drive_error_at(SaxDrive dr, const char *msg, int pos, int line, int col) {
     if (dr->has.error) {
         VALUE   args[3];
 
         args[0] = rb_str_new2(msg);
         args[1] = LONG2NUM(line);
         args[2] = LONG2NUM(col);
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, args[1]);
 	}
@@ -236,7 +244,7 @@ ox_sax_drive_error_at(SaxDrive dr, const char *msg, int line, int col) {
 
 void
 ox_sax_drive_error(SaxDrive dr, const char *msg) {
-    ox_sax_drive_error_at(dr, msg, dr->buf.line, dr->buf.col);
+    ox_sax_drive_error_at(dr, msg, dr->buf.pos, dr->buf.line, dr->buf.col);
 }
 
 static char
@@ -292,8 +300,9 @@ parse(SaxDrive dr) {
 		} else {
 		    int	i;
 		    int	spaced = 0;
+		    int	pos = dr->buf.pos + 1;
 		    int	line = dr->buf.line;
-		    int	col = dr->buf.col;
+		    int	col = dr->buf.col + 1;
 
 		    if (is_white(c)) {
 			spaced = 1;
@@ -305,7 +314,7 @@ parse(SaxDrive dr) {
 		    }
 		    if (0 == strncmp("DOCTYPE", dr->buf.str, 7)) {
 			if (spaced) {
-			    ox_sax_drive_error_at(dr, WRONG_CHAR "<!DOCTYPE can not included spaces", line, col);
+			    ox_sax_drive_error_at(dr, WRONG_CHAR "<!DOCTYPE can not included spaces", pos, line, col);
 			}
 			if (START_STATE != state) {
 			    ox_sax_drive_error(dr, OUT_OF_ORDER "DOCTYPE can not come after an element");
@@ -319,7 +328,7 @@ parse(SaxDrive dr) {
 			c = read_doctype(dr);
 		    } else if (0 == strncmp("[CDATA[", dr->buf.str, 7)) {
 			if (spaced) {
-			    ox_sax_drive_error_at(dr, WRONG_CHAR "<![CDATA[ can not included spaces", line, col);
+			    ox_sax_drive_error_at(dr, WRONG_CHAR "<![CDATA[ can not included spaces", pos, line, col);
 			}
 			c = read_cdata(dr);
 		    } else if (0 == strncasecmp("[CDATA[", dr->buf.str, 7)) {
@@ -331,7 +340,7 @@ parse(SaxDrive dr) {
 			if (0 != parent) {
 			    parent->childCnt++;
 			}
-			ox_sax_drive_error_at(dr, WRONG_CHAR "DOCTYPE, CDATA, or comment expected", line, col);
+			ox_sax_drive_error_at(dr, WRONG_CHAR "DOCTYPE, CDATA, or comment expected", pos, line, col);
 			c = read_name_token(dr);
 			if ('>' == c) {
 			    c = buf_get(&dr->buf);
@@ -341,8 +350,9 @@ parse(SaxDrive dr) {
 		break;
 	    case '/': /* element end */
 		parent = stack_peek(&dr->stack);
-		if (0 != parent && 0 == parent->childCnt) {
+		if (0 != parent && 0 == parent->childCnt && dr->has.text) {
 		    VALUE	args[1];
+		    int		pos = dr->buf.pos;
 		    int		line = dr->buf.line;
 		    int		col = dr->buf.col - 1;
 
@@ -356,6 +366,9 @@ parse(SaxDrive dr) {
 			rb_funcall(args[0], ox_force_encoding_id, 1, dr->encoding);
 		    }
 #endif
+		    if (dr->has.pos) {
+			rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+		    }
 		    if (dr->has.line) {
 			rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 		    }
@@ -393,6 +406,9 @@ parse(SaxDrive dr) {
 	char	msg[256];
 	Nv	sp;
 
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(dr->buf.pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(dr->buf.line));
 	}
@@ -401,7 +417,7 @@ parse(SaxDrive dr) {
 	}
 	for (sp = dr->stack.tail - 1; dr->stack.head <= sp; sp--) {
 	    snprintf(msg, sizeof(msg) - 1, "%selement '%s' not closed", EL_MISMATCH, sp->name);
-	    ox_sax_drive_error_at(dr, msg, dr->buf.line, dr->buf.col);
+	    ox_sax_drive_error_at(dr, msg, dr->buf.pos, dr->buf.line, dr->buf.col);
 	    if (dr->has.end_element) {
 		VALUE       args[1];
 
@@ -448,6 +464,7 @@ read_instruction(SaxDrive dr) {
     char	*cend;
     VALUE	target = Qnil;
     int		is_xml;
+    int		pos = dr->buf.pos - 1;
     int		line = dr->buf.line;
     int		col = dr->buf.col - 1;
 
@@ -462,6 +479,9 @@ read_instruction(SaxDrive dr) {
     if (dr->has.instruct) {
         VALUE       args[1];
 
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -472,6 +492,7 @@ read_instruction(SaxDrive dr) {
         rb_funcall2(dr->handler, ox_instruct_id, 1, args);
     }
     buf_protect(&dr->buf);
+    pos = dr->buf.pos;
     line = dr->buf.line;
     col = dr->buf.col;
     read_content(dr, content, sizeof(content) - 1);
@@ -487,7 +508,7 @@ read_instruction(SaxDrive dr) {
 	    VALUE   args[1];
 
 	    if (dr->options.convert_special) {
-		ox_sax_collapse_special(dr, content, line, col);
+		ox_sax_collapse_special(dr, content, pos, line, col);
 	    }
 	    args[0] = rb_str_new2(content);
 #if HAS_ENCODING_SUPPORT
@@ -502,6 +523,9 @@ read_instruction(SaxDrive dr) {
 	    if (dr->has.line) {
 		rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	    }
+	    if (dr->has.pos) {
+		rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	    }
 	    if (dr->has.column) {
 		rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	    }
@@ -510,13 +534,14 @@ read_instruction(SaxDrive dr) {
 	dr->buf.tail = cend;
 	c = buf_get(&dr->buf);
     } else {
+	pos = dr->buf.pos;
 	line = dr->buf.line;
 	col = dr->buf.col;
 	c = buf_next_non_white(&dr->buf);
 	if ('>' == c) {
 	    c = buf_get(&dr->buf);
 	} else {
-	    ox_sax_drive_error_at(dr, NO_TERM "instruction not terminated", line, col);
+	    ox_sax_drive_error_at(dr, NO_TERM "instruction not terminated", pos, line, col);
 	    if ('>' == c) {
 		c = buf_get(&dr->buf);
 	    }
@@ -525,6 +550,9 @@ read_instruction(SaxDrive dr) {
     if (dr->has.end_instruct) {
         VALUE       args[1];
 
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -580,16 +608,17 @@ read_delimited(SaxDrive dr, char end) {
     return c;
 }
 
-/* Entered after the "<!DOCTYPE" sequence. Ready to read the rest.
+/* Entered after the "<!DOCTYPE " sequence. Ready to read the rest.
  */
 static char
 read_doctype(SaxDrive dr) {
+    int		pos = dr->buf.pos - 9;
     int		line = dr->buf.line;
-    int		col = dr->buf.col - 10;
+    int		col = dr->buf.col - 9;
     char	*s;
     Nv		parent = stack_peek(&dr->stack);
 
-    buf_backup(&dr->buf); /* back up to the start in case the cdata is empty */
+    buf_backup(&dr->buf); /* back up to the start in case the doctype is empty */
     buf_protect(&dr->buf);
     read_delimited(dr, '>');
     if (dr->options.smart && 0 == dr->hints) {
@@ -605,6 +634,9 @@ read_doctype(SaxDrive dr) {
     if (dr->has.doctype) {
         VALUE       args[1];
 
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -626,8 +658,9 @@ read_cdata(SaxDrive dr) {
     char        	c;
     char        	zero = '\0';
     int         	end = 0;
+    int			pos = dr->buf.pos - 9;
     int			line = dr->buf.line;
-    int			col = dr->buf.col - 10;
+    int			col = dr->buf.col - 9;
     struct _CheckPt	cp = CHECK_PT_INIT;
     Nv			parent = stack_peek(&dr->stack);
 
@@ -691,6 +724,9 @@ read_cdata(SaxDrive dr) {
 	    rb_funcall(args[0], ox_force_encoding_id, 1, dr->encoding);
         }
 #endif
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -714,6 +750,7 @@ read_comment(SaxDrive dr) {
     char        	c;
     char        	zero = '\0';
     int         	end = 0;
+    int			pos = dr->buf.pos - 4;
     int			line = dr->buf.line;
     int			col = dr->buf.col - 4;
     struct _CheckPt	cp = CHECK_PT_INIT;
@@ -775,6 +812,9 @@ read_comment(SaxDrive dr) {
 	    rb_funcall(args[0], ox_force_encoding_id, 1, dr->encoding);
         }
 #endif
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -800,13 +840,13 @@ read_element_start(SaxDrive dr) {
     volatile VALUE	name = Qnil;
     char        	c;
     int			closed;
+    int			pos = dr->buf.pos;
     int			line = dr->buf.line;
-    int			col = dr->buf.col - 1;
+    int			col = dr->buf.col;
     Hint		h = 0;
     int			stackless = 0;
     Nv			parent = stack_peek(&dr->stack);
 
-
     if ('\0' == (c = read_name_token(dr))) {
         return '\0';
     }
@@ -838,7 +878,7 @@ read_element_start(SaxDrive dr) {
 			     INV_ELEMENT, dr->buf.str, dr->hints->name);
 		    ox_sax_drive_error(dr, msg);
 		    stack_pop(&dr->stack);
-		    end_element_cb(dr, top_nv->val, line, col);
+		    end_element_cb(dr, top_nv->val, pos, line, col);
 		    top_nv = stack_peek(&dr->stack);
 		}
 		if (0 != h->parents) {
@@ -864,6 +904,9 @@ read_element_start(SaxDrive dr) {
     if (dr->has.start_element) {
         VALUE       args[1];
 
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -890,11 +933,20 @@ read_element_start(SaxDrive dr) {
     }
     if (closed) {
 	c = buf_next_non_white(&dr->buf);
+	pos = dr->buf.pos;
 	line = dr->buf.line;
-	col = dr->buf.col - 1;
-	end_element_cb(dr, name, line, col);
+	col = dr->buf.col;
+	end_element_cb(dr, name, pos, line, col);
     } else if (stackless) {
-	end_element_cb(dr, name, line, col);
+	end_element_cb(dr, name, pos, line, col);
+    } else if (0 != h && h->jump) {
+	stack_push(&dr->stack, ename, name, h);
+	if ('>' != c) {
+	    ox_sax_drive_error(dr, WRONG_CHAR "element not closed");
+	    return c;
+	}
+	read_jump(dr, h->name);
+	return '<';
     } else {
 	stack_push(&dr->stack, ename, name, h);
     }
@@ -923,13 +975,17 @@ static char
 read_element_end(SaxDrive dr) {
     VALUE       name = Qnil;
     char        c;
+    int		pos = dr->buf.pos - 1;
     int		line = dr->buf.line;
-    int		col = dr->buf.col - 2;
+    int		col = dr->buf.col - 1;
     Nv		nv;
 
     if ('\0' == (c = read_name_token(dr))) {
         return '\0';
     }
+    if (is_white(c)) {
+	c = buf_next_non_white(&dr->buf);
+    }
     // c should be > and current is one past so read another char
     c = buf_get(&dr->buf);
     nv = stack_peek(&dr->stack);
@@ -949,15 +1005,18 @@ read_element_end(SaxDrive dr) {
 		// Just close normally
 		name = str2sym(dr, dr->buf.str, 0);
 		snprintf(msg, sizeof(msg) - 1, "%selement '%s' should not have a separate close element", EL_MISMATCH, dr->buf.str);
-		ox_sax_drive_error_at(dr, msg, line, col);
+		ox_sax_drive_error_at(dr, msg, pos, line, col);
 		return c;
 	    } else {
 		snprintf(msg, sizeof(msg) - 1, "%selement '%s' closed but not opened", EL_MISMATCH, dr->buf.str);
-		ox_sax_drive_error_at(dr, msg, line, col);
+		ox_sax_drive_error_at(dr, msg, pos, line, col);
 		name = str2sym(dr, dr->buf.str, 0);
 		if (dr->has.start_element) {
 		    VALUE       args[1];
 
+		    if (dr->has.pos) {
+			rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+		    }
 		    if (dr->has.line) {
 			rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 		    }
@@ -976,7 +1035,10 @@ read_element_end(SaxDrive dr) {
 		name = n2->val;
 	    } else {
 		snprintf(msg, sizeof(msg) - 1, "%selement '%s' close does not match '%s' open", EL_MISMATCH, dr->buf.str, nv->name);
-		ox_sax_drive_error_at(dr, msg, line, col);
+		ox_sax_drive_error_at(dr, msg, pos, line, col);
+		if (dr->has.pos) {
+		    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+		}
 		if (dr->has.line) {
 		    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 		}
@@ -992,7 +1054,7 @@ read_element_end(SaxDrive dr) {
 	    }
 	}
     }
-    end_element_cb(dr, name, line, col);
+    end_element_cb(dr, name, pos, line, col);
 
     return c;
 }
@@ -1001,6 +1063,7 @@ static char
 read_text(SaxDrive dr) {
     VALUE	args[1];
     char        c;
+    int		pos = dr->buf.pos;
     int		line = dr->buf.line;
     int		col = dr->buf.col - 1;
     Nv		parent = stack_peek(&dr->stack);
@@ -1044,6 +1107,9 @@ read_text(SaxDrive dr) {
 	parent->childCnt++;
     }
     if (dr->has.value) {
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -1054,7 +1120,7 @@ read_text(SaxDrive dr) {
         rb_funcall2(dr->handler, ox_value_id, 1, args);
     } else if (dr->has.text) {
         if (dr->options.convert_special) {
-            ox_sax_collapse_special(dr, dr->buf.str, line, col);
+            ox_sax_collapse_special(dr, dr->buf.str, pos, line, col);
         }
 	switch (dr->options.skip) {
 	case CrSkip:
@@ -1076,6 +1142,9 @@ read_text(SaxDrive dr) {
 	    rb_funcall(args[0], ox_force_encoding_id, 1, dr->encoding);
         }
 #endif
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}
@@ -1089,10 +1158,97 @@ read_text(SaxDrive dr) {
     return c;
 }
 
+static int
+read_jump_term(Buf buf, const char *pat) {
+    struct _CheckPt	cp;
+
+    buf_checkpoint(buf, &cp); // right after <
+    if ('/' != buf_next_non_white(buf)) {
+	return 0;
+    }
+    if (*pat != buf_next_non_white(buf)) {
+	return 0;
+    }
+    for (pat++; '\0' != *pat; pat++) {
+	if (*pat != buf_get(buf)) {
+	    return 0;
+	}
+    }
+    if ('>' != buf_next_non_white(buf)) {
+	return 0;
+    }
+    buf_checkback(buf, &cp);
+    return 1;
+}
+
+static char
+read_jump(SaxDrive dr, const char *pat) {
+    VALUE	args[1];
+    char        c;
+    int		pos = dr->buf.pos;
+    int		line = dr->buf.line;
+    int		col = dr->buf.col - 1;
+    Nv		parent = stack_peek(&dr->stack);
+
+    buf_protect(&dr->buf);
+    while (1) {
+	c = buf_get(&dr->buf);
+	switch(c) {
+	case '<':
+	    if (read_jump_term(&dr->buf, pat)) {
+		goto END_OF_BUF;
+		break;
+	    }
+	    break;
+	case '\0':
+            ox_sax_drive_error(dr, NO_TERM "not terminated");
+	    goto END_OF_BUF;
+	    break;
+	default:
+	    break;
+	}
+    }
+ END_OF_BUF:
+    if ('\0' != c) {
+	*(dr->buf.tail - 1) = '\0';
+    }
+    if (0 != parent) {
+	parent->childCnt++;
+    }
+    if (dr->has.text) {
+        args[0] = rb_str_new2(dr->buf.str);
+#if HAS_ENCODING_SUPPORT
+        if (0 != dr->encoding) {
+            rb_enc_associate(args[0], dr->encoding);
+        }
+#elif HAS_PRIVATE_ENCODING
+        if (Qnil != dr->encoding) {
+	    rb_funcall(args[0], ox_force_encoding_id, 1, dr->encoding);
+        }
+#endif
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
+	if (dr->has.line) {
+	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
+	}
+	if (dr->has.column) {
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
+	}
+        rb_funcall2(dr->handler, ox_text_id, 1, args);
+    }
+    dr->buf.str = 0;
+    if ('\0' != c) {
+	*(dr->buf.tail - 1) = '<';
+    }
+    return c;
+}
+
 static char
 read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req) {
     VALUE       name = Qnil;
     int         is_encoding = 0;
+    int		pos;
     int		line;
     int		col;
     char	*attr_value;
@@ -1108,8 +1264,9 @@ read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req)
 	    ox_sax_drive_error(dr, NO_TERM "attributes not terminated");
 	    return '\0';
         }
+	pos = dr->buf.pos + 1;
 	line = dr->buf.line;
-	col = dr->buf.col;
+	col = dr->buf.col + 1;
         if ('\0' == (c = read_name_token(dr))) {
 	    ox_sax_drive_error(dr, NO_TERM "error reading token");
 	    return '\0';
@@ -1132,8 +1289,9 @@ read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req)
 		attr_value = (char*)"";
 	    }
         } else {
+	    pos = dr->buf.pos + 1;
 	    line = dr->buf.line;
-	    col = dr->buf.col;
+	    col = dr->buf.col + 1;
 	    c = read_quoted_value(dr);
 	    attr_value = dr->buf.str;
 	    if (is_encoding) {
@@ -1150,6 +1308,9 @@ read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req)
         if (dr->has.attr_value) {
             VALUE       args[2];
 
+	    if (dr->has.pos) {
+		rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	    }
 	    if (dr->has.line) {
 		rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	    }
@@ -1163,7 +1324,7 @@ read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req)
             VALUE       args[2];
 
             args[0] = name;
-            ox_sax_collapse_special(dr, dr->buf.str, line, col);
+            ox_sax_collapse_special(dr, dr->buf.str, pos, line, col);
             args[1] = rb_str_new2(attr_value);
 #if HAS_ENCODING_SUPPORT
             if (0 != dr->encoding) {
@@ -1174,6 +1335,9 @@ read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req)
 		rb_funcall(args[1], ox_force_encoding_id, 1, dr->encoding);
 	    }
 #endif
+	    if (dr->has.pos) {
+		rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	    }
 	    if (dr->has.line) {
 		rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	    }
@@ -1319,7 +1483,7 @@ read_10_uint64(char *b, uint64_t *up) {
 }
 
 int
-ox_sax_collapse_special(SaxDrive dr, char *str, int line, int col) {
+ox_sax_collapse_special(SaxDrive dr, char *str, int pos, int line, int col) {
     char        *s = str;
     char        *b = str;
 
@@ -1407,7 +1571,7 @@ ox_sax_collapse_special(SaxDrive dr, char *str, int line, int col) {
                 c = '\'';
                 s += 5;
             } else {
-		ox_sax_drive_error_at(dr, NO_TERM "special character does not end with a semicolon", line, col);
+		ox_sax_drive_error_at(dr, NO_TERM "special character does not end with a semicolon", pos, line, col);
 		c = '&';
             }
             *b++ = (char)c;
@@ -1435,7 +1599,7 @@ hint_clear_empty(SaxDrive dr) {
 	    break;
 	}
 	if (nv->hint->empty) {
-	    end_element_cb(dr, nv->val, dr->buf.line, dr->buf.col);
+	    end_element_cb(dr, nv->val, dr->buf.pos, dr->buf.line, dr->buf.col);
 	    stack_pop(&dr->stack);
 	} else {
 	    break;
@@ -1460,7 +1624,7 @@ hint_try_close(SaxDrive dr, const char *name) {
 	    break;
 	}
 	if (nv->hint->empty) {
-	    end_element_cb(dr, nv->val, dr->buf.line, dr->buf.col);
+	    end_element_cb(dr, nv->val, dr->buf.pos, dr->buf.line, dr->buf.col);
 	    dr->stack.tail = nv;
 	} else {
 	    break;
@@ -1470,8 +1634,11 @@ hint_try_close(SaxDrive dr, const char *name) {
 }
 
 static void
-end_element_cb(SaxDrive dr, VALUE name, int line, int col) {
+end_element_cb(SaxDrive dr, VALUE name, int pos, int line, int col) {
     if (dr->has.end_element) {
+	if (dr->has.pos) {
+	    rb_ivar_set(dr->handler, ox_at_pos_id, LONG2NUM(pos));
+	}
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(line));
 	}

data/ext/ox/sax.h

@@ -41,7 +41,7 @@ extern void	ox_collapse_return(char *str);
 extern void	ox_sax_parse(VALUE handler, VALUE io, SaxOptions options);
 extern void	ox_sax_drive_cleanup(SaxDrive dr);
 extern void	ox_sax_drive_error(SaxDrive dr, const char *msg);
-extern int	ox_sax_collapse_special(SaxDrive dr, char *str, int line, int col);
+extern int	ox_sax_collapse_special(SaxDrive dr, char *str, int pos, int line, int col);
 
 extern VALUE	ox_sax_value_class;
 

data/ext/ox/sax_as.c

@@ -119,7 +119,7 @@ sax_value_as_s(VALUE self) {
 	return Qnil;
     }
     if (dr->options.convert_special) {
-	ox_sax_collapse_special(dr, dr->buf.str, dr->buf.line, dr->buf.col);
+	ox_sax_collapse_special(dr, dr->buf.str, dr->buf.pos, dr->buf.line, dr->buf.col);
     }
     switch (dr->options.skip) {
     case CrSkip:

data/ext/ox/sax_buf.c

@@ -57,8 +57,10 @@ ox_sax_buf_init(Buf buf, VALUE io) {
     buf->read_end = buf->head;
     buf->pro = 0;
     buf->str = 0;
+    buf->pos = 0;
     buf->line = 1;
     buf->col = 0;
+    buf->pro_pos = 1;
     buf->pro_line = 1;
     buf->pro_col = 0;
     buf->dr = 0;

data/ext/ox/sax_buf.h

@@ -14,8 +14,10 @@ typedef struct _Buf {
     char	*read_end;      /* one past last character read */
     char       	*pro;           /* protection start, buffer can not slide past this point */
     char        *str;           /* start of current string being read */
+    int		pos;
     int		line;
     int		col;
+    int		pro_pos;
     int		pro_line;
     int		pro_col;
     int		(*read_func)(struct _Buf *buf);
@@ -29,12 +31,13 @@ typedef struct _Buf {
 
 typedef struct _CheckPt {
     int		pro_dif;
+    int		pos;
     int		line;
     int		col;
     char	c;
 } *CheckPt;
 
-#define CHECK_PT_INIT { -1, 0, 0, '\0' }
+#define CHECK_PT_INIT { -1, 0, 0, 0, '\0' }
 
 extern void	ox_sax_buf_init(Buf buf, VALUE io);
 extern int	ox_sax_buf_read(Buf buf);
@@ -50,9 +53,11 @@ buf_get(Buf buf) {
     if ('\n' == *buf->tail) {
         buf->line++;
         buf->col = 0;
+    } else {
+	buf->col++;
     }
-    buf->col++;
-    
+    buf->pos++;
+
     return *buf->tail++;
 }
 
@@ -60,6 +65,7 @@ static inline void
 buf_backup(Buf buf) {
     buf->tail--;
     buf->col--;
+    buf->pos--;
     if (0 >= buf->col) {
 	buf->line--;
 	// allow col to be negative since we never backup twice in a row
@@ -70,6 +76,7 @@ static inline void
 buf_protect(Buf buf) {
     buf->pro = buf->tail;
     buf->str = buf->tail; // can't have str before pro
+    buf->pro_pos = buf->pos;
     buf->pro_line = buf->line;
     buf->pro_col = buf->col;
 }
@@ -77,6 +84,7 @@ buf_protect(Buf buf) {
 static inline void
 buf_reset(Buf buf) {
     buf->tail = buf->pro;
+    buf->pos = buf->pro_pos;
     buf->line = buf->pro_line;
     buf->col = buf->pro_col;
 }
@@ -152,6 +160,7 @@ is_white(char c) {
 static inline void
 buf_checkpoint(Buf buf, CheckPt cp) {
     cp->pro_dif = (int)(buf->tail - buf->pro);
+    cp->pos = buf->pos;
     cp->line = buf->line;
     cp->col = buf->col;
     cp->c = *(buf->tail - 1);
@@ -165,6 +174,7 @@ buf_checkset(CheckPt cp) {
 static inline char
 buf_checkback(Buf buf, CheckPt cp) {
     buf->tail = buf->pro + cp->pro_dif;
+    buf->pos = cp->pos;
     buf->line = cp->line;
     buf->col = cp->col;
     return cp->c;

data/ext/ox/sax_has.h

@@ -20,6 +20,7 @@ typedef struct _Has {
     int         start_element;
     int         end_element;
     int         error;
+    int		pos;
     int		line;
     int		column;
 } *Has;
@@ -44,6 +45,7 @@ has_init(Has has, VALUE handler) {
     has->start_element = respond_to(handler, ox_start_element_id);
     has->end_element = respond_to(handler, ox_end_element_id);
     has->error = respond_to(handler, ox_error_id);
+    has->pos = (Qtrue == rb_ivar_defined(handler, ox_at_pos_id));
     has->line = (Qtrue == rb_ivar_defined(handler, ox_at_line_id));
     has->column = (Qtrue == rb_ivar_defined(handler, ox_at_column_id));
 }

data/ext/ox/sax_hint.c

@@ -26,127 +26,127 @@ static const char	*table_0[] = { "table", 0 };
 static const char	*tr_0[] = { "tr", 0 };
 
 static struct _Hint	html_hint_array[] = {
-    { "a", 0, 0, 0 },
-    { "abbr", 0, 0, 0 },
-    { "acronym", 0, 0, 0 },
-    { "address", 0, 0, 0 },
-    { "applet", 0, 0, 0 },
-    { "area", 1, 0, map_0 },
-    { "article", 0, 0, 0 },
-    { "aside", 0, 0, 0 },
-    { "audio", 0, 0, 0 },
-    { "b", 0, 0, 0 },
-    { "base", 1, 0, head_0 },
-    { "basefont", 1, 0, head_0 },
-    { "bdi", 0, 0, 0 },
-    { "bdo", 0, 1, 0 },
-    { "big", 0, 0, 0 },
-    { "blockquote", 0, 0, 0 },
-    { "body", 0, 0, html_0 },
-    { "br", 1, 0, 0 },
-    { "button", 0, 0, 0 },
-    { "canvas", 0, 0, 0 },
-    { "caption", 0, 0, table_0 },
-    { "center", 0, 0, 0 },
-    { "cite", 0, 0, 0 },
-    { "code", 0, 0, 0 },
-    { "col", 1, 0, colgroup_0 },
-    { "colgroup", 0, 0, 0 },
-    { "command", 1, 0, 0 },
-    { "datalist", 0, 0, 0 },
-    { "dd", 0, 0, dl_0 },
-    { "del", 0, 0, 0 },
-    { "details", 0, 0, 0 },
-    { "dfn", 0, 0, 0 },
-    { "dialog", 0, 0, dt_th_0 },
-    { "dir", 0, 0, 0 },
-    { "div", 0, 1, 0 },
-    { "dl", 0, 0, 0 },
-    { "dt", 0, 1, dl_0 },
-    { "em", 0, 0, 0 },
-    { "embed", 1, 0, 0 },
-    { "fieldset", 0, 0, 0 },
-    { "figcaption", 0, 0, figure_0 },
-    { "figure", 0, 0, 0 },
-    { "font", 0, 1, 0 },
-    { "footer", 0, 0, 0 },
-    { "form", 0, 0, 0 },
-    { "frame", 1, 0, frameset_0 },
-    { "frameset", 0, 0, 0 },
-    { "h1", 0, 0, 0 },
-    { "h2", 0, 0, 0 },
-    { "h3", 0, 0, 0 },
-    { "h4", 0, 0, 0 },
-    { "h5", 0, 0, 0 },
-    { "h6", 0, 0, 0 },
-    { "head", 0, 0, html_0 },
-    { "header", 0, 0, 0 },
-    { "hgroup", 0, 0, 0 },
-    { "hr", 1, 0, 0 },
-    { "html", 0, 0, 0 },
-    { "i", 0, 0, 0 },
-    { "iframe", 1, 0, 0 },
-    { "img", 1, 0, 0 },
-    { "input", 1, 0, 0 }, // somewhere under a form_0
-    { "ins", 0, 0, 0 },
-    { "kbd", 0, 0, 0 },
-    { "keygen", 1, 0, 0 },
-    { "label", 0, 0, 0 }, // somewhere under a form_0
-    { "legend", 0, 0, fieldset_0 },
-    { "li", 0, 0, ol_ul_menu_0 },
-    { "link", 1, 0, head_0 },
-    { "map", 0, 0, 0 },
-    { "mark", 0, 0, 0 },
-    { "menu", 0, 0, 0 },
-    { "meta", 1, 0, head_0 },
-    { "meter", 0, 0, 0 },
-    { "nav", 0, 0, 0 },
-    { "noframes", 0, 0, 0 },
-    { "noscript", 0, 0, 0 },
-    { "object", 0, 0, 0 },
-    { "ol", 0, 1, 0 },
-    { "optgroup", 0, 0, 0 },
-    { "option", 0, 0, optgroup_select_datalist_0 },
-    { "output", 0, 0, 0 },
-    { "p", 0, 0, 0 },
-    { "param", 1, 0, 0 },
-    { "pre", 0, 0, 0 },
-    { "progress", 0, 0, 0 },
-    { "q", 0, 0, 0 },
-    { "rp", 0, 0, ruby_0 },
-    { "rt", 0, 0, ruby_0 },
-    { "ruby", 0, 0, 0 },
-    { "s", 0, 0, 0 },
-    { "samp", 0, 0, 0 },
-    { "script", 0, 0, 0 },
-    { "section", 0, 1, 0 },
-    { "select", 0, 0, 0 },
-    { "small", 0, 0, 0 },
-    { "source", 0, 0, audio_video_0 },
-    { "span", 0, 1, 0 },
-    { "strike", 0, 0, 0 },
-    { "strong", 0, 0, 0 },
-    { "style", 0, 0, 0 },
-    { "sub", 0, 0, 0 },
-    { "summary", 0, 0, details_0 },
-    { "sup", 0, 0, 0 },
-    { "table", 0, 0, 0 },
-    { "tbody", 0, 0, table_0 },
-    { "td", 0, 0, tr_0 },
-    { "textarea", 0, 0, 0 },
-    { "tfoot", 0, 0, table_0 },
-    { "th", 0, 0, tr_0 },
-    { "thead", 0, 0, table_0 },
-    { "time", 0, 0, 0 },
-    { "title", 0, 0, head_0 },
-    { "tr", 0, 0, table_0 },
-    { "track", 1, 0, audio_video_0 },
-    { "tt", 0, 0, 0 },
-    { "u", 0, 0, 0 },
-    { "ul", 0, 0, 0 },
-    { "var", 0, 0, 0 },
-    { "video", 0, 0, 0 },
-    { "wbr", 1, 0, 0 },
+    { "a", 0, 0, 0, 0 },
+    { "abbr", 0, 0, 0, 0 },
+    { "acronym", 0, 0, 0, 0 },
+    { "address", 0, 0, 0, 0 },
+    { "applet", 0, 0, 0, 0 },
+    { "area", 1, 0, 0, map_0 },
+    { "article", 0, 0, 0, 0 },
+    { "aside", 0, 0, 0, 0 },
+    { "audio", 0, 0, 0, 0 },
+    { "b", 0, 0, 0, 0 },
+    { "base", 1, 0, 0, head_0 },
+    { "basefont", 1, 0, 0, head_0 },
+    { "bdi", 0, 0, 0, 0 },
+    { "bdo", 0, 1, 0, 0 },
+    { "big", 0, 0, 0, 0 },
+    { "blockquote", 0, 0, 0, 0 },
+    { "body", 0, 0, 0, html_0 },
+    { "br", 1, 0, 0, 0 },
+    { "button", 0, 0, 0, 0 },
+    { "canvas", 0, 0, 0, 0 },
+    { "caption", 0, 0, 0, table_0 },
+    { "center", 0, 0, 0, 0 },
+    { "cite", 0, 0, 0, 0 },
+    { "code", 0, 0, 0, 0 },
+    { "col", 1, 0, 0, colgroup_0 },
+    { "colgroup", 0, 0, 0, 0 },
+    { "command", 1, 0, 0, 0 },
+    { "datalist", 0, 0, 0, 0 },
+    { "dd", 0, 0, 0, dl_0 },
+    { "del", 0, 0, 0, 0 },
+    { "details", 0, 0, 0, 0 },
+    { "dfn", 0, 0, 0, 0 },
+    { "dialog", 0, 0, 0, dt_th_0 },
+    { "dir", 0, 0, 0, 0 },
+    { "div", 0, 1, 0, 0 },
+    { "dl", 0, 0, 0, 0 },
+    { "dt", 0, 1, 0, dl_0 },
+    { "em", 0, 0, 0, 0 },
+    { "embed", 1, 0, 0, 0 },
+    { "fieldset", 0, 0, 0, 0 },
+    { "figcaption", 0, 0, 0, figure_0 },
+    { "figure", 0, 0, 0, 0 },
+    { "font", 0, 1, 0, 0 },
+    { "footer", 0, 0, 0, 0 },
+    { "form", 0, 0, 0, 0 },
+    { "frame", 1, 0, 0, frameset_0 },
+    { "frameset", 0, 0, 0, 0 },
+    { "h1", 0, 0, 0, 0 },
+    { "h2", 0, 0, 0, 0 },
+    { "h3", 0, 0, 0, 0 },
+    { "h4", 0, 0, 0, 0 },
+    { "h5", 0, 0, 0, 0 },
+    { "h6", 0, 0, 0, 0 },
+    { "head", 0, 0, 0, html_0 },
+    { "header", 0, 0, 0, 0 },
+    { "hgroup", 0, 0, 0, 0 },
+    { "hr", 1, 0, 0, 0 },
+    { "html", 0, 0, 0, 0 },
+    { "i", 0, 0, 0, 0 },
+    { "iframe", 1, 0, 0, 0 },
+    { "img", 1, 0, 0, 0 },
+    { "input", 1, 0, 0, 0 }, // somewhere under a form_0
+    { "ins", 0, 0, 0, 0 },
+    { "kbd", 0, 0, 0, 0 },
+    { "keygen", 1, 0, 0, 0 },
+    { "label", 0, 0, 0, 0 }, // somewhere under a form_0
+    { "legend", 0, 0, 0, fieldset_0 },
+    { "li", 0, 0, 0, ol_ul_menu_0 },
+    { "link", 1, 0, 0, head_0 },
+    { "map", 0, 0, 0, 0 },
+    { "mark", 0, 0, 0, 0 },
+    { "menu", 0, 0, 0, 0 },
+    { "meta", 1, 0, 0, head_0 },
+    { "meter", 0, 0, 0, 0 },
+    { "nav", 0, 0, 0, 0 },
+    { "noframes", 0, 0, 0, 0 },
+    { "noscript", 0, 0, 0, 0 },
+    { "object", 0, 0, 0, 0 },
+    { "ol", 0, 1, 0, 0 },
+    { "optgroup", 0, 0, 0, 0 },
+    { "option", 0, 0, 0, optgroup_select_datalist_0 },
+    { "output", 0, 0, 0, 0 },
+    { "p", 0, 0, 0, 0 },
+    { "param", 1, 0, 0, 0 },
+    { "pre", 0, 0, 0, 0 },
+    { "progress", 0, 0, 0, 0 },
+    { "q", 0, 0, 0, 0 },
+    { "rp", 0, 0, 0, ruby_0 },
+    { "rt", 0, 0, 0, ruby_0 },
+    { "ruby", 0, 0, 0, 0 },
+    { "s", 0, 0, 0, 0 },
+    { "samp", 0, 0, 0, 0 },
+    { "script", 0, 0, 1, 0 },
+    { "section", 0, 1, 0, 0 },
+    { "select", 0, 0, 0, 0 },
+    { "small", 0, 0, 0, 0 },
+    { "source", 0, 0, 0, audio_video_0 },
+    { "span", 0, 1, 0, 0 },
+    { "strike", 0, 0, 0, 0 },
+    { "strong", 0, 0, 0, 0 },
+    { "style", 0, 0, 0, 0 },
+    { "sub", 0, 0, 0, 0 },
+    { "summary", 0, 0, 0, details_0 },
+    { "sup", 0, 0, 0, 0 },
+    { "table", 0, 0, 0, 0 },
+    { "tbody", 0, 0, 0, table_0 },
+    { "td", 0, 0, 0, tr_0 },
+    { "textarea", 0, 0, 0, 0 },
+    { "tfoot", 0, 0, 0, table_0 },
+    { "th", 0, 0, 0, tr_0 },
+    { "thead", 0, 0, 0, table_0 },
+    { "time", 0, 0, 0, 0 },
+    { "title", 0, 0, 0, head_0 },
+    { "tr", 0, 0, 0, table_0 },
+    { "track", 1, 0, 0, audio_video_0 },
+    { "tt", 0, 0, 0, 0 },
+    { "u", 0, 0, 0, 0 },
+    { "ul", 0, 0, 0, 0 },
+    { "var", 0, 0, 0, 0 },
+    { "video", 0, 0, 0, 0 },
+    { "wbr", 1, 0, 0, 0 },
 };
 static struct _Hints	html_hints = {
     "HTML",

data/ext/ox/sax_hint.h

@@ -10,6 +10,7 @@ typedef struct _Hint {
     const char	*name;
     char	empty;	// must be closed or close auto it, not error
     char	nest;	// nesting allowed
+    char	jump;	// jump to end <script> ... </script>
     const char	**parents;
 } *Hint;
 

data/lib/ox/sax.rb

@@ -48,10 +48,12 @@ module Ox
   # Initializing @line in the initializer will cause that variable to be updated
   # before each callback with the XML line number. The same is true for the
   # @column but it will be updated with the column in the XML file that is the
-  # start of the element or node just read.
+  # start of the element or node just read. @pos if defined will hold the number
+  # of bytes from the start of the document.
   class Sax
     # Create a new instance of the Sax handler class.
     def initialize()
+      #@pos = nil
       #@line = nil
       #@column = nil
     end

data/lib/ox/version.rb

@@ -1,5 +1,5 @@
 
 module Ox
   # Current version of the module. 
-  VERSION = '2.2.0'
+  VERSION = '2.2.1'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: ox
 version: !ruby/object:Gem::Version
-  version: 2.2.0
+  version: 2.2.1
 platform: ruby
 authors:
 - Peter Ohler
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-04-20 00:00:00.000000000 Z
+date: 2015-07-30 00:00:00.000000000 Z
 dependencies: []
 description: "A fast XML parser and object serializer that uses only standard C lib.\n
   \           \nOptimized XML (Ox), as the name implies was written to provide speed