Mailing-List: contact dev-help@ant.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Ant Developers List" <dev@ant.apache.org>
Received-SPF: pass (nike.apache.org: local policy)
From: "aborisevich" <aborisevich@beldts.de>
To: <dev@ant.apache.org>
Subject: Including national characters int Tar Entry names into archive.
Date: Tue, 19 May 2009 18:31:57 +0300
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_00DA_01C9D8B0.17208260"
Message-Id: <20090519153703.77F1B4FC747@mx1.dts-online.net>

------=_NextPart_000_00DA_01C9D8B0.17208260
Content-Type: text/plain;
	charset="koi8-r"
Content-Transfer-Encoding: quoted-printable

Hello everyone. I have found the next bug using present  =
org.apache.tools.tar package. Tar Archive was created on one system (for =
example Windows XP  - default charset CP-1251). This tar archive =
contains TarEntries were named with using national characters like =
German umlauts. Than this archive file was copied on Linux system =
(default charset UTF-8)  - after unpackin this archive file there - =
information was lost (TarEntries names were lost). There is possible =
solution for this problem.


Index: ant-core/src/main/org/apache/tools/tar/TarInputStream.java
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
--- ant-core/src/main/org/apache/tools/tar/TarInputStream.java (revision =
776302)
+++ ant-core/src/main/org/apache/tools/tar/TarInputStream.java (working =
copy)
@@ -264,10 +264,10 @@
         if (currEntry !=3D null && currEntry.isGNULongNameEntry()) {
             // read in the name
             StringBuffer longName =3D new StringBuffer();
-            byte[] buf =3D new byte[SMALL_BUFFER_SIZE];
+            byte[] buf =3D new byte[(int)currEntry.getSize()];
             int length =3D 0;
             while ((length =3D read(buf)) >=3D 0) {
-                longName.append(new String(buf, 0, length));
+                longName.append(new String(buf, 0, length, "UTF-8"));
             }
             getNextEntry();
             if (currEntry =3D=3D null) {
Index: ant-core/src/main/org/apache/tools/tar/TarOutputStream.java
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
--- ant-core/src/main/org/apache/tools/tar/TarOutputStream.java =
(revision 776302)
+++ ant-core/src/main/org/apache/tools/tar/TarOutputStream.java (working =
copy)
@@ -179,9 +179,10 @@
                 TarEntry longLinkEntry =3D new =
TarEntry(TarConstants.GNU_LONGLINK,
                                                       =
TarConstants.LF_GNUTYPE_LONGNAME);
=20
-                longLinkEntry.setSize(entry.getName().length() + 1);
+                byte[] nameBytes =3D entry.getName().getBytes("UTF-8");
+                longLinkEntry.setSize(nameBytes.length + 1);
                 putNextEntry(longLinkEntry);
-                write(entry.getName().getBytes());
+               write(nameBytes);
                 write(0);
                 closeEntry();
             } else if (longFileMode !=3D LONGFILE_TRUNCATE) {
Index: ant-core/src/main/org/apache/tools/tar/TarUtils.java
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
--- ant-core/src/main/org/apache/tools/tar/TarUtils.java (revision =
776302)
+++ ant-core/src/main/org/apache/tools/tar/TarUtils.java (working copy)
@@ -23,6 +23,8 @@
=20
 package org.apache.tools.tar;
=20
+import java.io.UnsupportedEncodingException;
+
 /**
  * This class provides static utility methods to work with byte =
streams.
  *
@@ -79,15 +81,21 @@
      * @return The header's entry name.
      */
     public static StringBuffer parseName(byte[] header, int offset, int =
length) {
-        StringBuffer result =3D new StringBuffer(length);
+        StringBuffer result =3D null;
+        int nameLen =3D length;
+
         int          end =3D offset + length;
-
         for (int i =3D offset; i < end; ++i) {
-            if (header[i] =3D=3D 0) {
+            if(header[i] =3D=3D 0) {
+                nameLen =3D i - offset;
                 break;
             }
+        }
=20
-            result.append((char) header[i]);
+        try {
+            result =3D new StringBuffer(new String(header, offset, =
nameLen, "UTF-8"));
+        } catch(UnsupportedEncodingException e) {
+           e.printStackTrace();
         }
=20
         return result;
@@ -103,18 +111,23 @@
      * @return The number of bytes in a header's entry name.
      */
     public static int getNameBytes(StringBuffer name, byte[] buf, int =
offset, int length) {
-        int i;
+         int nameLength =3D -1;
+         try
+         {
+             byte nameBytes[] =3D name.toString().getBytes("UTF-8");
+             nameLength =3D nameBytes.length ;
+             System.arraycopy(nameBytes, 0, buf, offset, nameLength);
+         } catch(UnsupportedEncodingException e) {
+             e.printStackTrace();
+         }
=20
-        for (i =3D 0; i < length && i < name.length(); ++i) {
-            buf[offset + i] =3D (byte) name.charAt(i);
-        }
=20
-        for (; i < length; ++i) {
-            buf[offset + i] =3D 0;
-        }
+         for (; nameLength < length; ++nameLength) {
+             buf[offset + nameLength] =3D 0;
+         }
=20
-        return offset + length;
-    }
+         return offset + length;
+     }
=20
     /**
      * Parse an octal integer from a header buffer.


Best Regards.
Alexander Borisevich =20
BelDTS Minsk Belarus
------=_NextPart_000_00DA_01C9D8B0.17208260--