Re: svn commit: r1503857 - in /tomcat/trunk: java/javax/el/ java/org/apache/el/parser/ test/org/apache/el/parser/
2013/7/16
>
> Author: markt
> Date: Tue Jul 16 20:01:47 2013
> New Revision: 1503857
>
> URL: http://svn.apache.org/r1503857
> Log:
> Fix handling of nested lambda expressions with method parameters.
>
> Removed:
>
tomcat/trunk/java/org/apache/el/parser/AstLambdaExpressionOrInvocation.java
> Modified:
> tomcat/trunk/java/javax/el/LambdaExpression.java
> tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
> tomcat/trunk/java/org/apache/el/parser/ELParser.java
> tomcat/trunk/java/org/apache/el/parser/ELParser.jjt
> tomcat/trunk/java/org/apache/el/parser/ELParserTreeConstants.java
> tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java
>
> Modified: tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
> URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java?rev=1503857&r1=1503856&r2=1503857&view=diff
>
==
> --- tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
(original)
> +++ tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java Tue
Jul 16 20:01:47 2013
> @@ -35,11 +35,15 @@ public class AstLambdaExpression extends
> @Override
> public Object getValue(EvaluationContext ctx) throws ELException {
>
> -// Two children - the formal parameters and the expression
> +// First child is always parameters even if there aren't any
> AstLambdaParameters formalParametersNode =
> (AstLambdaParameters) children[0];
> Node[] formalParamNodes = formalParametersNode.children;
>
> +// Second child is a value expression
> +ValueExpressionImpl ve = new ValueExpressionImpl("", children[1],
> +ctx.getFunctionMapper(), ctx.getVariableMapper(), null);
> +
> // Build a LambdaExpression
> List formalParameters = new ArrayList<>();
> if (formalParamNodes != null) {
> @@ -47,30 +51,29 @@ public class AstLambdaExpression extends
> formalParameters.add(formalParamNode.getImage());
> }
> }
> -
> -ValueExpressionImpl ve = new ValueExpressionImpl("", children[1],
> -ctx.getFunctionMapper(), ctx.getVariableMapper(), null);
> LambdaExpression le = new LambdaExpression(formalParameters, ve);
> le.setELContext(ctx);
>
> if (formalParameters.isEmpty()) {
> -// No formal parameters - should be able to simply invoke
this
> +// No formal parameters - invoke the expression
> return le.invoke(ctx, (Object[]) null);
> -} else {
> -// Has parameters but they aren't provided so return the
> -// LambdaExpression
> -return le;
> }
> -}
>
> -@Override
> -public Object invoke(EvaluationContext ctx, Class[] paramTypes,
> -Object[] paramValues) throws ELException {
> -
> -Object result = getValue(ctx);
> +// If there are method parameters, need to invoke the expression
with
> +// those parameters. If there are multiple method parameters
there
> +// should be that many nested expressions.
> +// If there are more nested expressions that parameters this
will return
> +// a LambdaExpression
> +Object result = le;
> +int i = 2;
> +while (result instanceof LambdaExpression && i <
jjtGetNumChildren()) {
> +result = ((LambdaExpression) result).invoke(
> +((AstMethodParameters)
children[i]).getParameters(ctx));
> +i++;
> +}
>
> -if (result instanceof LambdaExpression) {
> -result = ((LambdaExpression) result).invoke(ctx,
paramValues);
> +if (i < jjtGetNumChildren()) {
> +throw new ELException();
> }
>
> return result;
Having the example from the test:
(x->y->x-y)(2)(1)
We may transform it like this:
(()->y->2-y)()(1)
So instead of returning right way when formalParameters.isEmpty()
Shouldn't we have something like the one below in order to continue
evaluation of the nested expression:
--- org/apache/el/parser/AstLambdaExpression.java (revision 1504024)
+++ org/apache/el/parser/AstLambdaExpression.java (working copy)
@@ -54,9 +54,12 @@
LambdaExpression le = new LambdaExpression(formalParameters, ve);
le.setELContext(ctx);
+Object result = le;
+int i = 2;
if (formalParameters.isEmpty()) {
// No formal parameters - invoke the expression
-return le.invoke(ctx, (Object[]) null);
+result = ((LambdaExpression) result).invoke(ctx, (Object[])
null);
+i++;
}
// If there are method parameters, need to invoke the expression
with
@@ -64,8 +67,6 @@
// should be that many nested expressions.
// If there are more nested expressions that parameters this
Re: svn commit: r1503857 - in /tomcat/trunk: java/javax/el/ java/org/apache/el/parser/ test/org/apache/el/parser/
Violeta Georgieva wrote:
>Having the example from the test:
>(x->y->x-y)(2)(1)
>
>We may transform it like this:
>(()->y->2-y)()(1)
>
>So instead of returning right way when formalParameters.isEmpty()
>
>Shouldn't we have something like the one below in order to continue
>evaluation of the nested expression:
>
>
>
>--- org/apache/el/parser/AstLambdaExpression.java (revision 1504024)
>+++ org/apache/el/parser/AstLambdaExpression.java (working copy)
>@@ -54,9 +54,12 @@
> LambdaExpression le = new LambdaExpression(formalParameters, ve);
> le.setELContext(ctx);
>
>+Object result = le;
>+int i = 2;
> if (formalParameters.isEmpty()) {
> // No formal parameters - invoke the expression
>-return le.invoke(ctx, (Object[]) null);
>+result = ((LambdaExpression) result).invoke(ctx,
>(Object[])
>null);
>+i++;
> }
>
> // If there are method parameters, need to invoke the expression
>with
>@@ -64,8 +67,6 @@
> // should be that many nested expressions.
> // If there are more nested expressions that parameters this will
>return
> // a LambdaExpression
>-Object result = le;
>-int i = 2;
> while (result instanceof LambdaExpression && i <
>jjtGetNumChildren()) {
> result = ((LambdaExpression) result).invoke(
> ((AstMethodParameters)
>children[i]).getParameters(ctx));
I agree the current solution isn't right but I have some local test cases that
I'm fairly sure will still fail with the above patch. I'm working on a fix and
should have something to commit soon.
Mark
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: Various shell-scripting idioms in bin/daemon.sh
On 07/16/2013 11:42 PM, Christopher Schultz wrote: All, While doing the trivial fix for https://issues.apache.org/bugzilla/show_bug.cgi?id=55268, I noticed a few idioms being used in bin/daemon.sh that struck me as odd. For example: while [ ".$1" != . ] do case "$1" in --java-home ) JAVA_HOME="$2" shift; shift; continue ;; This example actually illustrates the two main questions I had: 1. Why use [ ".$FOO" != . ] instead of simply [ -n "$FOO" ] (Corollary: why use [ ".$FOO" = . ] instead of [ -z "$FOO" ])? Because some shell scripts does dot handle -z or -n well. 2. Why have a "continue" at the end of every case option, since the whole body of the while loop is nothing but the case construct? That might be an extra directive, true. Probably a leftover from multiple case directives in while loop. I may be spoiled by using Linux and bash for most of my career, but I believe these are fairly standard POSIX-compliant things that should work on all *NIX systems. Sadly that's not the case IMHO. Regards -- ^TM - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504123 - in /tomcat/trunk: java/org/apache/el/Messages.properties java/org/apache/el/parser/AstLambdaExpression.java test/org/apache/el/parser/TestAstLambdaExpression.java
Author: markt
Date: Wed Jul 17 13:24:09 2013
New Revision: 1504123
URL: http://svn.apache.org/r1504123
Log:
Further improve handling of nested lambda expressions.
Modified:
tomcat/trunk/java/org/apache/el/Messages.properties
tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java
Modified: tomcat/trunk/java/org/apache/el/Messages.properties
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/Messages.properties?rev=1504123&r1=1504122&r2=1504123&view=diff
==
--- tomcat/trunk/java/org/apache/el/Messages.properties (original)
+++ tomcat/trunk/java/org/apache/el/Messages.properties Wed Jul 17 13:24:09 2013
@@ -52,3 +52,4 @@ error.context.null=ELContext was null
# Parser
error.identifier.notjava=The identifier [{0}] is not a valid Java identifier
as required by section 1.19 of the EL specification (Identifier ::= Java
language identifier). This check can be disabled by setting the system property
org.apache.el.parser.SKIP_IDENTIFIER_CHECK to true.
+error.lambda.tooManyMethodParameterSets=There are more sets of method
parameters specified than there are nested lambda expressions
Modified: tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java?rev=1504123&r1=1504122&r2=1504123&view=diff
==
--- tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java (original)
+++ tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java Wed Jul 17
13:24:09 2013
@@ -25,6 +25,7 @@ import javax.el.LambdaExpression;
import org.apache.el.ValueExpressionImpl;
import org.apache.el.lang.EvaluationContext;
+import org.apache.el.util.MessageFactory;
public class AstLambdaExpression extends SimpleNode {
@@ -35,6 +36,24 @@ public class AstLambdaExpression extends
@Override
public Object getValue(EvaluationContext ctx) throws ELException {
+// Check that there are not more sets of method parameters than there
+// are nested lambda expressions
+int methodParameterSetCount = jjtGetNumChildren() - 2;
+if (methodParameterSetCount > 0) {
+// We know this node is an expression
+methodParameterSetCount--;
+Node n = this.jjtGetChild(1);
+while (methodParameterSetCount > 0) {
+if (n.jjtGetNumChildren() <2 ||
+!(n.jjtGetChild(0) instanceof AstLambdaParameters)) {
+throw new ELException(MessageFactory.get(
+"error.lambda.tooManyMethodParameterSets"));
+}
+n = n.jjtGetChild(1);
+methodParameterSetCount--;
+}
+}
+
// First child is always parameters even if there aren't any
AstLambdaParameters formalParametersNode =
(AstLambdaParameters) children[0];
@@ -54,16 +73,22 @@ public class AstLambdaExpression extends
LambdaExpression le = new LambdaExpression(formalParameters, ve);
le.setELContext(ctx);
-if (formalParameters.isEmpty()) {
+if (formalParameters.isEmpty() && jjtGetNumChildren() == 2) {
// No formal parameters - invoke the expression
return le.invoke(ctx, (Object[]) null);
}
// If there are method parameters, need to invoke the expression with
-// those parameters. If there are multiple method parameters there
-// should be that many nested expressions.
-// If there are more nested expressions that parameters this will
return
-// a LambdaExpression
+// those parameters. If there are multiple sets of method parameters
+// there should be at least that many nested expressions.
+// If there are more nested expressions than sets of method parameters
+// this may return a LambdaExpression.
+// If there are more sets of method parameters than nested expressions
+// an ELException will have been thrown by the check at the start of
+// this method.
+// If the inner most expression(s) do not require parameters then a
+// value will be returned once the outermost expression that does
+// require a parameter has been evaluated.
Object result = le;
int i = 2;
while (result instanceof LambdaExpression && i < jjtGetNumChildren()) {
@@ -72,10 +97,6 @@ public class AstLambdaExpression extends
i++;
}
-if (i < jjtGetNumChildren()) {
-throw new ELException();
-}
-
return result;
}
}
Modified: tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/test/org
svn commit: r1504124 - /tomcat/trunk/webapps/docs/config/systemprops.xml
Author: markt Date: Wed Jul 17 13:24:30 2013 New Revision: 1504124 URL: http://svn.apache.org/r1504124 Log: Whitespace police Modified: tomcat/trunk/webapps/docs/config/systemprops.xml Modified: tomcat/trunk/webapps/docs/config/systemprops.xml URL: http://svn.apache.org/viewvc/tomcat/trunk/webapps/docs/config/systemprops.xml?rev=1504124&r1=1504123&r2=1504124&view=diff == --- tomcat/trunk/webapps/docs/config/systemprops.xml (original) +++ tomcat/trunk/webapps/docs/config/systemprops.xml Wed Jul 17 13:24:30 2013 @@ -84,7 +84,7 @@ Number and Character types and false for Boolean as required by the EL 2.2 and earlier specifications. If this property is false the result of the coercion will be null as - required by the EL 3.0 specification. + required by the EL 3.0 specification. If not specified, the default value of false will be used. - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504127 - in /tomcat/trunk/java/org/apache/el: Messages.properties stream/Optional.java stream/Stream.java
Author: markt
Date: Wed Jul 17 13:28:25 2013
New Revision: 1504127
URL: http://svn.apache.org/r1504127
Log:
Provide messages with exceptions
Modified:
tomcat/trunk/java/org/apache/el/Messages.properties
tomcat/trunk/java/org/apache/el/stream/Optional.java
tomcat/trunk/java/org/apache/el/stream/Stream.java
Modified: tomcat/trunk/java/org/apache/el/Messages.properties
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/Messages.properties?rev=1504127&r1=1504126&r2=1504127&view=diff
==
--- tomcat/trunk/java/org/apache/el/Messages.properties (original)
+++ tomcat/trunk/java/org/apache/el/Messages.properties Wed Jul 17 13:28:25 2013
@@ -53,3 +53,7 @@ error.context.null=ELContext was null
# Parser
error.identifier.notjava=The identifier [{0}] is not a valid Java identifier
as required by section 1.19 of the EL specification (Identifier ::= Java
language identifier). This check can be disabled by setting the system property
org.apache.el.parser.SKIP_IDENTIFIER_CHECK to true.
error.lambda.tooManyMethodParameterSets=There are more sets of method
parameters specified than there are nested lambda expressions
+
+# Stream
+stream.optional.empty=It is illegal to call get() on an empty optional
+stream.compare.notComparable=Stream elements must implement Comparable
\ No newline at end of file
Modified: tomcat/trunk/java/org/apache/el/stream/Optional.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/stream/Optional.java?rev=1504127&r1=1504126&r2=1504127&view=diff
==
--- tomcat/trunk/java/org/apache/el/stream/Optional.java (original)
+++ tomcat/trunk/java/org/apache/el/stream/Optional.java Wed Jul 17 13:28:25
2013
@@ -19,6 +19,8 @@ package org.apache.el.stream;
import javax.el.ELException;
import javax.el.LambdaExpression;
+import org.apache.el.util.MessageFactory;
+
public class Optional {
private final Object obj;
@@ -32,7 +34,7 @@ public class Optional {
public Object get() throws ELException {
if (obj == null) {
-throw new ELException();
+throw new ELException(MessageFactory.get("stream.optional.empty"));
} else {
return obj;
}
Modified: tomcat/trunk/java/org/apache/el/stream/Stream.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/stream/Stream.java?rev=1504127&r1=1504126&r2=1504127&view=diff
==
--- tomcat/trunk/java/org/apache/el/stream/Stream.java (original)
+++ tomcat/trunk/java/org/apache/el/stream/Stream.java Wed Jul 17 13:28:25 2013
@@ -30,6 +30,7 @@ import javax.el.LambdaExpression;
import org.apache.el.lang.ELArithmetic;
import org.apache.el.lang.ELSupport;
+import org.apache.el.util.MessageFactory;
public class Stream {
@@ -412,7 +413,8 @@ public class Stream {
if ((obj instanceof Comparable)) {
result = (Comparable) obj;
} else {
-throw new ELException();
+throw new ELException(
+MessageFactory.get("stream.compare.notComparable"));
}
}
@@ -425,7 +427,8 @@ public class Stream {
result = (Comparable) obj;
}
} else {
-throw new ELException();
+throw new ELException(
+MessageFactory.get("stream.compare.notComparable"));
}
}
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: svn commit: r1503857 - in /tomcat/trunk: java/javax/el/ java/org/apache/el/parser/ test/org/apache/el/parser/
On 17/07/2013 12:43, Mark Thomas wrote:
> Violeta Georgieva wrote:
>
>> Having the example from the test:
>> (x->y->x-y)(2)(1)
>>
>> We may transform it like this:
>> (()->y->2-y)()(1)
>>
>> So instead of returning right way when formalParameters.isEmpty()
>>
>> Shouldn't we have something like the one below in order to continue
>> evaluation of the nested expression:
>>
>>
>>
>> --- org/apache/el/parser/AstLambdaExpression.java (revision 1504024)
>> +++ org/apache/el/parser/AstLambdaExpression.java (working copy)
>> @@ -54,9 +54,12 @@
>> LambdaExpression le = new LambdaExpression(formalParameters, ve);
>> le.setELContext(ctx);
>>
>> +Object result = le;
>> +int i = 2;
>> if (formalParameters.isEmpty()) {
>> // No formal parameters - invoke the expression
>> -return le.invoke(ctx, (Object[]) null);
>> +result = ((LambdaExpression) result).invoke(ctx,
>> (Object[])
>> null);
>> +i++;
>> }
>>
>> // If there are method parameters, need to invoke the expression
>> with
>> @@ -64,8 +67,6 @@
>> // should be that many nested expressions.
>> // If there are more nested expressions that parameters this will
>> return
>> // a LambdaExpression
>> -Object result = le;
>> -int i = 2;
>> while (result instanceof LambdaExpression && i <
>> jjtGetNumChildren()) {
>> result = ((LambdaExpression) result).invoke(
>> ((AstMethodParameters)
>> children[i]).getParameters(ctx));
>
> I agree the current solution isn't right but I have some local test cases
> that I'm fairly sure will still fail with the above patch. I'm working on a
> fix and should have something to commit soon.
Fix applied. What do you think?
Mark
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: Various shell-scripting idioms in bin/daemon.sh
On 17 July 2013 12:59, Mladen Turk wrote: > On 07/16/2013 11:42 PM, Christopher Schultz wrote: >> >> All, >> >> While doing the trivial fix for >> https://issues.apache.org/bugzilla/show_bug.cgi?id=55268, I noticed a >> few idioms being used in bin/daemon.sh that struck me as odd. For example: >> >> while [ ".$1" != . ] >> do >>case "$1" in >> --java-home ) >> JAVA_HOME="$2" >> shift; shift; >> continue >> ;; >> >> >> This example actually illustrates the two main questions I had: >> >> 1. Why use [ ".$FOO" != . ] instead of simply [ -n "$FOO" ] (Corollary: >> why use [ ".$FOO" = . ] instead of [ -z "$FOO" ])? >> > > Because some shell scripts does dot handle -z or -n well. > > >> 2. Why have a "continue" at the end of every case option, since the >> whole body of the while loop is nothing but the case construct? >> > > That might be an extra directive, true. > Probably a leftover from multiple case directives in while loop. > But safer than forgetting to add the continue if another case is ever added. > > >> I may be spoiled by using Linux and bash for most of my career, but I >> believe these are fairly standard POSIX-compliant things that should >> work on all *NIX systems. >> > > Sadly that's not the case IMHO. May I suggest a short comment is added to the script to document why -z and -n are not used? Someone else reading the script in the future is going to wonder the same. > > Regards > -- > ^TM > > - > To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org > For additional commands, e-mail: dev-h...@tomcat.apache.org > - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504148 - in /tomcat/trunk: java/org/apache/jasper/compiler/ java/org/apache/jasper/el/ java/org/apache/jasper/runtime/ test/javax/el/ test/org/apache/el/ test/org/apache/el/parser/ test/
Author: markt
Date: Wed Jul 17 14:14:28 2013
New Revision: 1504148
URL: http://svn.apache.org/r1504148
Log:
Add the two new resolver types (stream and static) to Jasper in the correct
order and modify JasperELResolver so the correct resolvers are skipped.
Modified:
tomcat/trunk/java/org/apache/jasper/compiler/Validator.java
tomcat/trunk/java/org/apache/jasper/el/ELContextImpl.java
tomcat/trunk/java/org/apache/jasper/el/ELResolverImpl.java
tomcat/trunk/java/org/apache/jasper/el/ExpressionEvaluatorImpl.java
tomcat/trunk/java/org/apache/jasper/el/ExpressionImpl.java
tomcat/trunk/java/org/apache/jasper/el/JasperELResolver.java
tomcat/trunk/java/org/apache/jasper/runtime/JspApplicationContextImpl.java
tomcat/trunk/test/javax/el/TestBeanELResolver.java
tomcat/trunk/test/javax/el/TestResourceBundleELResolver.java
tomcat/trunk/test/org/apache/el/TestELEvaluation.java
tomcat/trunk/test/org/apache/el/TestMethodExpressionImpl.java
tomcat/trunk/test/org/apache/el/TestValueExpressionImpl.java
tomcat/trunk/test/org/apache/el/parser/TestELParser.java
tomcat/trunk/test/org/apache/jasper/compiler/TestAttributeParser.java
tomcat/trunk/test/org/apache/jasper/el/TestJasperELResolver.java
Modified: tomcat/trunk/java/org/apache/jasper/compiler/Validator.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/jasper/compiler/Validator.java?rev=1504148&r1=1504147&r2=1504148&view=diff
==
--- tomcat/trunk/java/org/apache/jasper/compiler/Validator.java (original)
+++ tomcat/trunk/java/org/apache/jasper/compiler/Validator.java Wed Jul 17
14:14:28 2013
@@ -1215,7 +1215,8 @@ class Validator {
attrs.getQName(i), attrs.getURI(i),
attrs.getLocalName(i),
attrs.getValue(i), false, el,
false);
-ELContextImpl ctx = new ELContextImpl();
+ELContextImpl ctx = new ELContextImpl(
+expressionFactory);
ctx.setFunctionMapper(getFunctionMapper(el));
try {
jspAttrs[i].validateEL(this.pageInfo.getExpressionFactory(), ctx);
@@ -1377,7 +1378,8 @@ class Validator {
result = new Node.JspAttribute(tai, qName, uri,
localName, value, false, el, dynamic);
-ELContextImpl ctx = new ELContextImpl();
+ELContextImpl ctx =
+new ELContextImpl(expressionFactory);
ctx.setFunctionMapper(getFunctionMapper(el));
try {
@@ -1564,7 +1566,7 @@ class Validator {
validateFunctions(el, n);
// test it out
-ELContextImpl ctx = new ELContextImpl();
+ELContextImpl ctx = new ELContextImpl(expressionFactory);
ctx.setFunctionMapper(this.getFunctionMapper(el));
ExpressionFactory ef = this.pageInfo.getExpressionFactory();
try {
Modified: tomcat/trunk/java/org/apache/jasper/el/ELContextImpl.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/jasper/el/ELContextImpl.java?rev=1504148&r1=1504147&r2=1504148&view=diff
==
--- tomcat/trunk/java/org/apache/jasper/el/ELContextImpl.java (original)
+++ tomcat/trunk/java/org/apache/jasper/el/ELContextImpl.java Wed Jul 17
14:14:28 2013
@@ -25,10 +25,12 @@ import javax.el.BeanELResolver;
import javax.el.CompositeELResolver;
import javax.el.ELContext;
import javax.el.ELResolver;
+import javax.el.ExpressionFactory;
import javax.el.FunctionMapper;
import javax.el.ListELResolver;
import javax.el.MapELResolver;
import javax.el.ResourceBundleELResolver;
+import javax.el.StaticFieldELResolver;
import javax.el.ValueExpression;
import javax.el.VariableMapper;
@@ -95,8 +97,8 @@ public final class ELContextImpl extends
private VariableMapper variableMapper;
-public ELContextImpl() {
-this(getDefaultResolver());
+public ELContextImpl(ExpressionFactory factory) {
+this(getDefaultResolver(factory));
}
public ELContextImpl(ELResolver resolver) {
@@ -129,11 +131,11 @@ public final class ELContextImpl extends
this.variableMapper = variableMapper;
}
-public static ELResolver getDefaultResolver() {
+public static ELResolver getDefaultResolver(ExpressionFactory factory) {
if (Constants.IS_SECURITY_ENABLED) {
CompositeELResolver defaultResolver = new CompositeELResolver();
-// TODO ExpressionFactory.getStreamELResolver()
-// TODO javax.el.StaticFieldResolver
+
Re: svn commit: r1504148 - in /tomcat/trunk: java/org/apache/jasper/compiler/ java/org/apache/jasper/el/ java/org/apache/jasper/runtime/ test/javax/el/ test/org/apache/el/ test/org/apache/el/parser/ t
On Jul 17, 2013, at 9:21 AM, Caldarale, Charles R wrote: >> From: ma...@apache.org [mailto:ma...@apache.org] >> Subject: svn commit: r1504148 > >> URL: http://svn.apache.org/r1504148 >> Log: >> Add the two new resolver types (stream and static) to Jasper in the correct >> order and modify JasperELResolver so the correct resolvers are skipped. > > Did you really want to say "so the correct resolvers are skipped"? Seems odd > to be skipping the proper ones, or perhaps I just don't understand how this > should work. > > - Chuck I read "so the correct resolvers are skipped" as "so the resolvers that should be skipped are skipped." Nick - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: svn commit: r1504148 - in /tomcat/trunk: java/org/apache/jasper/compiler/ java/org/apache/jasper/el/ java/org/apache/jasper/runtime/ test/javax/el/ test/org/apache/el/ test/org/apache/el/parser/ t
On 17/07/2013 15:21, Caldarale, Charles R wrote: >> From: ma...@apache.org [mailto:ma...@apache.org] Subject: svn >> commit: r1504148 > >> URL: http://svn.apache.org/r1504148 Log: Add the two new resolver >> types (stream and static) to Jasper in the correct order and modify >> JasperELResolver so the correct resolvers are skipped. > > Did you really want to say "so the correct resolvers are skipped"? > Seems odd to be skipping the proper ones, or perhaps I just don't > understand how this should work. Yes. I was trying to say "Make sure we skip the resolvers that are meant to be skipped." since the new resolvers were inserted the indexes of the entries on the array have changed. Mark - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
RE: svn commit: r1504148 - in /tomcat/trunk: java/org/apache/jasper/compiler/ java/org/apache/jasper/el/ java/org/apache/jasper/runtime/ test/javax/el/ test/org/apache/el/ test/org/apache/el/parser/ t
> From: Nick Williams [mailto:nicho...@nicholaswilliams.net] > Subject: Re: svn commit: r1504148 > I read "so the correct resolvers are skipped" as "so the resolvers > that should be skipped are skipped." That interpretation makes much more sense. I think the use of the word "correct" twice in the sentence to refer to completely different sets of resolvers creates the confusion. - Chuck THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY MATERIAL and is thus for use only by the intended recipient. If you received this in error, please contact the sender and delete the e-mail and its attachments from all computers. - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504153 - in /tomcat/trunk/test: org/apache/el/TestELInJsp.java webapp/el-misc.jsp
Author: markt Date: Wed Jul 17 14:32:00 2013 New Revision: 1504153 URL: http://svn.apache.org/r1504153 Log: Add unit tests that check (very basically) that the new EL syntax works in a JSP. There are single tests for lambda expressions, streams and imports. TODO: Figure out why the import test currently fails. Modified: tomcat/trunk/test/org/apache/el/TestELInJsp.java tomcat/trunk/test/webapp/el-misc.jsp Modified: tomcat/trunk/test/org/apache/el/TestELInJsp.java URL: http://svn.apache.org/viewvc/tomcat/trunk/test/org/apache/el/TestELInJsp.java?rev=1504153&r1=1504152&r2=1504153&view=diff == --- tomcat/trunk/test/org/apache/el/TestELInJsp.java (original) +++ tomcat/trunk/test/org/apache/el/TestELInJsp.java Wed Jul 17 14:32:00 2013 @@ -388,6 +388,10 @@ public class TestELInJsp extends TomcatB assertEcho(result, "15-foo\\bar\\baz"); assertEcho(result, "16-foo\\bar\\baz"); assertEcho(result, "17-foo\\bar\\baz"); +assertEcho(result, "18-3"); +assertEcho(result, "19-4"); +// TODO Figure out why this doesn't work. +// assertEcho(result, "20-" + Integer.MAX_VALUE); } @Test Modified: tomcat/trunk/test/webapp/el-misc.jsp URL: http://svn.apache.org/viewvc/tomcat/trunk/test/webapp/el-misc.jsp?rev=1504153&r1=1504152&r2=1504153&view=diff == --- tomcat/trunk/test/webapp/el-misc.jsp (original) +++ tomcat/trunk/test/webapp/el-misc.jsp Wed Jul 17 14:32:00 2013 @@ -36,5 +36,8 @@ + + + \ No newline at end of file - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
RE: svn commit: r1504148 - in /tomcat/trunk: java/org/apache/jasper/compiler/ java/org/apache/jasper/el/ java/org/apache/jasper/runtime/ test/javax/el/ test/org/apache/el/ test/org/apache/el/parser/ t
> From: ma...@apache.org [mailto:ma...@apache.org] > Subject: svn commit: r1504148 > URL: http://svn.apache.org/r1504148 > Log: > Add the two new resolver types (stream and static) to Jasper in the correct > order and modify JasperELResolver so the correct resolvers are skipped. Did you really want to say "so the correct resolvers are skipped"? Seems odd to be skipping the proper ones, or perhaps I just don't understand how this should work. - Chuck THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY MATERIAL and is thus for use only by the intended recipient. If you received this in error, please contact the sender and delete the e-mail and its attachments from all computers. - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: Various shell-scripting idioms in bin/daemon.sh
On 07/17/2013 04:01 PM, sebb wrote: May I suggest a short comment is added to the script to document why -z and -n are not used? Someone else reading the script in the future is going to wonder the same-z and -n Some shells do not work well if variable is not set when using -z and -n. In case FOO was not set at all they break, but work if FOO="" as expected. Anyhow, IMHO 'if [ ".$FOO" != . ]' is more readable then 'if [ -n "$FOO" ]' I always end wondering if I should use -z or -n :) Regards -- ^TM - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
buildbot failure in ASF Buildbot on tomcat-trunk
The Buildbot has detected a new failure on builder tomcat-trunk while building ASF Buildbot. Full details are available at: http://ci.apache.org/builders/tomcat-trunk/builds/4707 Buildbot URL: http://ci.apache.org/ Buildslave for this Build: bb-vm_ubuntu Build Reason: scheduler Build Source Stamp: [branch tomcat/trunk] 1504148 Blamelist: markt BUILD FAILED: failed compile_1 sincerely, -The Buildbot
Re: Various shell-scripting idioms in bin/daemon.sh
On 17 July 2013 15:50, Mladen Turk wrote: > On 07/17/2013 04:01 PM, sebb wrote: >> >> May I suggest a short comment is added to the script to document why >> -z and -n are not used? >> Someone else reading the script in the future is going to wonder the >> same-z and -n > > > Some shells do not work well if variable is not set when using -z and -n. > In case FOO was not set at all they break, but work if FOO="" as expected. OK, well that information could be added to the script. > Anyhow, IMHO 'if [ ".$FOO" != . ]' is more readable then 'if [ -n "$FOO" ]' Why not use: if [ "$FOO" != "" ] The . is not very obvious in some fonts, and it's more symmetrical to quote both sides. > I always end wondering if I should use -z or -n :) > > > Regards > -- > ^TM > > - > To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org > For additional commands, e-mail: dev-h...@tomcat.apache.org > - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: Various shell-scripting idioms in bin/daemon.sh
On 07/17/2013 06:05 PM, sebb wrote: Why not use: if [ "$FOO" != "" ] Some shells do not allow that (comparing empty strings) The . is not very obvious in some fonts, and it's more symmetrical to quote both sides. Some time its used [ "x$FOO" != x ] The same way as '[ "x$FOO" = xyes ]', etc... Quoting both sides is just wasting two bytes, but I saw that as well, but it should be consistent trough the script. Anyhow, so far I saw no complaints about that. As far as its consistent, it's OK. Regards -- ^TM - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: svn commit: r1503857 - in /tomcat/trunk: java/javax/el/ java/org/apache/el/parser/ test/org/apache/el/parser/
2013/7/17 Mark Thomas wrote:
>
> On 17/07/2013 12:43, Mark Thomas wrote:
> > Violeta Georgieva wrote:
> >
> >> Having the example from the test:
> >> (x->y->x-y)(2)(1)
> >>
> >> We may transform it like this:
> >> (()->y->2-y)()(1)
> >>
> >> So instead of returning right way when formalParameters.isEmpty()
> >>
> >> Shouldn't we have something like the one below in order to continue
> >> evaluation of the nested expression:
> >>
> >>
> >>
> >> --- org/apache/el/parser/AstLambdaExpression.java (revision 1504024)
> >> +++ org/apache/el/parser/AstLambdaExpression.java (working copy)
> >> @@ -54,9 +54,12 @@
> >> LambdaExpression le = new LambdaExpression(formalParameters, ve);
> >> le.setELContext(ctx);
> >>
> >> +Object result = le;
> >> +int i = 2;
> >> if (formalParameters.isEmpty()) {
> >> // No formal parameters - invoke the expression
> >> -return le.invoke(ctx, (Object[]) null);
> >> +result = ((LambdaExpression) result).invoke(ctx,
> >> (Object[])
> >> null);
> >> +i++;
> >> }
> >>
> >> // If there are method parameters, need to invoke the expression
> >> with
> >> @@ -64,8 +67,6 @@
> >> // should be that many nested expressions.
> >> // If there are more nested expressions that parameters this will
> >> return
> >> // a LambdaExpression
> >> -Object result = le;
> >> -int i = 2;
> >> while (result instanceof LambdaExpression && i <
> >> jjtGetNumChildren()) {
> >> result = ((LambdaExpression) result).invoke(
> >> ((AstMethodParameters)
> >> children[i]).getParameters(ctx));
> >
> > I agree the current solution isn't right but I have some local test
cases that I'm fairly sure will still fail with the above patch. I'm
working on a fix and should have something to commit soon.
>
> Fix applied. What do you think?
Why it is expected that (()->y->()->x->x-y)()(1)()(2) should throw
ELException? (TestAstLambdaExpression.testNested04())
Isn't it a correct one?
svn commit: r1504232 - in /tomcat/trunk/test: org/apache/el/TestELInJsp.java webapp/el-misc.jsp
Author: markt
Date: Wed Jul 17 18:49:31 2013
New Revision: 1504232
URL: http://svn.apache.org/r1504232
Log:
The specification does not appear to allow the use of imports in a JSP page.
EL 3.0 section 1.5.1 states that the ELResolvers process identifiers before the
ImportHandler
JSP 2.3 section JSP.2.9 defines a set of ELResolvers that will always produce a
result hence the resolution of identifiers never progress to imports.
Modified:
tomcat/trunk/test/org/apache/el/TestELInJsp.java
tomcat/trunk/test/webapp/el-misc.jsp
Modified: tomcat/trunk/test/org/apache/el/TestELInJsp.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/test/org/apache/el/TestELInJsp.java?rev=1504232&r1=1504231&r2=1504232&view=diff
==
--- tomcat/trunk/test/org/apache/el/TestELInJsp.java (original)
+++ tomcat/trunk/test/org/apache/el/TestELInJsp.java Wed Jul 17 18:49:31 2013
@@ -370,6 +370,7 @@ public class TestELInJsp extends TomcatB
ByteChunk res = getUrl("http://localhost:"; + getPort() +
"/test/el-misc.jsp");
String result = res.toString();
+
assertEcho(result, "00-\\\"${'hello world'}");
assertEcho(result, "01-\\\"\\${'hello world'}");
assertEcho(result, "02-\\\"${'hello world'}");
@@ -390,8 +391,6 @@ public class TestELInJsp extends TomcatB
assertEcho(result, "17-foo\\bar\\baz");
assertEcho(result, "18-3");
assertEcho(result, "19-4");
-// TODO Figure out why this doesn't work.
-// assertEcho(result, "20-" + Integer.MAX_VALUE);
}
@Test
Modified: tomcat/trunk/test/webapp/el-misc.jsp
URL:
http://svn.apache.org/viewvc/tomcat/trunk/test/webapp/el-misc.jsp?rev=1504232&r1=1504231&r2=1504232&view=diff
==
--- tomcat/trunk/test/webapp/el-misc.jsp (original)
+++ tomcat/trunk/test/webapp/el-misc.jsp Wed Jul 17 18:49:31 2013
@@ -38,6 +38,5 @@
-
\ No newline at end of file
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504234 - /tomcat/trunk/java/javax/el/LambdaExpression.java
Author: markt
Date: Wed Jul 17 18:58:17 2013
New Revision: 1504234
URL: http://svn.apache.org/r1504234
Log:
Use the correct message key
Modified:
tomcat/trunk/java/javax/el/LambdaExpression.java
Modified: tomcat/trunk/java/javax/el/LambdaExpression.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/javax/el/LambdaExpression.java?rev=1504234&r1=1504233&r2=1504234&view=diff
==
--- tomcat/trunk/java/javax/el/LambdaExpression.java (original)
+++ tomcat/trunk/java/javax/el/LambdaExpression.java Wed Jul 17 18:58:17 2013
@@ -58,7 +58,7 @@ public class LambdaExpression {
if (formalParamCount > argCount) {
throw new ELException(Util.message(context,
-"error.lambda.args.tooFew",
+"lambdaExpression.tooFewArgs",
Integer.valueOf(argCount),
Integer.valueOf(formalParamCount)));
}
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: svn commit: r1503857 - in /tomcat/trunk: java/javax/el/ java/org/apache/el/parser/ test/org/apache/el/parser/
On 17/07/2013 19:17, Violeta Georgieva wrote:
> 2013/7/17 Mark Thomas wrote:
>>
>> On 17/07/2013 12:43, Mark Thomas wrote:
>>> Violeta Georgieva wrote:
>>>
Having the example from the test:
(x->y->x-y)(2)(1)
We may transform it like this:
(()->y->2-y)()(1)
So instead of returning right way when formalParameters.isEmpty()
Shouldn't we have something like the one below in order to continue
evaluation of the nested expression:
--- org/apache/el/parser/AstLambdaExpression.java (revision 1504024)
+++ org/apache/el/parser/AstLambdaExpression.java (working copy)
@@ -54,9 +54,12 @@
LambdaExpression le = new LambdaExpression(formalParameters, ve);
le.setELContext(ctx);
+Object result = le;
+int i = 2;
if (formalParameters.isEmpty()) {
// No formal parameters - invoke the expression
-return le.invoke(ctx, (Object[]) null);
+result = ((LambdaExpression) result).invoke(ctx,
(Object[])
null);
+i++;
}
// If there are method parameters, need to invoke the expression
with
@@ -64,8 +67,6 @@
// should be that many nested expressions.
// If there are more nested expressions that parameters this will
return
// a LambdaExpression
-Object result = le;
-int i = 2;
while (result instanceof LambdaExpression && i <
jjtGetNumChildren()) {
result = ((LambdaExpression) result).invoke(
((AstMethodParameters)
children[i]).getParameters(ctx));
>>>
>>> I agree the current solution isn't right but I have some local test
> cases that I'm fairly sure will still fail with the above patch. I'm
> working on a fix and should have something to commit soon.
>>
>> Fix applied. What do you think?
>
>
> Why it is expected that (()->y->()->x->x-y)()(1)()(2) should throw
> ELException? (TestAstLambdaExpression.testNested04())
> Isn't it a correct one?
That is a copy and paste error that is masking at least two bugs. I've
fixed one bug and am trying to track down why this expression won't
evaluate.
I need to look at your proposal above again.
Mark
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: Various shell-scripting idioms in bin/daemon.sh
Mladen, On 7/17/13 1:07 PM, Mladen Turk wrote: > On 07/17/2013 06:05 PM, sebb wrote: >> >> Why not use: >> >> if [ "$FOO" != "" ] >> > > Some shells do not allow that (comparing empty strings) I don't know of a shell where "$FOO" would be unset and yet expand to some non-zero-length string in a command. If that were the case, ".$FOO" would then expand to the same non-zero-length string with a dot pre-pended to it and the comparison wouldn't work, anyway. In my patch, I used a "-z". I'll remove it if there is significant concern that it is very non-standard. Sebb, one reason to use the ".$FOO" trick can be found in O'Reilly's /Unix Power Tools/. They point out that by testing for "$FOO" like this: if [ "$FOO" = "somestring" ] ; You run the risk of $FOO expanding to something like "-z" where it would not result in a string comparison. I like the use of ".$FOO" (leading dot) because it avoids this issue, but I think it's moot because of the availability of the "-z" test. I couldn't find a man page online for the "test" program (but of course searching for "test man page" yields completely useless results). -chris signature.asc Description: OpenPGP digital signature
svn commit: r1504256 - in /tomcat/trunk: java/org/apache/el/parser/AstLambdaExpression.java java/org/apache/el/parser/AstLambdaParameters.java java/org/apache/el/parser/AstMethodParameters.java test/o
Author: markt
Date: Wed Jul 17 20:07:48 2013
New Revision: 1504256
URL: http://svn.apache.org/r1504256
Log:
More nested lambda expression fixes.
Includes another test case and some simple debug code aimed at IDE debuggers.
Modified:
tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
tomcat/trunk/java/org/apache/el/parser/AstLambdaParameters.java
tomcat/trunk/java/org/apache/el/parser/AstMethodParameters.java
tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java
Modified: tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java?rev=1504256&r1=1504255&r2=1504256&view=diff
==
--- tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java (original)
+++ tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java Wed Jul 17
20:07:48 2013
@@ -95,9 +95,24 @@ public class AstLambdaExpression extends
result = ((LambdaExpression) result).invoke(
((AstMethodParameters) children[i]).getParameters(ctx));
i++;
+while (i < jjtGetNumChildren() && children[i].jjtGetNumChildren()
== 0) {
+i++;
+}
}
return result;
}
+
+
+@Override
+public String toString() {
+// Purely for debug purposes. May not be complete or correct. Certainly
+// is not efficient. Be sure not to call this from 'real' code.
+StringBuilder result = new StringBuilder();
+for (Node n : children) {
+result.append(n.toString());
+}
+return result.toString();
+}
}
/* JavaCC - OriginalChecksum=071159eff10c8e15ec612c765ae4480a (do not edit
this line) */
Modified: tomcat/trunk/java/org/apache/el/parser/AstLambdaParameters.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/parser/AstLambdaParameters.java?rev=1504256&r1=1504255&r2=1504256&view=diff
==
--- tomcat/trunk/java/org/apache/el/parser/AstLambdaParameters.java (original)
+++ tomcat/trunk/java/org/apache/el/parser/AstLambdaParameters.java Wed Jul 17
20:07:48 2013
@@ -17,10 +17,27 @@
/* Generated By:JJTree: Do not edit this line. AstLambdaParameters.java
Version 4.3 */
package org.apache.el.parser;
-public
-class AstLambdaParameters extends SimpleNode {
- public AstLambdaParameters(int id) {
-super(id);
- }
+public class AstLambdaParameters extends SimpleNode {
+
+public AstLambdaParameters(int id) {
+super(id);
+}
+
+@Override
+public String toString() {
+// Purely for debug purposes. May not be complete or correct. Certainly
+// is not efficient. Be sure not to call this from 'real' code.
+StringBuilder result = new StringBuilder();
+result.append('(');
+if (children != null) {
+for (Node n : children) {
+result.append(n.toString());
+result.append(',');
+}
+}
+result.append(")->");
+return result.toString();
+}
+
}
/* JavaCC - OriginalChecksum=a8c1609257dac59e41c43d6ed91072c6 (do not edit
this line) */
Modified: tomcat/trunk/java/org/apache/el/parser/AstMethodParameters.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/parser/AstMethodParameters.java?rev=1504256&r1=1504255&r2=1504256&view=diff
==
--- tomcat/trunk/java/org/apache/el/parser/AstMethodParameters.java (original)
+++ tomcat/trunk/java/org/apache/el/parser/AstMethodParameters.java Wed Jul 17
20:07:48 2013
@@ -37,4 +37,20 @@ public final class AstMethodParameters e
}
return params.toArray(new Object[params.size()]);
}
+
+@Override
+public String toString() {
+// Purely for debug purposes. May not be complete or correct. Certainly
+// is not efficient. Be sure not to call this from 'real' code.
+StringBuilder result = new StringBuilder();
+result.append('(');
+if (children != null) {
+for (Node n : children) {
+result.append(n.toString());
+result.append(',');
+}
+}
+result.append(')');
+return result.toString();
+}
}
Modified: tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java?rev=1504256&r1=1504255&r2=1504256&view=diff
==
--- tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java
(original)
+++ tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java Wed Jul
17 20:07:48 2013
@@ -116,7 +116,7 @@ public cl
Re: svn commit: r1503857 - in /tomcat/trunk: java/javax/el/ java/org/apache/el/parser/ test/org/apache/el/parser/
On 17/07/2013 20:12, Mark Thomas wrote: > On 17/07/2013 19:17, Violeta Georgieva wrote: >> Why it is expected that (()->y->()->x->x-y)()(1)()(2) should throw >> ELException? (TestAstLambdaExpression.testNested04()) >> Isn't it a correct one? > > That is a copy and paste error that is masking at least two bugs. I've > fixed one bug and am trying to track down why this expression won't > evaluate. > > I need to look at your proposal above again. I added some debug code that helped track down what is going on. I've fixed that issue but I'm fairly sure the logic could be clarified. I'm planning on looking at that next. Mark - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504281 - in /tomcat/trunk: java/org/apache/el/parser/AstLambdaExpression.java test/org/apache/el/parser/TestAstLambdaExpression.java
Author: markt
Date: Wed Jul 17 21:12:22 2013
New Revision: 1504281
URL: http://svn.apache.org/r1504281
Log:
More fun and games with nested lambda expressions.
Each invocation of a nested expression consumes a set of method parameters.
Therefore nested lambda expressions that are invoked immediately (because they
have no formal parameetrs) need to inform the outer expression of the
invocation so the next invocation uses the correct method parameters.
Modified:
tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
tomcat/trunk/test/org/apache/el/parser/TestAstLambdaExpression.java
Modified: tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java?rev=1504281&r1=1504280&r2=1504281&view=diff
==
--- tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java (original)
+++ tomcat/trunk/java/org/apache/el/parser/AstLambdaExpression.java Wed Jul 17
21:12:22 2013
@@ -29,6 +29,8 @@ import org.apache.el.util.MessageFactory
public class AstLambdaExpression extends SimpleNode {
+private int methodParameterIndex = 0;
+
public AstLambdaExpression(int id) {
super(id);
}
@@ -73,37 +75,70 @@ public class AstLambdaExpression extends
LambdaExpression le = new LambdaExpression(formalParameters, ve);
le.setELContext(ctx);
-if (formalParameters.isEmpty() && jjtGetNumChildren() == 2) {
-// No formal parameters - invoke the expression
-return le.invoke(ctx, (Object[]) null);
-}
-
-// If there are method parameters, need to invoke the expression with
-// those parameters. If there are multiple sets of method parameters
-// there should be at least that many nested expressions.
-// If there are more nested expressions than sets of method parameters
-// this may return a LambdaExpression.
-// If there are more sets of method parameters than nested expressions
-// an ELException will have been thrown by the check at the start of
-// this method.
-// If the inner most expression(s) do not require parameters then a
-// value will be returned once the outermost expression that does
-// require a parameter has been evaluated.
-Object result = le;
-int i = 2;
-while (result instanceof LambdaExpression && i < jjtGetNumChildren()) {
-result = ((LambdaExpression) result).invoke(
-((AstMethodParameters) children[i]).getParameters(ctx));
-i++;
-while (i < jjtGetNumChildren() && children[i].jjtGetNumChildren()
== 0) {
-i++;
+if (jjtGetNumChildren() == 2) {
+if (formalParameters.isEmpty()) {
+// No formal parameters or method parameters so invoke the
+// expression. If this is a nested expression inform the outer
+// expression that an invocation has occurred so the correct
set
+// of method parameters are used for the next invocation.
+incMethodParameterIndex();
+return le.invoke(ctx, (Object[]) null);
+} else {
+// Has formal parameters but no method parameters so return the
+// expression for later evaluation
+return le;
}
}
+
+// Always have to invoke the outer-most expression
+methodParameterIndex = 2;
+Object result = le.invoke(((AstMethodParameters)
+children[methodParameterIndex]).getParameters(ctx));
+methodParameterIndex++;
+
+/*
+ * If there are multiple sets of method parameters there should be at
+ * least that many nested expressions.
+ *
+ * If there are more nested expressions than sets of method parameters
+ * this may return a LambdaExpression.
+ *
+ * If there are more sets of method parameters than nested expressions
+ * an ELException will have been thrown by the check at the start of
+ * this method.
+ *
+ * If the inner most expression(s) do not require parameters then a
+ * value will be returned once the outermost expression that does
+ * require a parameter has been evaluated.
+ *
+ * When invoking an expression if it has nested expressions that do not
+ * have formal parameters then they will be evaluated as as part of
that
+ * invocation. In this case the method parameters associated with those
+ * nested expressions need to be skipped.
+ */
+while (result instanceof LambdaExpression &&
+methodParameterIndex < jjtGetNumChildren()) {
+result = ((LambdaExpression) result).invoke(((AstMethodParameters)
+
Re: svn commit: r1503857 - in /tomcat/trunk: java/javax/el/ java/org/apache/el/parser/ test/org/apache/el/parser/
On 17/07/2013 21:10, Mark Thomas wrote: > On 17/07/2013 20:12, Mark Thomas wrote: >> On 17/07/2013 19:17, Violeta Georgieva wrote: > >>> Why it is expected that (()->y->()->x->x-y)()(1)()(2) should throw >>> ELException? (TestAstLambdaExpression.testNested04()) >>> Isn't it a correct one? >> >> That is a copy and paste error that is masking at least two bugs. I've >> fixed one bug and am trying to track down why this expression won't >> evaluate. >> >> I need to look at your proposal above again. > > I added some debug code that helped track down what is going on. > > I've fixed that issue but I'm fairly sure the logic could be clarified. > I'm planning on looking at that next. As I was writing the comments for the simpler code, I realised there was still a way to break it. I added to test case to prove that and then fixed the evaluation again. At the moment I think lambda expressions are working correctly but to be frank I'm just waiting for someone to find the next tricky nested expression that breaks things again. Violeta, over to you :) Mark - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504286 - in /tomcat/trunk: java/org/apache/el/parser/AstIdentifier.java test/org/apache/el/parser/TestAstIdentifier.java
Author: markt
Date: Wed Jul 17 21:38:48 2013
New Revision: 1504286
URL: http://svn.apache.org/r1504286
Log:
Add some basic import tests and fix a couple of identified issues.
Added:
tomcat/trunk/test/org/apache/el/parser/TestAstIdentifier.java (with props)
Modified:
tomcat/trunk/java/org/apache/el/parser/AstIdentifier.java
Modified: tomcat/trunk/java/org/apache/el/parser/AstIdentifier.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/el/parser/AstIdentifier.java?rev=1504286&r1=1504285&r2=1504286&view=diff
==
--- tomcat/trunk/java/org/apache/el/parser/AstIdentifier.java (original)
+++ tomcat/trunk/java/org/apache/el/parser/AstIdentifier.java Wed Jul 17
21:38:48 2013
@@ -18,6 +18,7 @@
package org.apache.el.parser;
+import javax.el.ELClass;
import javax.el.ELException;
import javax.el.MethodExpression;
import javax.el.MethodInfo;
@@ -78,11 +79,27 @@ public final class AstIdentifier extends
// EL Resolvers
ctx.setPropertyResolved(false);
Object result = ctx.getELResolver().getValue(ctx, null, this.image);
-if (!ctx.isPropertyResolved()) {
-throw new PropertyNotFoundException(MessageFactory.get(
-"error.resolver.unhandled.null", this.image));
+if (ctx.isPropertyResolved()) {
+return result;
}
-return result;
+
+// Import
+result = ctx.getImportHandler().resolveClass(this.image);
+if (result != null) {
+return new ELClass((Class) result);
+}
+result = ctx.getImportHandler().resolveStatic(this.image);
+if (result != null) {
+try {
+return ((Class) result).getField(this.image).get(null);
+} catch (IllegalArgumentException | IllegalAccessException
+| NoSuchFieldException | SecurityException e) {
+throw new ELException(e);
+}
+}
+
+throw new PropertyNotFoundException(MessageFactory.get(
+"error.resolver.unhandled.null", this.image));
}
@Override
Added: tomcat/trunk/test/org/apache/el/parser/TestAstIdentifier.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/test/org/apache/el/parser/TestAstIdentifier.java?rev=1504286&view=auto
==
--- tomcat/trunk/test/org/apache/el/parser/TestAstIdentifier.java (added)
+++ tomcat/trunk/test/org/apache/el/parser/TestAstIdentifier.java Wed Jul 17
21:38:48 2013
@@ -0,0 +1,46 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.el.parser;
+
+import javax.el.ELProcessor;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+public class TestAstIdentifier {
+
+@Test
+public void testImport01() {
+ELProcessor processor = new ELProcessor();
+Object result =
+processor.getValue("Integer.MAX_VALUE",
+Integer.class);
+Assert.assertEquals(Integer.valueOf(Integer.MAX_VALUE), result);
+}
+
+
+@Test
+public void testImport02() {
+ELProcessor processor = new ELProcessor();
+
processor.getELManager().getELContext().getImportHandler().importStatic(
+"java.lang.Integer.MAX_VALUE");
+Object result =
+processor.getValue("MAX_VALUE",
+Integer.class);
+Assert.assertEquals(Integer.valueOf(Integer.MAX_VALUE), result);
+}
+}
Propchange: tomcat/trunk/test/org/apache/el/parser/TestAstIdentifier.java
--
svn:eol-style = native
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504287 - /tomcat/trunk/test/org/apache/jasper/el/TestJasperELResolver.java
Author: markt
Date: Wed Jul 17 21:40:19 2013
New Revision: 1504287
URL: http://svn.apache.org/r1504287
Log:
Fix failing CI build. Another resolver was added.
Modified:
tomcat/trunk/test/org/apache/jasper/el/TestJasperELResolver.java
Modified: tomcat/trunk/test/org/apache/jasper/el/TestJasperELResolver.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/test/org/apache/jasper/el/TestJasperELResolver.java?rev=1504287&r1=1504286&r2=1504287&view=diff
==
--- tomcat/trunk/test/org/apache/jasper/el/TestJasperELResolver.java (original)
+++ tomcat/trunk/test/org/apache/jasper/el/TestJasperELResolver.java Wed Jul 17
21:40:19 2013
@@ -58,9 +58,9 @@ public class TestJasperELResolver {
Assert.assertEquals(Integer.valueOf(count),
getField("appResolversSize", resolver));
-Assert.assertEquals(8 + count,
+Assert.assertEquals(9 + count,
((ELResolver[])getField("resolvers", resolver)).length);
-Assert.assertEquals(Integer.valueOf(8 + count),
+Assert.assertEquals(Integer.valueOf(9 + count),
getField("size", resolver));
}
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
svn commit: r1504292 - /tomcat/trunk/RELEASE-NOTES
Author: markt
Date: Wed Jul 17 21:46:15 2013
New Revision: 1504292
URL: http://svn.apache.org/r1504292
Log:
Prep for an 8.0.0-alpha release
Remove section on JNI. Not a frequent enough issue to mentioned in the release
notes.
Clean up the contents list.
Modified:
tomcat/trunk/RELEASE-NOTES
Modified: tomcat/trunk/RELEASE-NOTES
URL:
http://svn.apache.org/viewvc/tomcat/trunk/RELEASE-NOTES?rev=1504292&r1=1504291&r2=1504292&view=diff
==
--- tomcat/trunk/RELEASE-NOTES (original)
+++ tomcat/trunk/RELEASE-NOTES Wed Jul 17 21:46:15 2013
@@ -28,11 +28,8 @@ CONTENTS:
* Dependency Changes
* API Stability
-* JNI Based Applications
* Bundled APIs
* Web application reloading and static fields in shared libraries
-* Tomcat on Linux
-* Enabling SSI and CGI Support
* Security manager URLs
* Symlinking static resources
* Viewing the Tomcat Change Log
@@ -43,20 +40,16 @@ CONTENTS:
===
Dependency Changes:
===
-Tomcat @VERSION_MAJOR_MINOR@ is designed to run on Java SE 6 and later.
-
-In addition, Tomcat @VERSION_MAJOR_MINOR@ uses the Eclipse JDT Java compiler
for
-compiling JSP pages. This means you no longer need to have the complete
-Java Development Kit (JDK) to run Tomcat, but a Java Runtime Environment
-(JRE) is sufficient. The Eclipse JDT Java compiler is bundled with the
-binary Tomcat distributions. Tomcat can also be configured to use the
-compiler from the JDK to compile JSPs, or any other Java compiler supported
-by Apache Ant.
+Tomcat @VERSION_MAJOR_MINOR@ is designed to run on Java SE 7 and later.
==
API Stability:
==
+
+*** This is an alpha release. Currently no guarantees are made regarding API
***
+*** stability.
***
+
The public interfaces for the following classes are fixed and will not be
changed at all during the remaining lifetime of the @VERSION_MAJOR@.x series:
- None
@@ -73,27 +66,6 @@ The remaining classes are considered par
without notice between point releases.
-===
-JNI Based Applications:
-===
-Applications that require native libraries must ensure that the libraries have
-been loaded prior to use. Typically, this is done with a call like:
-
- static {
-System.loadLibrary("path-to-library-file");
- }
-
-in some class. However, the application must also ensure that the library is
-not loaded more than once. If the above code were placed in a class inside
-the web application (i.e. under /WEB-INF/classes or /WEB-INF/lib), and the
-application were reloaded, the loadLibrary() call would be attempted a second
-time.
-
-To avoid this problem, place classes that load native libraries outside of the
-web application, and ensure that the loadLibrary() call is executed only once
-during the lifetime of a particular JVM.
-
-
=
Bundled APIs:
=
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
Re: Various shell-scripting idioms in bin/daemon.sh
Regardless, please consider documenting the script to explain why it does not use -n/-z if that is necessary to avoid bugs. On 17 July 2013 20:26, Christopher Schultz wrote: > Mladen, > > On 7/17/13 1:07 PM, Mladen Turk wrote: >> On 07/17/2013 06:05 PM, sebb wrote: >>> >>> Why not use: >>> >>> if [ "$FOO" != "" ] >>> >> >> Some shells do not allow that (comparing empty strings) > > I don't know of a shell where "$FOO" would be unset and yet expand to > some non-zero-length string in a command. If that were the case, ".$FOO" > would then expand to the same non-zero-length string with a dot > pre-pended to it and the comparison wouldn't work, anyway. > > In my patch, I used a "-z". I'll remove it if there is significant > concern that it is very non-standard. > > Sebb, one reason to use the ".$FOO" trick can be found in O'Reilly's > /Unix Power Tools/. They point out that by testing for "$FOO" like this: > > if [ "$FOO" = "somestring" ] ; > > You run the risk of $FOO expanding to something like "-z" where it would > not result in a string comparison. I like the use of ".$FOO" (leading > dot) because it avoids this issue, but I think it's moot because of the > availability of the "-z" test. > > I couldn't find a man page online for the "test" program (but of course > searching for "test man page" yields completely useless results). > > -chris > - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
buildbot success in ASF Buildbot on tomcat-trunk
The Buildbot has detected a restored build on builder tomcat-trunk while building ASF Buildbot. Full details are available at: http://ci.apache.org/builders/tomcat-trunk/builds/4713 Buildbot URL: http://ci.apache.org/ Buildslave for this Build: bb-vm_ubuntu Build Reason: scheduler Build Source Stamp: [branch tomcat/trunk] 1504292 Blamelist: markt Build succeeded! sincerely, -The Buildbot
Re: Various shell-scripting idioms in bin/daemon.sh
On 07/17/2013 11:59 PM, sebb wrote: Regardless, please consider documenting the script to explain why it does not use -n/-z if that is necessary to avoid bugs. It would be the same as documenting why one uses a+=1 instead a++ :) I don't see where its written that one *must* use -n/-z at the first place. Regards -- ^TM - To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
