archive-access-cvs

[Archive-access-cvs] SF.net SVN: archive-access: [2434] branches/wayback-1_2_2/wayback/ wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java

[<bra...@us...>]

Revision: 2434
          http://archive-access.svn.sourceforge.net/archive-access/?rev=2434&view=rev
Author:   bradtofel
Date:     2008-07-13 18:40:33 -0700 (Sun, 13 Jul 2008)

Log Message:
-----------
BUGFIX (ACC-21): now rewrites all(or at least more) url(XXX) tags in CSS.

Modified Paths:
--------------
    branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java

Modified: branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java
===================================================================
--- branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java	2008-07-14 01:40:16 UTC (rev 2433)
+++ branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java	2008-07-14 01:40:33 UTC (rev 2434)
@@ -76,14 +76,14 @@
 	private static String cssUrlPatString = 
 		"url\\s*\\(\\s*([\\\\\"']*.+?[\\\\\"']*)\\s*\\)";
 
-	private static String cssImportPatString = 
-		"@import\\s+" + cssUrlPatString;
+//	private static String cssImportPatString = 
+//		"@import\\s+" + cssUrlPatString;
 
 	private static String cssImportNoUrlPatString =
 		"@import\\s+([\"'].+?[\"'])";
 	
-	private static Pattern cssImportPattern = 
-		Pattern.compile(cssImportPatString);
+//	private static Pattern cssImportPattern = 
+//		Pattern.compile(cssImportPatString);
 
 	private static Pattern cssImportNoUrlPattern = 
 		Pattern.compile(cssImportNoUrlPatString);
@@ -161,8 +161,9 @@
 	public static void markupCSSImports(StringBuilder page,
 			ResultURIConverter uriConverter, String captureDate,
 			String baseUrl) {
-		markupTagREURIC(page,uriConverter,captureDate,baseUrl,cssImportPattern);
+//		markupTagREURIC(page,uriConverter,captureDate,baseUrl,cssImportPattern);
 		markupTagREURIC(page,uriConverter,captureDate,baseUrl,cssImportNoUrlPattern);
+		markupTagREURIC(page,uriConverter,captureDate,baseUrl,cssUrlPattern);
 	}
 	
 	public static void markupStyleUrls(StringBuilder page,


This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

[Archive-access-cvs] SF.net SVN: archive-access: [2442] branches/wayback-1_2_2/wayback/ wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java

[<bra...@us...>]

Revision: 2442
          http://archive-access.svn.sourceforge.net/archive-access/?rev=2442&view=rev
Author:   bradtofel
Date:     2008-07-14 16:19:40 -0700 (Mon, 14 Jul 2008)

Log Message:
-----------
BUGFIX(ACC-26): string OOB exception in javascript generated escaped HTML attributes.

Modified Paths:
--------------
    branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java

Modified: branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java
===================================================================
--- branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java	2008-07-14 21:24:37 UTC (rev 2441)
+++ branches/wayback-1_2_2/wayback/wayback-core/src/main/java/org/archive/wayback/replay/TagMagix.java	2008-07-14 23:19:40 UTC (rev 2442)
@@ -42,9 +42,9 @@
  */
 public class TagMagix {
 
-	// minimum length XXXX in a 'style=XXXX' declaration... mostly handy
-	// to keep us from trying to mark up javascript generated style code.
-	private static int MIN_STYLE_LENGTH = 3;
+	// minimum length XXXX in a 'ATTR=XXXX' declaration... mostly handy
+	// to keep us from trying to mark up javascript generated HTML/CSS code.
+	private static int MIN_ATTR_LENGTH = 3;
 	
 	private static HashMap<String, Pattern> pcPatterns = 
 		new HashMap<String, Pattern>();
@@ -179,7 +179,7 @@
 			int attrStart = matcher.start(1);
 			int attrEnd = matcher.end(1);
 			idx = attrEnd;
-			if(origAttrLength < MIN_STYLE_LENGTH) {
+			if(origAttrLength < MIN_ATTR_LENGTH) {
 				continue;
 			}
 
@@ -227,10 +227,12 @@
 	 * Alter the HTML document in page, updating URLs in the attrName attributes
 	 * of all tagName tags such that:
 	 * 
-	 * 1) absolute URLs are prefixed with: wmPrefix + pageTS 2) server-relative
-	 * URLs are prefixed with: wmPrefix + pageTS + (host of page) 3)
-	 * path-relative URLs are prefixed with: wmPrefix + pageTS + (attribute URL
-	 * resolved against pageUrl)
+	 * 1) absolute URLs are prefixed with: 
+	 * 				wmPrefix + pageTS
+	 * 2) server-relative URLs are prefixed with: 
+	 * 				wmPrefix + pageTS + (host of page)
+	 * 3) path-relative URLs are prefixed with: 
+	 * 				wmPrefix + pageTS + (attribute URL resolved against baseUrl)
 	 * 
 	 * @param page
 	 * @param uriConverter
@@ -258,6 +260,10 @@
 			int origUrlLength = url.length();
 			int attrStart = matcher.start(1);
 			int attrEnd = matcher.end(1);
+			if(origUrlLength < MIN_ATTR_LENGTH) {
+				idx = attrEnd;
+				continue;
+			}
 			String quote = "";
 			if (url.charAt(0) == '"') {
 				quote = "\"";


This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.