Author: rjung
Date: Sun Nov 13 16:14:02 2011
New Revision: 1201452
URL: http://svn.apache.org/viewvc?rev=1201452&view=rev
Log:
Clean up handling multi-byte chars in the connectors.
Backport of r1201069+1201087 from trunk
resp. r1201076+r1201088 from TC7
plus making the JK connector consistent with the
newer ones.
Modified:
tomcat/tc6.0.x/trunk/STATUS.txt
tomcat/tc6.0.x/trunk/java/org/apache/coyote/ajp/AjpMessage.java
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalAprOutputBuffer.java
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalNioOutputBuffer.java
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalOutputBuffer.java
tomcat/tc6.0.x/trunk/java/org/apache/jk/common/JkInputStream.java
tomcat/tc6.0.x/trunk/java/org/apache/jk/common/MsgAjp.java
tomcat/tc6.0.x/trunk/webapps/docs/changelog.xml
Modified: tomcat/tc6.0.x/trunk/STATUS.txt
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/STATUS.txt?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
--- tomcat/tc6.0.x/trunk/STATUS.txt (original)
+++ tomcat/tc6.0.x/trunk/STATUS.txt Sun Nov 13 16:14:02 2011
@@ -95,13 +95,6 @@ PATCHES PROPOSED TO BACKPORT:
-1:
-* Clean up handling multi-byte chars in the connectors.
- Backport of r1201069+1201087 from trunk resp. r1201076+r1201088 from TC7
- plus making the JK connector consistent with the Coyote one.
-
http://people.apache.org/~rjung/patches/connectors_multi-byte_handling_cleanup-v2.patch
- +1: rjung, kkolinko, jfclere
- -1:
-
PATCHES/ISSUES THAT ARE STALLED
* Backport JSP unloading patch (BZ48358).
Modified: tomcat/tc6.0.x/trunk/java/org/apache/coyote/ajp/AjpMessage.java
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/java/org/apache/coyote/ajp/AjpMessage.java?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
--- tomcat/tc6.0.x/trunk/java/org/apache/coyote/ajp/AjpMessage.java (original)
+++ tomcat/tc6.0.x/trunk/java/org/apache/coyote/ajp/AjpMessage.java Sun Nov 13
16:14:02 2011
@@ -218,9 +218,7 @@ public class AjpMessage {
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
appendByte(c);
@@ -253,9 +251,7 @@ public class AjpMessage {
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
appendByte(c);
Modified:
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalAprOutputBuffer.java
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalAprOutputBuffer.java?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
---
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalAprOutputBuffer.java
(original)
+++
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalAprOutputBuffer.java
Sun Nov 13 16:14:02 2011
@@ -634,9 +634,7 @@ public class InternalAprOutputBuffer
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
buf[pos++] = (byte) c;
@@ -681,9 +679,7 @@ public class InternalAprOutputBuffer
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
buf[pos++] = (byte) c;
Modified:
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalNioOutputBuffer.java
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalNioOutputBuffer.java?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
---
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalNioOutputBuffer.java
(original)
+++
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalNioOutputBuffer.java
Sun Nov 13 16:14:02 2011
@@ -713,9 +713,7 @@ public class InternalNioOutputBuffer
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
buf[pos++] = (byte) c;
@@ -760,9 +758,7 @@ public class InternalNioOutputBuffer
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
buf[pos++] = (byte) c;
Modified:
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalOutputBuffer.java
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalOutputBuffer.java?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
---
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalOutputBuffer.java
(original)
+++
tomcat/tc6.0.x/trunk/java/org/apache/coyote/http11/InternalOutputBuffer.java
Sun Nov 13 16:14:02 2011
@@ -679,9 +679,7 @@ public class InternalOutputBuffer
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
buf[pos++] = (byte) c;
@@ -726,9 +724,7 @@ public class InternalOutputBuffer
// but is the only consistent approach within the current
// servlet framework. It must suffice until servlet output
// streams properly encode their output.
- if ((c <= 31) && (c != 9)) {
- c = ' ';
- } else if (c == 127) {
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
c = ' ';
}
buf[pos++] = (byte) c;
Modified: tomcat/tc6.0.x/trunk/java/org/apache/jk/common/JkInputStream.java
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/java/org/apache/jk/common/JkInputStream.java?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
--- tomcat/tc6.0.x/trunk/java/org/apache/jk/common/JkInputStream.java (original)
+++ tomcat/tc6.0.x/trunk/java/org/apache/jk/common/JkInputStream.java Sun Nov
13 16:14:02 2011
@@ -316,11 +316,9 @@ public class JkInputStream implements In
MessageBytes hN=headers.getName(i);
// no header to sc conversion - there's little benefit
// on this direction
- c2b.convert ( hN );
outputMsg.appendBytes( hN );
MessageBytes hV=headers.getValue(i);
- c2b.convert( hV );
outputMsg.appendBytes( hV );
}
mc.getSource().send( outputMsg, mc );
Modified: tomcat/tc6.0.x/trunk/java/org/apache/jk/common/MsgAjp.java
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/java/org/apache/jk/common/MsgAjp.java?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
--- tomcat/tc6.0.x/trunk/java/org/apache/jk/common/MsgAjp.java (original)
+++ tomcat/tc6.0.x/trunk/java/org/apache/jk/common/MsgAjp.java Sun Nov 13
16:14:02 2011
@@ -21,6 +21,7 @@ import java.io.IOException;
import org.apache.jk.core.Msg;
import org.apache.tomcat.util.buf.ByteChunk;
+import org.apache.tomcat.util.buf.CharChunk;
import org.apache.tomcat.util.buf.MessageBytes;
/**
@@ -149,10 +150,15 @@ public class MsgAjp extends Msg {
appendByte(0);
return;
}
-
- // XXX Convert !!
- ByteChunk bc= mb.getByteChunk();
- appendByteChunk(bc);
+ if (mb.getType() == MessageBytes.T_BYTES) {
+ ByteChunk bc = mb.getByteChunk();
+ appendByteChunk(bc);
+ } else if (mb.getType() == MessageBytes.T_CHARS) {
+ CharChunk cc = mb.getCharChunk();
+ appendCharChunk(cc);
+ } else {
+ appendString(mb.toString());
+ }
}
public void appendByteChunk(ByteChunk bc) throws IOException {
@@ -171,6 +177,66 @@ public class MsgAjp extends Msg {
appendByte(0);
}
+ /**
+ * Write a CharChunk out at the current write position.
+ * A null CharChunk is encoded as a string with length 0.
+ */
+ private void appendCharChunk(CharChunk cc) {
+ if (cc == null) {
+ log.error("appendCharChunk() null");
+ appendInt(0);
+ appendByte(0);
+ return;
+ }
+ int start = cc.getStart();
+ int end = cc.getEnd();
+ appendInt(end - start);
+ char[] cbuf = cc.getBuffer();
+ for (int i = start; i < end; i++) {
+ char c = cbuf[i];
+ // Note: This is clearly incorrect for many strings,
+ // but is the only consistent approach within the current
+ // servlet framework. It must suffice until servlet output
+ // streams properly encode their output.
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
+ c = ' ';
+ }
+ appendByte((byte)c);
+ }
+ appendByte(0);
+ }
+
+ /**
+ * Write a String out at the current write position. Strings are
+ * encoded with the length in two bytes first, then the string, and
+ * then a terminating \0 (which is <B>not</B> included in the
+ * encoded length). The terminator is for the convenience of the C
+ * code, where it saves a round of copying. A null string is
+ * encoded as a string with length 0.
+ */
+ private void appendString(String str) {
+ if (str == null) {
+ log.error("appendString() null");
+ appendInt(0);
+ appendByte(0);
+ return;
+ }
+ int len = str.length();
+ appendInt(len);
+ for (int i = 0; i < len; i++) {
+ char c = str.charAt (i);
+ // Note: This is clearly incorrect for many strings,
+ // but is the only consistent approach within the current
+ // servlet framework. It must suffice until servlet output
+ // streams properly encode their output.
+ if (((c <= 31) && (c != 9)) || c == 127 || c > 255) {
+ c = ' ';
+ }
+ appendByte((byte)c);
+ }
+ appendByte(0);
+ }
+
/**
* Copy a chunk of bytes into the packet, starting at the current
* write position. The chunk of bytes is encoded with the length
Modified: tomcat/tc6.0.x/trunk/webapps/docs/changelog.xml
URL:
http://svn.apache.org/viewvc/tomcat/tc6.0.x/trunk/webapps/docs/changelog.xml?rev=1201452&r1=1201451&r2=1201452&view=diff
==============================================================================
--- tomcat/tc6.0.x/trunk/webapps/docs/changelog.xml (original)
+++ tomcat/tc6.0.x/trunk/webapps/docs/changelog.xml Sun Nov 13 16:14:02 2011
@@ -175,6 +175,9 @@
Allow the BIO HTTP connector to be used with SSL when running under
Java
7. (markt)
</fix>
+ <fix>
+ Improve multi-byte character handling in all connectors. (rjung)
+ </fix>
</changelog>
</subsection>
<subsection name="Jasper">
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
