[jira] [Commented] (GROOVY-9599) CliBuilder: Option with "type: String, defaultValue ''" (empty String) results in NullObject if default value is applied
[
https://issues.apache.org/jira/browse/GROOVY-9599?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141714#comment-17141714
]
Dirk Heinrichs commented on GROOVY-9599:
Great, thanks a lot [~rpopma]!
> CliBuilder: Option with "type: String, defaultValue ''" (empty String)
> results in NullObject if default value is applied
>
>
> Key: GROOVY-9599
> URL: https://issues.apache.org/jira/browse/GROOVY-9599
> Project: Groovy
> Issue Type: Bug
> Components: command line processing
>Affects Versions: 2.5.11, 3.0.4
>Reporter: Remko Popma
>Assignee: Remko Popma
>Priority: Major
> Fix For: 3.0.5, 2.5.13
>
>
> This is a follow-up ticket for GROOVY-9519:
> After checking with Groovy 3.0.4, it seems that only the "Integer with
> defaultValue '0'" case was fixed, but not "String with defaultValue ''". The
> latter became even worse. It still ignores the defaultValue, but instead of
> creating a Boolean, it now creates a NullObject. Try this code:
> {code:java}
> @Grab('info.picocli:picocli-groovy:4.3.2')
> @GrabConfig(systemClassLoader=true)
> import groovy.cli.picocli.CliBuilder
> def cli = new CliBuilder(name: 'cliTest.groovy', stopAtNonOption: false)
> cli.h(type: Boolean, longOpt: 'help', usageHelp: true, required: false, 'Show
> usage information')
> cli.a(type: String, longOpt: 'optA', required: false, args: 1, defaultValue:
> '', 'Option a (optional)')
> def opts = cli.parse(args)
> opts || System.exit(1)
> if(opts.h) {
> cli.usage()
> System.exit(0)
> }
> println(opts.a.getClass())
> if (opts.a) {
> println(opts.a)
> }
> {code}
> with Groovy 3.0.2:
> {code:java}
> % ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy
> class java.lang.Boolean
> {code}
> and with Groovy 3.0.4:
> {code:java}
> % ~/tmp/groovy-3.0.4/bin/groovy ./cliTest.groovy
> class org.codehaus.groovy.runtime.NullObject
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GROOVY-9596) Cannot access protected field of Java superclass with --illegal-access=deny
[
https://issues.apache.org/jira/browse/GROOVY-9596?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141711#comment-17141711
]
Paul King commented on GROOVY-9596:
---
Using `this.@in` or `super.@in` results (for both static and dynamic variants)
in:
{code}
getfield #gg// Field
java/io/FilterReader.in:Ljava/io/Reader;
{code}
So, that would be the correct workaround for now.
> Cannot access protected field of Java superclass with --illegal-access=deny
> ---
>
> Key: GROOVY-9596
> URL: https://issues.apache.org/jira/browse/GROOVY-9596
> Project: Groovy
> Issue Type: Bug
> Components: bytecode
>Affects Versions: 3.0.4, 2.5.12
> Environment: JDK 14
> Groovy 2.5.12 or 3.0.4
>Reporter: Kevin Rushforth
>Priority: Major
> Attachments: Main.groovy, Main2.groovy, MyAWTEvent.groovy,
> MyFilterReader.groovy
>
>
> When using the Java option {{--illegal-access=deny}}, a Groovy subclass that
> extends a Java superclass in the Java runtime (e.g., a class in the
> {{java.base}} module) cannot access a protected field in the parent class
> using {{this.fieldName}}.
> To reproduce this bug, use JDK 14 (or later) and run the following using
> either Groovy 2.5.12 or 3.0.4:
> {{$ groovyc Main.groovy MyFilterReader.groovy}}
> {{$ _JAVA_OPTIONS="--illegal-access=deny" groovy Main}}
> It works correctly without {{--illegal-access=deny}}. You can add
> {{-Dsun.reflect.debugModuleAccessChecks=access}} to get more detailed
> information about the illegal access exceptions that are being swallowed by
> Groovy.
> {{$ _JAVA_OPTIONS="--illegal-access=deny
> -Dsun.reflect.debugModuleAccessChecks=access" groovy Main}}
> WORKAROUND: change the following line in the Groovy subclass:
> line = ((BufferedReader)this.in).readLine();
> to:
> line = ((BufferedReader)super.in).readLine();
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GROOVY-4856) Closure resolveStrategy does not work the same way for properties and methods
[
https://issues.apache.org/jira/browse/GROOVY-4856?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141680#comment-17141680
]
Eric Milles commented on GROOVY-4856:
-
AsmClassGenerator checks for "this", "super" or defined variables before
anything else.
{code:java}
public void visitVariableExpression(final VariableExpression expression) {
final String variableName = expression.getName();
if (expression.isThisExpression()) {
...
}
if (expression.isSuperExpression()) {
...
}
BytecodeVariable variable =
controller.getCompileStack().getVariable(variableName, false);
if (variable != null) {
controller.getOperandStack().loadOrStoreVariable(variable,
expression.isUseReferenceDirectly());
} else ...
{code}
I'm not sure if it is intended behavior to have local or script variables take
precedence over delegate or owner properties. However, you can add a
"delegate." qualifier to disambiguate.
> Closure resolveStrategy does not work the same way for properties and methods
> -
>
> Key: GROOVY-4856
> URL: https://issues.apache.org/jira/browse/GROOVY-4856
> Project: Groovy
> Issue Type: Bug
> Components: Compiler
>Affects Versions: 1.7.6
> Environment: Mac OS 10.6.7, Groovy installed via Ports
>Reporter: Guven Demir
>Priority: Major
>
> When the resolveStrategy is set to DELEGATE_FIRST and the delegate is set
> property, a method with the same name as a global function is preferred
> instead of the global function. The same is not true for properties.
> The following code fragment demonstrates this issue:
> -
> {noformat}
> class ClosureTest
> {
> def someProperty = ""
> def anotherProperty = ""
> public void test(Closure closure)
> {
> closure.delegate = this
> closure.resolveStrategy = Closure.DELEGATE_FIRST
> closure()
> }
>
> public void someMethod()
> {
> println "ClosureTest.someMethod()"
> }
> public void anotherMethod()
> {
> println "ClosureTest.anotherMethod()"
> }
>
> public String toString()
> {
> return "ClosureTest: sp: " + someProperty + ", ap: " + anotherProperty
> }
> }
> def someProperty = "global someProperty"
> public void someMethod()
> {
> println "global someMethod()"
> }
> def ct = new ClosureTest()
> ct.test() {
> someMethod()
> anotherMethod()
> someProperty = "foo"
> anotherProperty = "bar"
> }
> println "ct.someProperty : ${ct.someProperty}"
> println "ct.anotherProperty : ${ct.anotherProperty}"
> println "global: someProperty: " + someProperty
> {noformat}
> -
> And the output of this script run is:
> {noformat}
> ClosureTest.someMethod()
> ClosureTest.anotherMethod()
> ct.someProperty :
> ct.anotherProperty : bar
> global: someProperty: foo
> {noformat}
> -
> As seen above, the methods are called correctly, but the properties are not
> set correctly.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GROOVY-7480) ClosureMetaClass wrong routing calls logic
[
https://issues.apache.org/jira/browse/GROOVY-7480?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141675#comment-17141675
]
Eric Milles commented on GROOVY-7480:
-
When making an implicit-this method call from within a closure, methods on
Closure itself take precedence. This is so you can call "getDelegate()" or
"trampoline()" for example. In this case, you are effectively calling
"c.leftShift('*')", since leftShift is a method implemented by
groovy.lang.Closure.
You can add a "delegate." or "getDelegate()." qualifier to your leftShift call
to direct it away from Closure.
{code:java}
/**
* Alias for calling a Closure for non-closure arguments.
*
* Typical usage:
*
* def times2 = { a {@code ->} a * 2 }
* def add3 = { a {@code ->} a + 3 }
* assert add3 {@code <<} times2 {@code <<} 3 == 9
*
*
* @param arg the argument to call the closure with
* @return the result of calling the Closure
*/
public V leftShift(final Object arg) {
return call(arg);
}
{code}
> ClosureMetaClass wrong routing calls logic
> --
>
> Key: GROOVY-7480
> URL: https://issues.apache.org/jira/browse/GROOVY-7480
> Project: Groovy
> Issue Type: Bug
> Components: groovy-runtime
>Affects Versions: 2.4.2
>Reporter: Paolo Piersanti
>Priority: Major
>
> Setting up the resolve strategy to Closure.DELEGATE_FIRST does not
> get the delegate looked up at first. For instance:
> target = 'aTarget'
> c = new GroovyShell().evaluate("{ -> leftShift('*') }");
> c.resolveStrategy = Closure.DELEGATE_FIRST
> c.delegate = target
> c.call()
> results in
> groovy.lang.MissingMethodException: No signature of method:
> Script1$_run_closure1.doCall() is applicable for argument types:
> (java.lang.String) values: [*]
> Possible solutions: doCall(), findAll(), findAll(),
> isCase(java.lang.Object), isCase(java.lang.Object)
> at Script1$_run_closure1.doCall(Script1.groovy:1)
> at ConsoleScript33.run(ConsoleScript33:5)
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GROOVY-9605) leftShift operator does not work on BigInteger (throws UnsupportedOperationException)
[
https://issues.apache.org/jira/browse/GROOVY-9605?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141674#comment-17141674
]
Paul King commented on GROOVY-9605:
---
Both leftShift and rightShift can be accommodated for BigInteger by adjusting
BigIntegerMath as you spotted. It's an easy change which I have made locally.
I'll just check tests and doco before creating a PR for review.
> leftShift operator does not work on BigInteger (throws
> UnsupportedOperationException)
> -
>
> Key: GROOVY-9605
> URL: https://issues.apache.org/jira/browse/GROOVY-9605
> Project: Groovy
> Issue Type: Bug
> Components: groovy-jdk
>Affects Versions: 3.0.4
> Environment: Groovy 3.0.4 Windows and Linux
>Reporter: Bertrand Martin
>Assignee: Paul King
>Priority: Major
>
> h1. Problem
> The _leftShift_ operator ({{<<}}) does not work on *BigInteger* (see
> [StackOverflow
> question|https://stackoverflow.com/questions/62496971/leftshift-method-on-biginteger-in-groovy]).
> The *leftShift()* method is however described in [Groovy's JDK
> enhancements|http://docs.groovy-lang.org/latest/html/api/org/codehaus/groovy/runtime/DefaultGroovyMethods.html#leftShift(java.lang.Number,java.lang.Number)].
> The below code fails with Groovy 3.0.4:
> {code:groovy}
> Integer i = new Integer(3)
> assert 12 == i << 2
> Number b = new BigInteger("3")
> assert 12 == b << 2
> {code}
> It throws an exception:
> {noformat}
> Caught: java.lang.UnsupportedOperationException: Cannot use leftShift() on
> this number type: java.math.BigInteger with value: 3
> java.lang.UnsupportedOperationException: Cannot use leftShift() on this
> number type: java.math.BigInteger with value: 3
> at test.run(test.groovy:5)
> {noformat}
> Consequence: you cannot use the {{<<}} operator on a _Number_ if this number
> is a _BigInteger_.
> h1. Understanding of the problem
> (I'm not familiar with the code)
> The
> [BigIntegerMath|https://github.com/groovy/groovy-core/blob/master/src/main/org/codehaus/groovy/runtime/typehandling/BigIntegerMath.java]
> class does not implement the _leftShift()_ method.
> Also, please note that JDK's
> [BigInteger|https://docs.oracle.com/javase/8/docs/api/java/math/BigInteger.html#shiftLeft-int-]
> method for this operation is *not* _leftShift()_ but _shiftLeft()_!
> h1. Workaround
> Using mixins, the below code works:
> {code:groovy}
> class EnhancedNumber {
> static def originalLeftShift =
> org.codehaus.groovy.runtime.DefaultGroovyMethods.
> static Number leftShift(Number self, Number operand) {
> self.class == BigInteger ? self.shiftLeft(operand) :
> originalLeftShift(self, operand)
> }
> }
> Number.mixin(EnhancedNumber)
> Integer i = new Integer(3)
> assert 12 == i << 2
> Number b = new BigInteger("3")
> assert 12 == b << 2
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[GitHub] [groovy] eric-milles opened a new pull request #1286: GROOVY-8274: allow user-supplied methodMissing in non-static inner class
eric-milles opened a new pull request #1286: URL: https://github.com/apache/groovy/pull/1286 InnerClassCompletionVisitor overwrote user's methodMissing or propertyMissing with synthetic dispatcher. https://issues.apache.org/jira/browse/GROOVY-8274 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[jira] [Assigned] (GROOVY-8274) methodMissing not executed when the delegate is an inner class
[
https://issues.apache.org/jira/browse/GROOVY-8274?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Eric Milles reassigned GROOVY-8274:
---
Assignee: Eric Milles
> methodMissing not executed when the delegate is an inner class
> --
>
> Key: GROOVY-8274
> URL: https://issues.apache.org/jira/browse/GROOVY-8274
> Project: Groovy
> Issue Type: Bug
> Components: groovy-runtime
>Affects Versions: 2.4.12
>Reporter: James Kleeh
>Assignee: Eric Milles
>Priority: Major
>
> {code}
> class Foo {
>
> void executeClosure(Closure c) {
> c.resolveStrategy = Closure.DELEGATE_ONLY
> c.delegate = new Bar()
> c.call()
> }
> class Bar {
> def methodMissing(String name, args) {
> println name
> }
> }
> }
> Closure c = {
> go()
> }
> new Foo().executeClosure(c)
> {code}
> Executing the above results in:
> {{groovy.lang.MissingMethodException: No signature of method: Foo.go() is
> applicable for argument types: () values: []}}
> If the Bar class is moved out of Foo, it works as I expect.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Assigned] (GROOVY-9605) leftShift operator does not work on BigInteger (throws UnsupportedOperationException)
[
https://issues.apache.org/jira/browse/GROOVY-9605?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Paul King reassigned GROOVY-9605:
-
Assignee: Paul King
> leftShift operator does not work on BigInteger (throws
> UnsupportedOperationException)
> -
>
> Key: GROOVY-9605
> URL: https://issues.apache.org/jira/browse/GROOVY-9605
> Project: Groovy
> Issue Type: Bug
> Components: groovy-jdk
>Affects Versions: 3.0.4
> Environment: Groovy 3.0.4 Windows and Linux
>Reporter: Bertrand Martin
>Assignee: Paul King
>Priority: Major
>
> h1. Problem
> The _leftShift_ operator ({{<<}}) does not work on *BigInteger* (see
> [StackOverflow
> question|https://stackoverflow.com/questions/62496971/leftshift-method-on-biginteger-in-groovy]).
> The *leftShift()* method is however described in [Groovy's JDK
> enhancements|http://docs.groovy-lang.org/latest/html/api/org/codehaus/groovy/runtime/DefaultGroovyMethods.html#leftShift(java.lang.Number,java.lang.Number)].
> The below code fails with Groovy 3.0.4:
> {code:groovy}
> Integer i = new Integer(3)
> assert 12 == i << 2
> Number b = new BigInteger("3")
> assert 12 == b << 2
> {code}
> It throws an exception:
> {noformat}
> Caught: java.lang.UnsupportedOperationException: Cannot use leftShift() on
> this number type: java.math.BigInteger with value: 3
> java.lang.UnsupportedOperationException: Cannot use leftShift() on this
> number type: java.math.BigInteger with value: 3
> at test.run(test.groovy:5)
> {noformat}
> Consequence: you cannot use the {{<<}} operator on a _Number_ if this number
> is a _BigInteger_.
> h1. Understanding of the problem
> (I'm not familiar with the code)
> The
> [BigIntegerMath|https://github.com/groovy/groovy-core/blob/master/src/main/org/codehaus/groovy/runtime/typehandling/BigIntegerMath.java]
> class does not implement the _leftShift()_ method.
> Also, please note that JDK's
> [BigInteger|https://docs.oracle.com/javase/8/docs/api/java/math/BigInteger.html#shiftLeft-int-]
> method for this operation is *not* _leftShift()_ but _shiftLeft()_!
> h1. Workaround
> Using mixins, the below code works:
> {code:groovy}
> class EnhancedNumber {
> static def originalLeftShift =
> org.codehaus.groovy.runtime.DefaultGroovyMethods.
> static Number leftShift(Number self, Number operand) {
> self.class == BigInteger ? self.shiftLeft(operand) :
> originalLeftShift(self, operand)
> }
> }
> Number.mixin(EnhancedNumber)
> Integer i = new Integer(3)
> assert 12 == i << 2
> Number b = new BigInteger("3")
> assert 12 == b << 2
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[GitHub] [groovy] paulk-asert commented on pull request #1269: GROOVY-7971: do not save instanceof types under logical or
paulk-asert commented on pull request #1269: URL: https://github.com/apache/groovy/pull/1269#issuecomment-647243171 I have been doing a few spikes but nothing finalised. I tried using Union and Intersection types to better model the type relationships. A logical OR with instanceof on both sides for example should have different semantics to a logical AND with instanceof on both sides. And we'd need a NotType as well at some point to have better handling of not in combination with instanceof for nested cases. I can get specific parts of our type checking processing to work as I'd hope with these in place but there are numerous places in the code which make different assumptions to what the new types would need. There are places where we ignore Object for instance which isn't what we'd now need but just removing the places where we make such tests breaks other things because various assumptions flow to other parts of the code. So, as an intermediate step, I have a slightly improved mechanism that looks for any IfStatement where the test is a binary logical OR and treats that as two individual IfStatement's corresponding to each half of the OR. It then does a separate pass to get the UnionType we currently expect for the AND case. The UnionType needs to be catered for so that we have an invokeDynamic instruction produced when @CS is in place. I think it will be a few more days before I have anything ready for review. This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[jira] [Commented] (GROOVY-9519) CLIBuilder: Option with "type: Integer, defaultValue '0'" results in Boolean if default value is applied
[
https://issues.apache.org/jira/browse/GROOVY-9519?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141628#comment-17141628
]
Remko Popma commented on GROOVY-9519:
-
[~dheinric], you are right, I did miss it when you raised it here originally.
Thanks for following up!
> CLIBuilder: Option with "type: Integer, defaultValue '0'" results in Boolean
> if default value is applied
>
>
> Key: GROOVY-9519
> URL: https://issues.apache.org/jira/browse/GROOVY-9519
> Project: Groovy
> Issue Type: Bug
> Components: command line processing
>Affects Versions: 3.0.2
> Environment: WIndows/Linux
>Reporter: Dirk Heinrichs
>Assignee: Remko Popma
>Priority: Major
> Fix For: 3.0.4, 2.5.12
>
>
> When using an option with {{type: Integer}} and {{defaultValue: '0'}} the
> resulting variable is of type Boolean if the option is not specified on the
> command line. Here's a little Groovy program to demonstrate the issue:
> {code:java}
> @Grab('info.picocli:picocli-groovy:4.2.0')
> @GrabConfig(systemClassLoader=true)
> import groovy.cli.picocli.CliBuilder
> def cli = new CliBuilder(name: 'cliTest.groovy')
> cli.h(type: Boolean, longOpt: 'help', usageHelp: true, required: false, 'Show
> usage information')
> cli.i(type: Integer, longOpt: 'intTest', required: false, args: 1,
> defaultValue: '0', 'Testing integer with default value 0')
> def opts = cli.parse(args)
> opts || System.exit(1)
> if(opts.h) {
> cli.usage()
> System.exit(0)
> }
> println(opts.i.getClass())
> println(opts.i)
> // Need to explicitely convert if default value applies
> def i = (opts.i instanceof Boolean ? 0 : opts.i)
> println(i.getClass())
> println(i){code}
> Executing this w/o providing {{-i}} prints:
> {code}% ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy
> class java.lang.Boolean
> false
> class java.lang.Integer
> 0{code}
> But if {{-i}} IS provided (even with the same value as the default), it
> correctly prints:
> {code}% ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy -i 0
> class java.lang.Integer
> 0
> class java.lang.Integer
> 0{code}
> I would expect the type of {{opts.i}} to be Integer in both cases.
> NOTE: I've opened the same for PicoCLI, since I wasn't sure where it belongs.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Resolved] (GROOVY-9599) CliBuilder: Option with "type: String, defaultValue ''" (empty String) results in NullObject if default value is applied
[
https://issues.apache.org/jira/browse/GROOVY-9599?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Remko Popma resolved GROOVY-9599.
-
Resolution: Fixed
> CliBuilder: Option with "type: String, defaultValue ''" (empty String)
> results in NullObject if default value is applied
>
>
> Key: GROOVY-9599
> URL: https://issues.apache.org/jira/browse/GROOVY-9599
> Project: Groovy
> Issue Type: Bug
> Components: command line processing
>Affects Versions: 2.5.11, 3.0.4
>Reporter: Remko Popma
>Assignee: Remko Popma
>Priority: Major
> Fix For: 3.0.5, 2.5.13
>
>
> This is a follow-up ticket for GROOVY-9519:
> After checking with Groovy 3.0.4, it seems that only the "Integer with
> defaultValue '0'" case was fixed, but not "String with defaultValue ''". The
> latter became even worse. It still ignores the defaultValue, but instead of
> creating a Boolean, it now creates a NullObject. Try this code:
> {code:java}
> @Grab('info.picocli:picocli-groovy:4.3.2')
> @GrabConfig(systemClassLoader=true)
> import groovy.cli.picocli.CliBuilder
> def cli = new CliBuilder(name: 'cliTest.groovy', stopAtNonOption: false)
> cli.h(type: Boolean, longOpt: 'help', usageHelp: true, required: false, 'Show
> usage information')
> cli.a(type: String, longOpt: 'optA', required: false, args: 1, defaultValue:
> '', 'Option a (optional)')
> def opts = cli.parse(args)
> opts || System.exit(1)
> if(opts.h) {
> cli.usage()
> System.exit(0)
> }
> println(opts.a.getClass())
> if (opts.a) {
> println(opts.a)
> }
> {code}
> with Groovy 3.0.2:
> {code:java}
> % ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy
> class java.lang.Boolean
> {code}
> and with Groovy 3.0.4:
> {code:java}
> % ~/tmp/groovy-3.0.4/bin/groovy ./cliTest.groovy
> class org.codehaus.groovy.runtime.NullObject
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GROOVY-9599) CliBuilder: Option with "type: String, defaultValue ''" (empty String) results in NullObject if default value is applied
[
https://issues.apache.org/jira/browse/GROOVY-9599?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141625#comment-17141625
]
Remko Popma commented on GROOVY-9599:
-
Fix and additional test committed to the
[master|https://github.com/apache/groovy/commit/16a5c42e53c440d3675d7b7cb91113a584663b86],
[GROOVY_3_0_X|https://github.com/apache/groovy/commit/829b19c7bcc91b3aa80f16379f980dc5dcdf9c64]
and
[GROOVY_2_5_X|https://github.com/apache/groovy/commit/ede7ffca2bf151471cfc888ee41d4c4f68d3d906]
branches.
Test results look good in [CI|https://groovy.apache.org/buildstatus.html]
([master|https://ci.groovy-lang.org/viewLog.html?buildId=2154=MasterTestJdk8],
[3.0|https://ci.groovy-lang.org/viewLog.html?buildId=2151=Groovy30xTestAllJdk8],
[2.5|https://ci.groovy-lang.org/viewLog.html?buildId=2159=Groovy25xCheckJdk8])
Resolving this ticket as fixed.
> CliBuilder: Option with "type: String, defaultValue ''" (empty String)
> results in NullObject if default value is applied
>
>
> Key: GROOVY-9599
> URL: https://issues.apache.org/jira/browse/GROOVY-9599
> Project: Groovy
> Issue Type: Bug
> Components: command line processing
>Affects Versions: 2.5.11, 3.0.4
>Reporter: Remko Popma
>Assignee: Remko Popma
>Priority: Major
> Fix For: 3.0.5, 2.5.13
>
>
> This is a follow-up ticket for GROOVY-9519:
> After checking with Groovy 3.0.4, it seems that only the "Integer with
> defaultValue '0'" case was fixed, but not "String with defaultValue ''". The
> latter became even worse. It still ignores the defaultValue, but instead of
> creating a Boolean, it now creates a NullObject. Try this code:
> {code:java}
> @Grab('info.picocli:picocli-groovy:4.3.2')
> @GrabConfig(systemClassLoader=true)
> import groovy.cli.picocli.CliBuilder
> def cli = new CliBuilder(name: 'cliTest.groovy', stopAtNonOption: false)
> cli.h(type: Boolean, longOpt: 'help', usageHelp: true, required: false, 'Show
> usage information')
> cli.a(type: String, longOpt: 'optA', required: false, args: 1, defaultValue:
> '', 'Option a (optional)')
> def opts = cli.parse(args)
> opts || System.exit(1)
> if(opts.h) {
> cli.usage()
> System.exit(0)
> }
> println(opts.a.getClass())
> if (opts.a) {
> println(opts.a)
> }
> {code}
> with Groovy 3.0.2:
> {code:java}
> % ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy
> class java.lang.Boolean
> {code}
> and with Groovy 3.0.4:
> {code:java}
> % ~/tmp/groovy-3.0.4/bin/groovy ./cliTest.groovy
> class org.codehaus.groovy.runtime.NullObject
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (GROOVY-9599) CliBuilder: Option with "type: String, defaultValue ''" (empty String) results in NullObject if default value is applied
[
https://issues.apache.org/jira/browse/GROOVY-9599?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Remko Popma updated GROOVY-9599:
Affects Version/s: 2.5.11
> CliBuilder: Option with "type: String, defaultValue ''" (empty String)
> results in NullObject if default value is applied
>
>
> Key: GROOVY-9599
> URL: https://issues.apache.org/jira/browse/GROOVY-9599
> Project: Groovy
> Issue Type: Bug
> Components: command line processing
>Affects Versions: 2.5.11, 3.0.4
>Reporter: Remko Popma
>Assignee: Remko Popma
>Priority: Major
> Fix For: 3.0.5
>
>
> This is a follow-up ticket for GROOVY-9519:
> After checking with Groovy 3.0.4, it seems that only the "Integer with
> defaultValue '0'" case was fixed, but not "String with defaultValue ''". The
> latter became even worse. It still ignores the defaultValue, but instead of
> creating a Boolean, it now creates a NullObject. Try this code:
> {code:java}
> @Grab('info.picocli:picocli-groovy:4.3.2')
> @GrabConfig(systemClassLoader=true)
> import groovy.cli.picocli.CliBuilder
> def cli = new CliBuilder(name: 'cliTest.groovy', stopAtNonOption: false)
> cli.h(type: Boolean, longOpt: 'help', usageHelp: true, required: false, 'Show
> usage information')
> cli.a(type: String, longOpt: 'optA', required: false, args: 1, defaultValue:
> '', 'Option a (optional)')
> def opts = cli.parse(args)
> opts || System.exit(1)
> if(opts.h) {
> cli.usage()
> System.exit(0)
> }
> println(opts.a.getClass())
> if (opts.a) {
> println(opts.a)
> }
> {code}
> with Groovy 3.0.2:
> {code:java}
> % ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy
> class java.lang.Boolean
> {code}
> and with Groovy 3.0.4:
> {code:java}
> % ~/tmp/groovy-3.0.4/bin/groovy ./cliTest.groovy
> class org.codehaus.groovy.runtime.NullObject
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (GROOVY-9599) CliBuilder: Option with "type: String, defaultValue ''" (empty String) results in NullObject if default value is applied
[
https://issues.apache.org/jira/browse/GROOVY-9599?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Remko Popma updated GROOVY-9599:
Fix Version/s: 2.5.13
> CliBuilder: Option with "type: String, defaultValue ''" (empty String)
> results in NullObject if default value is applied
>
>
> Key: GROOVY-9599
> URL: https://issues.apache.org/jira/browse/GROOVY-9599
> Project: Groovy
> Issue Type: Bug
> Components: command line processing
>Affects Versions: 2.5.11, 3.0.4
>Reporter: Remko Popma
>Assignee: Remko Popma
>Priority: Major
> Fix For: 3.0.5, 2.5.13
>
>
> This is a follow-up ticket for GROOVY-9519:
> After checking with Groovy 3.0.4, it seems that only the "Integer with
> defaultValue '0'" case was fixed, but not "String with defaultValue ''". The
> latter became even worse. It still ignores the defaultValue, but instead of
> creating a Boolean, it now creates a NullObject. Try this code:
> {code:java}
> @Grab('info.picocli:picocli-groovy:4.3.2')
> @GrabConfig(systemClassLoader=true)
> import groovy.cli.picocli.CliBuilder
> def cli = new CliBuilder(name: 'cliTest.groovy', stopAtNonOption: false)
> cli.h(type: Boolean, longOpt: 'help', usageHelp: true, required: false, 'Show
> usage information')
> cli.a(type: String, longOpt: 'optA', required: false, args: 1, defaultValue:
> '', 'Option a (optional)')
> def opts = cli.parse(args)
> opts || System.exit(1)
> if(opts.h) {
> cli.usage()
> System.exit(0)
> }
> println(opts.a.getClass())
> if (opts.a) {
> println(opts.a)
> }
> {code}
> with Groovy 3.0.2:
> {code:java}
> % ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy
> class java.lang.Boolean
> {code}
> and with Groovy 3.0.4:
> {code:java}
> % ~/tmp/groovy-3.0.4/bin/groovy ./cliTest.groovy
> class org.codehaus.groovy.runtime.NullObject
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GROOVY-9599) CliBuilder: Option with "type: String, defaultValue ''" (empty String) results in NullObject if default value is applied
[
https://issues.apache.org/jira/browse/GROOVY-9599?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141605#comment-17141605
]
Remko Popma commented on GROOVY-9599:
-
This turned out to have a separate, very different cause from GROOVY-9519.
Given code like this:
{code:java}
def cli = new CliBuilder()
cli.s(type: String, longOpt: 'str', required: false, args: 1, defaultValue: '',
'description')
{code}
the {{groovy.cli.picocli.CliBuilder}} class constructs a
{{picocli.CommandLine.Model.OptionSpec}} from the method arguments.
These method arguments are given to the conversion logic as a {{Map}}. During this conversion, there is the following conditional that
prevents all options from getting an empty String value as their default value:
{code:java}
// Map m
if (!m.defaultValue) { //<-- (line 943) BAD: evaluates to false if value is
empty String.
m.remove('defaultValue') // don't default the picocli model to empty string
}
{code}
This works fine when no default is specified, or when the default is specified
as something that does not evaluate to {{false}} as per Groovy Truthiness, but
when an empty String is intentionally specified as the default, what actually
happens is that the default value becomes {{null}}.
The correct code should just look at the absence or presence of the
"defaultValue" key in the map:
{code:java}
if (!m.containsKey('defaultValue')) { // GROOVY-9599
m.remove('defaultValue') // don't default the picocli model to empty string
}
{code}
I will commit a fix and an additional test soon.
> CliBuilder: Option with "type: String, defaultValue ''" (empty String)
> results in NullObject if default value is applied
>
>
> Key: GROOVY-9599
> URL: https://issues.apache.org/jira/browse/GROOVY-9599
> Project: Groovy
> Issue Type: Bug
> Components: command line processing
>Affects Versions: 3.0.4
>Reporter: Remko Popma
>Assignee: Remko Popma
>Priority: Major
> Fix For: 3.0.5
>
>
> This is a follow-up ticket for GROOVY-9519:
> After checking with Groovy 3.0.4, it seems that only the "Integer with
> defaultValue '0'" case was fixed, but not "String with defaultValue ''". The
> latter became even worse. It still ignores the defaultValue, but instead of
> creating a Boolean, it now creates a NullObject. Try this code:
> {code:java}
> @Grab('info.picocli:picocli-groovy:4.3.2')
> @GrabConfig(systemClassLoader=true)
> import groovy.cli.picocli.CliBuilder
> def cli = new CliBuilder(name: 'cliTest.groovy', stopAtNonOption: false)
> cli.h(type: Boolean, longOpt: 'help', usageHelp: true, required: false, 'Show
> usage information')
> cli.a(type: String, longOpt: 'optA', required: false, args: 1, defaultValue:
> '', 'Option a (optional)')
> def opts = cli.parse(args)
> opts || System.exit(1)
> if(opts.h) {
> cli.usage()
> System.exit(0)
> }
> println(opts.a.getClass())
> if (opts.a) {
> println(opts.a)
> }
> {code}
> with Groovy 3.0.2:
> {code:java}
> % ~/tmp/groovy-3.0.2/bin/groovy ./cliTest.groovy
> class java.lang.Boolean
> {code}
> and with Groovy 3.0.4:
> {code:java}
> % ~/tmp/groovy-3.0.4/bin/groovy ./cliTest.groovy
> class org.codehaus.groovy.runtime.NullObject
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Resolved] (GROOVY-9592) Replace "groovy.generate.stub.in.memory" with "groovy.mem.stub"
[
https://issues.apache.org/jira/browse/GROOVY-9592?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Daniel Sun resolved GROOVY-9592.
Resolution: Fixed
> Replace "groovy.generate.stub.in.memory" with "groovy.mem.stub"
> ---
>
> Key: GROOVY-9592
> URL: https://issues.apache.org/jira/browse/GROOVY-9592
> Project: Groovy
> Issue Type: Improvement
>Reporter: Daniel Sun
>Assignee: Daniel Sun
>Priority: Major
> Fix For: 4.0.0-alpha-1, 3.0.5
>
> Time Spent: 1h 10m
> Remaining Estimate: 0h
>
> The JVM option {{groovy.generate.stub.in.memory}} is too long for users to
> remember, replace it with {{groovy.mem.stub}}
> For Groovy 3, {{groovy.generate.stub.in.memory}} will be deprecated.
> Since Groovy 4, {{groovy.generate.stub.in.memory}} will be removed.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[GitHub] [groovy] danielsun1106 merged pull request #1278: GROOVY-9592: Replace "groovy.generate.stub.in.memory" with "groovy.me…
danielsun1106 merged pull request #1278: URL: https://github.com/apache/groovy/pull/1278 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[GitHub] [groovy] danielsun1106 commented on pull request #1269: GROOVY-7971: do not save instanceof types under logical or
danielsun1106 commented on pull request #1269: URL: https://github.com/apache/groovy/pull/1269#issuecomment-647197753 @paulk-asert Any further comments on the PR? This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[jira] [Resolved] (GROOVY-8100) MarkupBuilder cant yield before first tag
[
https://issues.apache.org/jira/browse/GROOVY-8100?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Daniel Sun resolved GROOVY-8100.
Fix Version/s: 3.0.5
4.0.0-alpha-1
Assignee: Daniel Sun
Resolution: Fixed
The proposed PR has been merged. Thanks for your contribution!
> MarkupBuilder cant yield before first tag
> -
>
> Key: GROOVY-8100
> URL: https://issues.apache.org/jira/browse/GROOVY-8100
> Project: Groovy
> Issue Type: Bug
> Components: XML Processing
>Affects Versions: 2.4.8, 3.0.0-rc-2
> Environment: Oracle Java 1.8.0_112, Windows 10 x64
>Reporter: Jochen Eddelbuettel
>Assignee: Daniel Sun
>Priority: Major
> Fix For: 4.0.0-alpha-1, 3.0.5
>
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> Trying to produce an HTML5 compliant !DOCTYPE before starting with the
> HTML-Elements. Without setting the private field 'state' to 2, the output
> does not show the yielded output.
> {code}
> def html = new groovy.xml.MarkupBuilder(new PrintWriter(System.out))
> //html.state = 2
> html.mkp.yieldUnescaped("\n")
> html.h1("Header")
> {code}
> I suggest fixing this by changing the yield-Method to not check the state
> before doing the output, so that we see some output when the initial state is
> still 0:
> {code}
> void yield(String value, boolean escaping) {
> if (state == 1) {
> state = 2;
> this.nodeIsEmpty = false;
> out.print(">");
> }
> // -- remove -- if (state == 2 || state == 3) {
> out.print(escaping ? escapeElementContent(value) : value);
> // -- remove -- }
> }
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[GitHub] [groovy] danielsun1106 commented on pull request #1284: [GROOVY-8100] MarkupBuilder - yield before any markup (state == 0)
danielsun1106 commented on pull request #1284: URL: https://github.com/apache/groovy/pull/1284#issuecomment-647197082 Merged. Thanks. This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[GitHub] [groovy] danielsun1106 merged pull request #1284: [GROOVY-8100] MarkupBuilder - yield before any markup (state == 0)
danielsun1106 merged pull request #1284: URL: https://github.com/apache/groovy/pull/1284 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[jira] [Resolved] (GROOVY-9344) CCE in closure shared variable assignment
[
https://issues.apache.org/jira/browse/GROOVY-9344?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Daniel Sun resolved GROOVY-9344.
Fix Version/s: 3.0.5
4.0.0-alpha-1
Resolution: Fixed
> CCE in closure shared variable assignment
> -
>
> Key: GROOVY-9344
> URL: https://issues.apache.org/jira/browse/GROOVY-9344
> Project: Groovy
> Issue Type: Bug
> Components: Static compilation, Static Type Checker
>Affects Versions: 2.5.8
>Reporter: Daniil Ovchinnikov
>Assignee: Eric Milles
>Priority: Major
> Fix For: 4.0.0-alpha-1, 3.0.5
>
> Time Spent: 1h
> Remaining Estimate: 0h
>
> {code}
> class A {}
> class B {}
> @groovy.transform.CompileStatic
> def cs() {
> def var
> var = new A()
> def c = {
> var = new B() // Cannot cast object 'B@4e234c52' with class 'B' to
> class 'A'
> }
> c()
> var.toString()
> }
> assert cs() != null
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Resolved] (GROOVY-9516) Static compiler infers wrong type for implicitly shared variables
[
https://issues.apache.org/jira/browse/GROOVY-9516?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Daniel Sun resolved GROOVY-9516.
Fix Version/s: 3.0.5
4.0.0-alpha-1
Resolution: Fixed
> Static compiler infers wrong type for implicitly shared variables
> -
>
> Key: GROOVY-9516
> URL: https://issues.apache.org/jira/browse/GROOVY-9516
> Project: Groovy
> Issue Type: Bug
> Components: Static compilation
>Affects Versions: 3.0.0, 2.5.11
>Reporter: Konstantin Nisht
>Assignee: Eric Milles
>Priority: Major
> Fix For: 4.0.0-alpha-1, 3.0.5
>
>
> Consider the following code
> {code:groovy}
> class A {}
> class B extends A{ def b() {}}
> class C extends A{}
> @CompileStatic
> static foo() {
> def x = new B()
> ({ x = new C() })()
> def z = x
> z.b()
> }
> {code}
> Here type of {{z}} in the last line of the method is deduced to be {{B}}.
> Attempt to execute {{z.b()}} leads to cast exception, which is unexpected
> under static compilation.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[GitHub] [groovy] danielsun1106 commented on pull request #1279: GROOVY-9344, GROOVY-9516: track shared variable assignment in closure like conditional branch
danielsun1106 commented on pull request #1279: URL: https://github.com/apache/groovy/pull/1279#issuecomment-647196334 Merged. Thanks. This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[GitHub] [groovy] danielsun1106 merged pull request #1279: GROOVY-9344, GROOVY-9516: track shared variable assignment in closure like conditional branch
danielsun1106 merged pull request #1279: URL: https://github.com/apache/groovy/pull/1279 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[jira] [Resolved] (GROOVY-9591) Compiler error for use of variable expression within tap closure
[
https://issues.apache.org/jira/browse/GROOVY-9591?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Daniel Sun resolved GROOVY-9591.
Fix Version/s: 3.0.5
4.0.0-alpha-1
Resolution: Fixed
> Compiler error for use of variable expression within tap closure
>
>
> Key: GROOVY-9591
> URL: https://issues.apache.org/jira/browse/GROOVY-9591
> Project: Groovy
> Issue Type: Bug
>Affects Versions: 3.0.4, 2.5.12
>Reporter: Eric Milles
>Assignee: Eric Milles
>Priority: Major
> Fix For: 4.0.0-alpha-1, 3.0.5
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> Consider the following:
> {code:groovy}
> @groovy.transform.ToString
> class A {
> A(a) {}
> def b
> }
> class C {
> C() {
> this(new A(null).tap { b = 42 })
> }
> C(x) {
> print x
> }
> }
> new C() // should output "A(42)"
> {code}
> A has no default constructor, so properties are initialized in tap closure.
> Because the "tap" expression is passed as argument to special constructor
> call, extra static checking is performed and the error "Apparent variable 'b'
> was found in a static scope but doesn't refer to a local variable, static
> field or class." is emitted.
> Creation of A cannot be moved to a local variable because special constructor
> call must be the first statement/expression in the constructor body.
> This was discovered while investigating variations of GROOVY-8327,
> GROOVY-8389, and GROOVY-9587.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[GitHub] [groovy] danielsun1106 commented on pull request #1276: GROOVY-9591: no error for dynamic variable in closure in static context
danielsun1106 commented on pull request #1276: URL: https://github.com/apache/groovy/pull/1276#issuecomment-647196217 Merged. Thanks. This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[GitHub] [groovy] danielsun1106 merged pull request #1276: GROOVY-9591: no error for dynamic variable in closure in static context
danielsun1106 merged pull request #1276: URL: https://github.com/apache/groovy/pull/1276 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[jira] [Created] (GROOVY-9605) leftShift operator does not work on BigInteger (throws UnsupportedOperationException)
Bertrand Martin created GROOVY-9605:
---
Summary: leftShift operator does not work on BigInteger (throws
UnsupportedOperationException)
Key: GROOVY-9605
URL: https://issues.apache.org/jira/browse/GROOVY-9605
Project: Groovy
Issue Type: Bug
Components: groovy-jdk
Affects Versions: 3.0.4
Environment: Groovy 3.0.4 Windows and Linux
Reporter: Bertrand Martin
h1. Problem
The _leftShift_ operator ({{<<}}) does not work on *BigInteger* (see
[StackOverflow
question|https://stackoverflow.com/questions/62496971/leftshift-method-on-biginteger-in-groovy]).
The *leftShift()* method is however described in [Groovy's JDK
enhancements|http://docs.groovy-lang.org/latest/html/api/org/codehaus/groovy/runtime/DefaultGroovyMethods.html#leftShift(java.lang.Number,java.lang.Number)].
The below code fails with Groovy 3.0.4:
{code:groovy}
Integer i = new Integer(3)
assert 12 == i << 2
Number b = new BigInteger("3")
assert 12 == b << 2
{code}
It throws an exception:
{noformat}
Caught: java.lang.UnsupportedOperationException: Cannot use leftShift() on this
number type: java.math.BigInteger with value: 3
java.lang.UnsupportedOperationException: Cannot use leftShift() on this number
type: java.math.BigInteger with value: 3
at test.run(test.groovy:5)
{noformat}
Consequence: you cannot use the {{<<}} operator on a _Number_ if this number is
a _BigInteger_.
h1. Understanding of the problem
(I'm not familiar with the code)
The
[BigIntegerMath|https://github.com/groovy/groovy-core/blob/master/src/main/org/codehaus/groovy/runtime/typehandling/BigIntegerMath.java]
class does not implement the _leftShift()_ method.
Also, please note that JDK's
[BigInteger|https://docs.oracle.com/javase/8/docs/api/java/math/BigInteger.html#shiftLeft-int-]
method for this operation is *not* _leftShift()_ but _shiftLeft()_!
h1. Workaround
Using mixins, the below code works:
{code:groovy}
class EnhancedNumber {
static def originalLeftShift =
org.codehaus.groovy.runtime.DefaultGroovyMethods.
static Number leftShift(Number self, Number operand) {
self.class == BigInteger ? self.shiftLeft(operand) :
originalLeftShift(self, operand)
}
}
Number.mixin(EnhancedNumber)
Integer i = new Integer(3)
assert 12 == i << 2
Number b = new BigInteger("3")
assert 12 == b << 2
{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[GitHub] [groovy] eric-milles opened a new pull request #1285: GROOVY-9604: STC: add support for accessing additional closure methods and properties
eric-milles opened a new pull request #1285: URL: https://github.com/apache/groovy/pull/1285 https://issues.apache.org/jira/browse/GROOVY-9604 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
[jira] [Created] (GROOVY-9604) STC: method or property missing errors for Closure's getResolveStrategy(), etc.
Eric Milles created GROOVY-9604:
---
Summary: STC: method or property missing errors for Closure's
getResolveStrategy(), etc.
Key: GROOVY-9604
URL: https://issues.apache.org/jira/browse/GROOVY-9604
Project: Groovy
Issue Type: Bug
Reporter: Eric Milles
Assignee: Eric Milles
Consider the following:
{code:groovy}
@groovy.transform.CompileStatic
class C {
void m() {
{ ->
getResolveStrategy() // and other Closure methods/properties except
"owner", "delegate" and "thisObject"
}()
}
}
{code}
Trying to access the enclosing closure's metadata results in STC compiler
errors.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GROOVY-8845) @DelegatesTo works only for the first vararg
[
https://issues.apache.org/jira/browse/GROOVY-8845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17141508#comment-17141508
]
Eric Milles commented on GROOVY-8845:
-
GROOVY-6022 seems to suggest that a compiler error should be emitted for
@DelegatesTo applied to array (aka varargs parameter).
> @DelegatesTo works only for the first vararg
>
>
> Key: GROOVY-8845
> URL: https://issues.apache.org/jira/browse/GROOVY-8845
> Project: Groovy
> Issue Type: Bug
> Components: Static Type Checker
>Affects Versions: 2.5.3
>Reporter: Daniil Ovchinnikov
>Priority: Major
>
> {code:groovy}
> def md(@DelegatesTo.Target Object target, @DelegatesTo(strategy =
> Closure.DELEGATE_FIRST) Closure... arg) {
> for (Closure a : arg) {
> a.delegate = target
> a.resolveStrategy = Closure.DELEGATE_FIRST
> a()
> }
> }
> class D {
> def foo() { 42 }
> }
> @groovy.transform.CompileStatic
> def test() {
> md(
> new D(),
> { print(foo()) },
> // { print(foo()) }, // [Static type checking] - Cannot find matching
> method ConsoleScript15#foo()
> // { print(foo()) }, // [Static type checking] - Cannot find matching
> method ConsoleScript15#foo()
> )
> }
> test()
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
