More fine tuning of the regular expressions that detect URLs. We now allow '.' and ',' in URLs, as long as they are not at the very end of the URL, where they would probably be part of the enclosing sentence and not actually part of the URL.

git-svn-id: https://shellinabox.googlecode.com/svn/trunk@143 0da03de8-d603-11dd-86c2-0f8696b7b6f9

ChangeLog

@@ -3,6 +3,11 @@
 	* Optionally compress large responses, if the browser accepts
 	deflate compression. This mainly improves start up time.
 
+	* More fine tuning of the regular expressions that detect URLs.
+	We now allow '.' and ',' in URLs, as long as they are not at the
+	very end of the URL, where they would probably be part of the
+	enclosing sentence and not actually part of the URL.
+
 2009-07-06  Markus Gutschke  <markus@shellinabox.com>
 
 	* Making it easier to host the terminal on non-root URLs by always

config.h

@@ -135,7 +135,7 @@
 #define STDC_HEADERS 1
 
 /* Most recent revision number in the version control system */
-#define VCS_REVISION "142"
+#define VCS_REVISION "143"
 
 /* Version number of package */
 #define VERSION "2.9"

configure

@@ -2037,7 +2037,7 @@ ac_link='$CC -o conftest$ac_exeext $CFLAGS $CPPFLAGS $LDFLAGS conftest.$ac_ext $
 ac_compiler_gnu=$ac_cv_c_compiler_gnu
 
 
-VCS_REVISION=142
+VCS_REVISION=143
 
 
 cat >>confdefs.h <<_ACEOF

configure.ac

@@ -2,7 +2,7 @@ AC_PREREQ(2.57)
 
 dnl This is the one location where the authoritative version number is stored
 AC_INIT(shellinabox, 2.9, markus@shellinabox.com)
-VCS_REVISION=142
+VCS_REVISION=143
 AC_SUBST(VCS_REVISION)
 AC_DEFINE_UNQUOTED(VCS_REVISION, "${VCS_REVISION}",
                    [Most recent revision number in the version control system])

demo/vt100.js

@@ -104,7 +104,7 @@ function VT100(container) {
     '(?:http|https|ftp)://' +
 
     // Optionally allow username and passwords.
-    '(?:[^:@/ ]*(?::[^@/ ]*)?@)?' +
+    '(?:[^:@/ \u00A0]*(?::[^@/ \u00A0]*)?@)?' +
 
     // Hostname.
     '(?:[1-9][0-9]{0,2}(?:[.][1-9][0-9]{0,2}){3}|' +
@@ -115,12 +115,12 @@ function VT100(container) {
     '(?::[1-9][0-9]*)?' +
 
     // Path.
-    '(?:/[^/,.) ]*)*|' +
+    '(?:/(?:(?![/ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)*|' +
 
     (linkifyURLs <= 1 ? '' :
     // Also support URLs without a protocol (assume "http").
     // Optional username and password.
-    '(?:[^:@/ ]*(?::[^@/ ]*)?@)?' +
+    '(?:[^:@/ \u00A0]*(?::[^@/ \u00A0]*)?@)?' +
 
     // Hostnames must end with a well-known top-level domain or must be
     // numeric.
@@ -146,7 +146,7 @@ function VT100(container) {
     '(?::[1-9][0-9]{0,4})?' +
 
     // Path.
-    '(?:/[^/,.) ]*)*|') +
+    '(?:/(?:(?![/ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)*|') +
 
     // In addition, support e-mail address. Optionally, recognize "mailto:"
     '(?:mailto:)' + (linkifyURLs <= 1 ? '' : '?') +
@@ -171,7 +171,7 @@ function VT100(container) {
     'yu|za|zm|zw|arpa)(?![a-zA-Z0-9])|[Xx][Nn]--[-a-zA-Z0-9]+)' +
 
     // Optional arguments
-    '(?:[?][^/,.) ]+)?');
+    '(?:[?](?:(?![ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)?');
   }
   this.initializeElements(container);
   this.initializeAnsiColors();
@@ -980,7 +980,7 @@ VT100.prototype.truncateLines = function(width) {
     if (line.tagName == 'DIV') {
       var x           = 0;
 
-      // Traverse current line and truncate if once we saw "width" characters
+      // Traverse current line and truncate it once we saw "width" characters
       for (var span = line.firstChild; span;
            span = span.nextSibling) {
         var s         = this.getTextContent(span);
@@ -1000,7 +1000,7 @@ VT100.prototype.truncateLines = function(width) {
         // Scan backwards looking for first non-space character
         var s         = this.getTextContent(span);
         for (var i = s.length; i--; ) {
-          if (s.charAt(i) != ' ') {
+          if (s.charAt(i) != ' ' && s.charAt(i) != '\u00A0') {
             if (i+1 != s.length) {
               this.setTextContent(s.substr(0, i+1));
             }
@@ -1253,7 +1253,7 @@ VT100.prototype.putString = function(x, y, text, style) {
       // Scan backwards looking for first non-space character
       s                             = this.getTextContent(span);
       for (var i = s.length; i--; ) {
-        if (s.charAt(i) != ' ') {
+        if (s.charAt(i) != ' ' && s.charAt(i) != '\u00A0') {
           if (i+1 != s.length) {
             this.setTextContent(s.substr(0, i+1));
           }
@@ -1693,7 +1693,7 @@ VT100.prototype.toggleBell = function() {
 };
 
 VT100.prototype.about = function() {
-  alert("VT100 Terminal Emulator " + "2.9 (revision 142)" +
+  alert("VT100 Terminal Emulator " + "2.9 (revision 143)" +
         "\nCopyright 2008-2009 by Markus Gutschke\n" +
         "For more information check http://shellinabox.com");
 };

shellinabox/shell_in_a_box.js

@@ -355,7 +355,7 @@ ShellInABox.prototype.extendContextMenu = function(entries, actions) {
 };
 
 ShellInABox.prototype.about = function() {
-  alert("Shell In A Box version " + "2.9 (revision 142)" +
+  alert("Shell In A Box version " + "2.9 (revision 143)" +
         "\nCopyright 2008-2009 by Markus Gutschke\n" +
         "For more information check http://shellinabox.com" +
         (typeof serverSupportsSSL != 'undefined' && serverSupportsSSL ?

shellinabox/vt100.js

@@ -104,7 +104,7 @@ function VT100(container) {
     '(?:http|https|ftp)://' +
 
     // Optionally allow username and passwords.
-    '(?:[^:@/ ]*(?::[^@/ ]*)?@)?' +
+    '(?:[^:@/ \u00A0]*(?::[^@/ \u00A0]*)?@)?' +
 
     // Hostname.
     '(?:[1-9][0-9]{0,2}(?:[.][1-9][0-9]{0,2}){3}|' +
@@ -115,12 +115,12 @@ function VT100(container) {
     '(?::[1-9][0-9]*)?' +
 
     // Path.
-    '(?:/[^/,.) ]*)*|' +
+    '(?:/(?:(?![/ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)*|' +
 
     (linkifyURLs <= 1 ? '' :
     // Also support URLs without a protocol (assume "http").
     // Optional username and password.
-    '(?:[^:@/ ]*(?::[^@/ ]*)?@)?' +
+    '(?:[^:@/ \u00A0]*(?::[^@/ \u00A0]*)?@)?' +
 
     // Hostnames must end with a well-known top-level domain or must be
     // numeric.
@@ -146,7 +146,7 @@ function VT100(container) {
     '(?::[1-9][0-9]{0,4})?' +
 
     // Path.
-    '(?:/[^/,.) ]*)*|') +
+    '(?:/(?:(?![/ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)*|') +
 
     // In addition, support e-mail address. Optionally, recognize "mailto:"
     '(?:mailto:)' + (linkifyURLs <= 1 ? '' : '?') +
@@ -171,7 +171,7 @@ function VT100(container) {
     'yu|za|zm|zw|arpa)(?![a-zA-Z0-9])|[Xx][Nn]--[-a-zA-Z0-9]+)' +
 
     // Optional arguments
-    '(?:[?][^/,.) ]+)?');
+    '(?:[?](?:(?![ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)?');
   }
   this.initializeElements(container);
   this.initializeAnsiColors();
@@ -980,7 +980,7 @@ VT100.prototype.truncateLines = function(width) {
     if (line.tagName == 'DIV') {
       var x           = 0;
 
-      // Traverse current line and truncate if once we saw "width" characters
+      // Traverse current line and truncate it once we saw "width" characters
       for (var span = line.firstChild; span;
            span = span.nextSibling) {
         var s         = this.getTextContent(span);
@@ -1000,7 +1000,7 @@ VT100.prototype.truncateLines = function(width) {
         // Scan backwards looking for first non-space character
         var s         = this.getTextContent(span);
         for (var i = s.length; i--; ) {
-          if (s.charAt(i) != ' ') {
+          if (s.charAt(i) != ' ' && s.charAt(i) != '\u00A0') {
             if (i+1 != s.length) {
               this.setTextContent(s.substr(0, i+1));
             }
@@ -1253,7 +1253,7 @@ VT100.prototype.putString = function(x, y, text, style) {
       // Scan backwards looking for first non-space character
       s                             = this.getTextContent(span);
       for (var i = s.length; i--; ) {
-        if (s.charAt(i) != ' ') {
+        if (s.charAt(i) != ' ' && s.charAt(i) != '\u00A0') {
           if (i+1 != s.length) {
             this.setTextContent(s.substr(0, i+1));
           }
@@ -1693,7 +1693,7 @@ VT100.prototype.toggleBell = function() {
 };
 
 VT100.prototype.about = function() {
-  alert("VT100 Terminal Emulator " + "2.9 (revision 142)" +
+  alert("VT100 Terminal Emulator " + "2.9 (revision 143)" +
         "\nCopyright 2008-2009 by Markus Gutschke\n" +
         "For more information check http://shellinabox.com");
 };

shellinabox/vt100.jspp

@@ -104,7 +104,7 @@ function VT100(container) {
     '(?:http|https|ftp)://' +
 
     // Optionally allow username and passwords.
-    '(?:[^:@/ ]*(?::[^@/ ]*)?@)?' +
+    '(?:[^:@/ \u00A0]*(?::[^@/ \u00A0]*)?@)?' +
 
     // Hostname.
     '(?:[1-9][0-9]{0,2}(?:[.][1-9][0-9]{0,2}){3}|' +
@@ -115,12 +115,12 @@ function VT100(container) {
     '(?::[1-9][0-9]*)?' +
 
     // Path.
-    '(?:/[^/,.) ]*)*|' +
+    '(?:/(?:(?![/ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)*|' +
 
     (linkifyURLs <= 1 ? '' :
     // Also support URLs without a protocol (assume "http").
     // Optional username and password.
-    '(?:[^:@/ ]*(?::[^@/ ]*)?@)?' +
+    '(?:[^:@/ \u00A0]*(?::[^@/ \u00A0]*)?@)?' +
 
     // Hostnames must end with a well-known top-level domain or must be
     // numeric.
@@ -146,7 +146,7 @@ function VT100(container) {
     '(?::[1-9][0-9]{0,4})?' +
 
     // Path.
-    '(?:/[^/,.) ]*)*|') +
+    '(?:/(?:(?![/ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)*|') +
 
     // In addition, support e-mail address. Optionally, recognize "mailto:"
     '(?:mailto:)' + (linkifyURLs <= 1 ? '' : '?') +
@@ -171,7 +171,7 @@ function VT100(container) {
     'yu|za|zm|zw|arpa)(?![a-zA-Z0-9])|[Xx][Nn]--[-a-zA-Z0-9]+)' +
 
     // Optional arguments
-    '(?:[?][^/,.) ]+)?');
+    '(?:[?](?:(?![ \u00A0]|[,.)][ \u00A0]|[,.)]$).)*)?');
   }
   this.initializeElements(container);
   this.initializeAnsiColors();
@@ -980,7 +980,7 @@ VT100.prototype.truncateLines = function(width) {
     if (line.tagName == 'DIV') {
       var x           = 0;
 
-      // Traverse current line and truncate if once we saw "width" characters
+      // Traverse current line and truncate it once we saw "width" characters
       for (var span = line.firstChild; span;
            span = span.nextSibling) {
         var s         = this.getTextContent(span);
@@ -1000,7 +1000,7 @@ VT100.prototype.truncateLines = function(width) {
         // Scan backwards looking for first non-space character
         var s         = this.getTextContent(span);
         for (var i = s.length; i--; ) {
-          if (s.charAt(i) != ' ') {
+          if (s.charAt(i) != ' ' && s.charAt(i) != '\u00A0') {
             if (i+1 != s.length) {
               this.setTextContent(s.substr(0, i+1));
             }
@@ -1253,7 +1253,7 @@ VT100.prototype.putString = function(x, y, text, style) {
       // Scan backwards looking for first non-space character
       s                             = this.getTextContent(span);
       for (var i = s.length; i--; ) {
-        if (s.charAt(i) != ' ') {
+        if (s.charAt(i) != ' ' && s.charAt(i) != '\u00A0') {
           if (i+1 != s.length) {
             this.setTextContent(s.substr(0, i+1));
           }