ox 2.0.2 → 2.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: ae0dbc4e3a17970ece45a543a56181be5218e13f
-  data.tar.gz: a667157d26ff14f45d54cdc64b1f7f74a9dac965
+  metadata.gz: 9d03526c81d518faeee16a0866d7dcd58652cbb8
+  data.tar.gz: 606f087bcfadf31e8f3800424a7253bfee95dc17
 SHA512:
-  metadata.gz: 59334b95d88b70fe553ad03fd12adcb7c53d80aa04f743516515c841cb350aabad840a836b97a775c5da417e07a89047780ddfa2d5452ac06a311131ebf04fc4
-  data.tar.gz: 88c8e2abf0cdd10f738b43561b0b2ee4bd471540e6120f75aaa0499fa7a04634b16a2f71215786667229cb04e47716a87d1e2d9f76098c25d74a32448e010885
+  metadata.gz: decaadbf4e7024b3fbaef0694c49695d6c77739b617a8d24a63ed32f736f2cd4090d84ae1c0cfea2d32b7ef6419575b2b17c74672c96e3c995c1cb978095da44
+  data.tar.gz: 17fff8235ae614cc166d867c8fb983b953cb3bc2bc6e72a1f056f52d453c09516201199034c894843d3cb691f1cc8a3b4f01688e98737b1f4598c9e76a188721

data/README.md

@@ -34,9 +34,9 @@ A fast XML parser and Object marshaller as a Ruby gem.
 
 ## <a name="release">Release Notes</a>
 
-### Release 2.0.2
+### Release 2.0.3
 
- - Fixed buffer sliding window off by 1 error in the SAX parser.
+ - Fixed excessive memory allocation issue for very large file parsing (half a gig).
 
 ## <a name="description">Description</a>
 

data/ext/ox/sax.c

@@ -188,8 +188,8 @@ ox_sax_drive_error_at(SaxDrive dr, const char *msg, int line, int col) {
         VALUE   args[3];
 
         args[0] = rb_str_new2(msg);
-        args[1] = INT2FIX(line);
-        args[2] = INT2FIX(col);
+        args[1] = LONG2NUM(line);
+        args[2] = LONG2NUM(col);
 	if (dr->has.line) {
 	    rb_ivar_set(dr->handler, ox_at_line_id, args[1]);
 	}
@@ -331,10 +331,10 @@ parse(SaxDrive dr) {
 	Nv	sp;
 
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(dr->buf.line));
+	    rb_ivar_set(dr->handler, ox_at_line_id, LONG2NUM(dr->buf.line));
 	}
 	if (dr->has.column) {
-	    rb_ivar_set(dr->handler, ox_at_column_id, INT2FIX(dr->buf.col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(dr->buf.col));
 	}
 	for (sp = dr->stack.tail - 1; dr->stack.head <= sp; sp--) {
 	    snprintf(msg, sizeof(msg) - 1, "%selement '%s' not closed", EL_MISMATCH, sp->name);
@@ -400,10 +400,10 @@ read_instruction(SaxDrive dr) {
         VALUE       args[1];
 
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
         args[0] = target;
         rb_funcall2(dr->handler, ox_instruct_id, 1, args);
@@ -437,10 +437,10 @@ read_instruction(SaxDrive dr) {
 	    }
 #endif
 	    if (dr->has.line) {
-		rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+		rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	    }
 	    rb_funcall2(dr->handler, ox_text_id, 1, args);
 	}
@@ -463,10 +463,10 @@ read_instruction(SaxDrive dr) {
         VALUE       args[1];
 
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
         args[0] = target;
         rb_funcall2(dr->handler, ox_end_instruct_id, 1, args);
@@ -504,10 +504,10 @@ read_doctype(SaxDrive dr) {
         VALUE       args[1];
 
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
         args[0] = rb_str_new2(dr->buf.str);
         rb_funcall2(dr->handler, ox_doctype_id, 1, args);
@@ -584,10 +584,10 @@ read_cdata(SaxDrive dr) {
         }
 #endif
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
         rb_funcall2(dr->handler, ox_cdata_id, 1, args);
     }
@@ -663,10 +663,10 @@ read_comment(SaxDrive dr) {
         }
 #endif
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
         rb_funcall2(dr->handler, ox_comment_id, 1, args);
     }
@@ -744,10 +744,10 @@ read_element_start(SaxDrive dr) {
         VALUE       args[1];
 
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
         args[0] = name;
         rb_funcall2(dr->handler, ox_start_element_id, 1, args);
@@ -839,10 +839,10 @@ read_element_end(SaxDrive dr) {
 		    VALUE       args[1];
 
 		    if (dr->has.line) {
-			rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+			rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 		    }
 		    args[0] = name;
 		    rb_funcall2(dr->handler, ox_start_element_id, 1, args);
@@ -858,10 +858,10 @@ read_element_end(SaxDrive dr) {
 		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);
 		if (dr->has.line) {
-		    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+		    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 		}
 		for (nv = stack_pop(&dr->stack); match < nv; nv = stack_pop(&dr->stack)) {
 		    if (dr->has.end_element) {
@@ -898,10 +898,10 @@ read_text(SaxDrive dr) {
         VALUE   args[1];
 
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
 	*args = dr->value_obj;
         rb_funcall2(dr->handler, ox_value_id, 1, args);
@@ -922,10 +922,10 @@ read_text(SaxDrive dr) {
         }
 #endif
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
         rb_funcall2(dr->handler, ox_text_id, 1, args);
     }
@@ -994,10 +994,10 @@ read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req)
             VALUE       args[2];
 
 	    if (dr->has.line) {
-		rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+		rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	    }
             args[0] = name;
             args[1] = dr->value_obj;
@@ -1018,10 +1018,10 @@ read_attrs(SaxDrive dr, char c, char termc, char term2, int is_xml, int eq_req)
 	    }
 #endif
 	    if (dr->has.line) {
-		rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+		rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	    }
             rb_funcall2(dr->handler, ox_attr_id, 2, args);
         }
@@ -1239,10 +1239,10 @@ static void
 end_element_cb(SaxDrive dr, VALUE name, int line, int col) {
     if (dr->has.end_element) {
 	if (dr->has.line) {
-	    rb_ivar_set(dr->handler, ox_at_line_id, INT2FIX(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, INT2FIX(col));
+	    rb_ivar_set(dr->handler, ox_at_column_id, LONG2NUM(col));
 	}
 	rb_funcall(dr->handler, ox_end_element_id, 1, name);
     }

data/ext/ox/sax_buf.c

@@ -114,7 +114,8 @@ ox_sax_buf_read(Buf buf) {
     int         err;
     size_t      shift = 0;
     
-    if (buf->head < buf->tail) {
+    // if there is not much room to read into, shift or realloc a larger buffer.
+    if (buf->head < buf->tail && 4096 > buf->end - buf->tail) {
         if (0 == buf->pro) {
             shift = buf->tail - buf->head;
         } else {
@@ -170,7 +171,7 @@ rescue_cb(VALUE rbuf, VALUE err) {
 #endif
 	Buf	buf = (Buf)rbuf;
 
-        ox_sax_drive_cleanup(buf->dr);
+        //ox_sax_drive_cleanup(buf->dr); called after exiting protect
         rb_raise(err, "at line %d, column %d\n", buf->line, buf->col);
     }
 #endif
@@ -189,7 +190,7 @@ partial_io_cb(VALUE rbuf) {
     rstr = rb_funcall2(buf->io, ox_readpartial_id, 1, args);
     str = StringValuePtr(rstr);
     cnt = strlen(str);
-    /*printf("*** read %lu bytes, str: '%s'\n", cnt, str); */
+    //printf("*** read %lu bytes, str: '%s'\n", cnt, str);
     strcpy(buf->tail, str);
     buf->read_end = buf->tail + cnt;
 

data/ext/ox/sax_buf.h

@@ -153,8 +153,9 @@ buf_next_white(Buf buf) {
 
 static inline void
 buf_cleanup(Buf buf) {
-    if (buf->base != buf->head) {
+    if (buf->base != buf->head && 0 != buf->head) {
         xfree(buf->head);
+	buf->head = 0;
     }
 }
 

data/lib/ox/version.rb

@@ -1,5 +1,5 @@
 
 module Ox
   # Current version of the module. 
-  VERSION = '2.0.2'
+  VERSION = '2.0.3'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: ox
 version: !ruby/object:Gem::Version
-  version: 2.0.2
+  version: 2.0.3
 platform: ruby
 authors:
 - Peter Ohler
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-06-07 00:00:00.000000000 Z
+date: 2013-06-12 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