subject:"\[royale\-compiler\] 01\/03\: fix bracket access of XML\/XMLList"

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-07 Thread Harbs

I don’t have time right now, but I added an issue so we shouldn’t forget about 
it…

> On Oct 7, 2018, at 5:38 PM, Alex Harui  wrote:
> 
> Then add tests for string expressions.  I thought I'd already done that.  I 
> would expect that Tour De Flex is now broken.  I've been trying to fix the 
> compile to get TDF to work.
> 
> IExpressionNode.resolveType() should return the type of a function.  See the 
> Javadoc for it.
> 
> -Alex
> 
> On 10/7/18, 4:59 AM, "Harbs"  > wrote:
> 
>I just pushed a change to the compiler to swap around the check to check 
> if the type is a string instead of testing if it’s not a number. This fixes 
> number expressions, but likely breaks string expressions. I don’t know how to 
> get the evaluated type of the whole expression.
> 
>At least all current tests now pass…
> 
>Harbs
> 
>> On Oct 3, 2018, at 11:10 AM, Harbs  wrote:
>> 
>> OK.
>> 
>> I added some test cases for XMLList indexed access and assignment. One of 
>> those tests passed before your change and is now failing.
>> 
>> I think this is a summary of the cases we need to deal with (and we probably 
>> need more test cases…)
>> 
>> Given:
>> var xml:XML = 
>> var baz:XMLList = foo.baz;
>> 
>> baz = foo["baz”];
>> Should become:
>> baz = foo.child("baz”);
>> 
>> baz = foo[foo.length()-1];
>> Should remain unchanged:
>> baz = foo[foo.length()-1];
>> 
>> foo["baz"] = ;
>> Now becomes:
>> foo.child("baz") = new XML("> But that’s wrong. It should be:
>> foo.setChild("baz",new XML("> 
>> foo[0] = ;
>> Should remain unchanged:
>> foo[0] = new XML("");
>> 
>> I don’t understand the logic in the compiler well enough to really know how 
>> to go about fixing these. Numbers in brackets need to be handled differently 
>> that strings and left hand of assignment should be different than right-hand.
>> 
>> I hope this info is helpful…
>> 
>> Harbs
>> 
>>> On Oct 2, 2018, at 9:25 PM, Alex Harui >>  >> >> wrote:
>>> 
>>> The recommended practice is to add a test case and make sure it fails 
>>> appropriately.  Then set @ignore on it before committing so it doesn't 
>>> break the build. Whoever works on the fix can remove the @ignore locally 
>>> when working on a fix
>>> 
>>> I thought there were already test cases for handling number/int 
>>> appropriately, but maybe not.
>>> 
>>> It may be that there is additional code and tests needed to handle setting 
>>> a value via bracket access as well.  Feel free to add test cases for that 
>>> and make the necessary changes in compiler.
>>> 
>>> -Alex
>>> 
>>> On 10/2/18, 11:16 AM, "Harbs" >>  >> >> wrote:
>>> 
>>>   I guess I can add a test case, but that’ll make the build fail.
>>> 
>>>   I’m not sure what to propose for a fix though.
>>> 
>>>   Any indexed (i.e. Number, int or uint type) access should always remain 
>>> bracketed access. .child() seems wrong for assignment in any case. If 
>>> anything it should be .setChild(), but I’m not sure how well that works for 
>>> XMLList. It only does anything for single-item XMLList instances (which 
>>> behave like XML instances).
>>> 
>>>   Thanks,
>>>   Harbs
>>> 
 On Oct 2, 2018, at 7:22 PM, Alex Harui >>>  >> wrote:
 
 Add a test case to the test suite, propose the fix or ask for help on it.
 
 -Alex
 
 On 10/2/18, 1:03 AM, "Harbs" >>>  >   On Sep 28, 2018, at 6:34 AM, aha...@apache.org  
> > wrote:
> 
> This is an automated email from the ASF dual-hosted git repository.
> 
> aharui pushed a commit to branch develop
> in repository 
> https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitbox.apache.org%2Frepos%2Fasf%2Froyale-compiler.gitdata=02%7C01%7Caharui%40adobe.com%7C21d3c40fc0b8481819e608d62c4c4787%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636745103460602827sdata=pUyokCJAe7aUUdpXOd72VsvnnD8gKU4zw%2FmFd1AG6Ss%3Dreserved=0
>  
>

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-07 Thread Alex Harui

Then add tests for string expressions.  I thought I'd already done that.  I 
would expect that Tour De Flex is now broken.  I've been trying to fix the 
compile to get TDF to work.

IExpressionNode.resolveType() should return the type of a function.  See the 
Javadoc for it.

-Alex

On 10/7/18, 4:59 AM, "Harbs"  wrote:

I just pushed a change to the compiler to swap around the check to check if 
the type is a string instead of testing if it’s not a number. This fixes number 
expressions, but likely breaks string expressions. I don’t know how to get the 
evaluated type of the whole expression.

At least all current tests now pass…

Harbs

> On Oct 3, 2018, at 11:10 AM, Harbs  wrote:
> 
> OK.
> 
> I added some test cases for XMLList indexed access and assignment. One of 
those tests passed before your change and is now failing.
> 
> I think this is a summary of the cases we need to deal with (and we 
probably need more test cases…)
> 
> Given:
> var xml:XML = 
> var baz:XMLList = foo.baz;
> 
> baz = foo["baz”];
> Should become:
> baz = foo.child("baz”);
> 
> baz = foo[foo.length()-1];
> Should remain unchanged:
> baz = foo[foo.length()-1];
> 
> foo["baz"] = ;
> Now becomes:
> foo.child("baz") = new XML(" But that’s wrong. It should be:
> foo.setChild("baz",new XML(" 
> foo[0] = ;
> Should remain unchanged:
> foo[0] = new XML("");
> 
> I don’t understand the logic in the compiler well enough to really know 
how to go about fixing these. Numbers in brackets need to be handled 
differently that strings and left hand of assignment should be different than 
right-hand.
> 
> I hope this info is helpful…
> 
> Harbs
> 
>> On Oct 2, 2018, at 9:25 PM, Alex Harui mailto:aha...@adobe.com.INVALID>> wrote:
>> 
>> The recommended practice is to add a test case and make sure it fails 
appropriately.  Then set @ignore on it before committing so it doesn't break 
the build.  Whoever works on the fix can remove the @ignore locally when 
working on a fix
>> 
>> I thought there were already test cases for handling number/int 
appropriately, but maybe not.
>> 
>> It may be that there is additional code and tests needed to handle 
setting a value via bracket access as well.  Feel free to add test cases for 
that and make the necessary changes in compiler.
>> 
>> -Alex
>> 
>> On 10/2/18, 11:16 AM, "Harbs" mailto:harbs.li...@gmail.com>> wrote:
>> 
>>I guess I can add a test case, but that’ll make the build fail.
>> 
>>I’m not sure what to propose for a fix though.
>> 
>>Any indexed (i.e. Number, int or uint type) access should always 
remain bracketed access. .child() seems wrong for assignment in any case. If 
anything it should be .setChild(), but I’m not sure how well that works for 
XMLList. It only does anything for single-item XMLList instances (which behave 
like XML instances).
>> 
>>Thanks,
>>Harbs
>> 
>>> On Oct 2, 2018, at 7:22 PM, Alex Harui mailto:aha...@adobe.com.INVALID>> wrote:
>>> 
>>> Add a test case to the test suite, propose the fix or ask for help on 
it.
>>> 
>>> -Alex
>>> 
>>> On 10/2/18, 1:03 AM, "Harbs" mailto:harbs.li...@gmail.com> >> wrote:
>>> 
>>>   I don’t know the reason for this change, but this broke bracket 
access in XMLList.
>>> 
>>>   myList[myList.length()] = xml should be output unchanged.
>>> 
>>>   It’s now output as:
>>> 
>>>   myList.child(myList.length()) = xml
>>> 
>>>   That’s wrong…
>>> 
>>>   Harbs
>>> 
 On Sep 28, 2018, at 6:34 AM, aha...@apache.org 
 wrote:
 
 This is an automated email from the ASF dual-hosted git repository.
 
 aharui pushed a commit to branch develop
 in repository 
https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitbox.apache.org%2Frepos%2Fasf%2Froyale-compiler.gitdata=02%7C01%7Caharui%40adobe.com%7C21d3c40fc0b8481819e608d62c4c4787%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636745103460602827sdata=pUyokCJAe7aUUdpXOd72VsvnnD8gKU4zw%2FmFd1AG6Ss%3Dreserved=0

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-07 Thread Harbs

I just pushed a change to the compiler to swap around the check to check if the 
type is a string instead of testing if it’s not a number. This fixes number 
expressions, but likely breaks string expressions. I don’t know how to get the 
evaluated type of the whole expression.

At least all current tests now pass…

Harbs

> On Oct 3, 2018, at 11:10 AM, Harbs  wrote:
> 
> OK.
> 
> I added some test cases for XMLList indexed access and assignment. One of 
> those tests passed before your change and is now failing.
> 
> I think this is a summary of the cases we need to deal with (and we probably 
> need more test cases…)
> 
> Given:
> var xml:XML = 
> var baz:XMLList = foo.baz;
> 
> baz = foo["baz”];
> Should become:
> baz = foo.child("baz”);
> 
> baz = foo[foo.length()-1];
> Should remain unchanged:
> baz = foo[foo.length()-1];
> 
> foo["baz"] = ;
> Now becomes:
> foo.child("baz") = new XML(" But that’s wrong. It should be:
> foo.setChild("baz",new XML(" 
> foo[0] = ;
> Should remain unchanged:
> foo[0] = new XML("");
> 
> I don’t understand the logic in the compiler well enough to really know how 
> to go about fixing these. Numbers in brackets need to be handled differently 
> that strings and left hand of assignment should be different than right-hand.
> 
> I hope this info is helpful…
> 
> Harbs
> 
>> On Oct 2, 2018, at 9:25 PM, Alex Harui > > wrote:
>> 
>> The recommended practice is to add a test case and make sure it fails 
>> appropriately.  Then set @ignore on it before committing so it doesn't break 
>> the build.  Whoever works on the fix can remove the @ignore locally when 
>> working on a fix
>> 
>> I thought there were already test cases for handling number/int 
>> appropriately, but maybe not.
>> 
>> It may be that there is additional code and tests needed to handle setting a 
>> value via bracket access as well.  Feel free to add test cases for that and 
>> make the necessary changes in compiler.
>> 
>> -Alex
>> 
>> On 10/2/18, 11:16 AM, "Harbs" > > wrote:
>> 
>>I guess I can add a test case, but that’ll make the build fail.
>> 
>>I’m not sure what to propose for a fix though.
>> 
>>Any indexed (i.e. Number, int or uint type) access should always remain 
>> bracketed access. .child() seems wrong for assignment in any case. If 
>> anything it should be .setChild(), but I’m not sure how well that works for 
>> XMLList. It only does anything for single-item XMLList instances (which 
>> behave like XML instances).
>> 
>>Thanks,
>>Harbs
>> 
>>> On Oct 2, 2018, at 7:22 PM, Alex Harui >> > wrote:
>>> 
>>> Add a test case to the test suite, propose the fix or ask for help on it.
>>> 
>>> -Alex
>>> 
>>> On 10/2/18, 1:03 AM, "Harbs" >>  >> >> wrote:
>>> 
>>>   I don’t know the reason for this change, but this broke bracket access in 
>>> XMLList.
>>> 
>>>   myList[myList.length()] = xml should be output unchanged.
>>> 
>>>   It’s now output as:
>>> 
>>>   myList.child(myList.length()) = xml
>>> 
>>>   That’s wrong…
>>> 
>>>   Harbs
>>> 
 On Sep 28, 2018, at 6:34 AM, aha...@apache.org  
 wrote:
 
 This is an automated email from the ASF dual-hosted git repository.
 
 aharui pushed a commit to branch develop
 in repository 
 https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitbox.apache.org%2Frepos%2Fasf%2Froyale-compiler.gitdata=02%7C01%7Caharui%40adobe.com%7Cd1f87f436fbd44ebc97508d628933155%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636741009996972974sdata=59bTRbYx0R8V%2FOtR3BTuTFW3u1oxfMOca8yJNUaZy3w%3Dreserved=0
  
 
  
 >
 
 commit 86bfe0a6ef8789e8ce065c856ce6f888f7562776
 Author: Alex Harui mailto:aha...@apache.org>>
 AuthorDate: Tue Sep 25 12:19:02 2018 -0700
 
  fix bracket access of XML/XMLList
 ---

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-03 Thread Harbs

OK.

I added some test cases for XMLList indexed access and assignment. One of those 
tests passed before your change and is now failing.

I think this is a summary of the cases we need to deal with (and we probably 
need more test cases…)

Given:
var xml:XML = 
var baz:XMLList = foo.baz;

baz = foo["baz”];
Should become:
baz = foo.child("baz”);

baz = foo[foo.length()-1];
Should remain unchanged:
baz = foo[foo.length()-1];

foo["baz"] = ;
Now becomes:
foo.child("baz") = new XML(";
Should remain unchanged:
foo[0] = new XML("");

I don’t understand the logic in the compiler well enough to really know how to 
go about fixing these. Numbers in brackets need to be handled differently that 
strings and left hand of assignment should be different than right-hand.

I hope this info is helpful…

Harbs

> On Oct 2, 2018, at 9:25 PM, Alex Harui  wrote:
> 
> The recommended practice is to add a test case and make sure it fails 
> appropriately.  Then set @ignore on it before committing so it doesn't break 
> the build.  Whoever works on the fix can remove the @ignore locally when 
> working on a fix
> 
> I thought there were already test cases for handling number/int 
> appropriately, but maybe not.
> 
> It may be that there is additional code and tests needed to handle setting a 
> value via bracket access as well.  Feel free to add test cases for that and 
> make the necessary changes in compiler.
> 
> -Alex
> 
> On 10/2/18, 11:16 AM, "Harbs"  > wrote:
> 
>I guess I can add a test case, but that’ll make the build fail.
> 
>I’m not sure what to propose for a fix though.
> 
>Any indexed (i.e. Number, int or uint type) access should always remain 
> bracketed access. .child() seems wrong for assignment in any case. If 
> anything it should be .setChild(), but I’m not sure how well that works for 
> XMLList. It only does anything for single-item XMLList instances (which 
> behave like XML instances).
> 
>Thanks,
>Harbs
> 
>> On Oct 2, 2018, at 7:22 PM, Alex Harui  wrote:
>> 
>> Add a test case to the test suite, propose the fix or ask for help on it.
>> 
>> -Alex
>> 
>> On 10/2/18, 1:03 AM, "Harbs" >  > >> wrote:
>> 
>>   I don’t know the reason for this change, but this broke bracket access in 
>> XMLList.
>> 
>>   myList[myList.length()] = xml should be output unchanged.
>> 
>>   It’s now output as:
>> 
>>   myList.child(myList.length()) = xml
>> 
>>   That’s wrong…
>> 
>>   Harbs
>> 
>>> On Sep 28, 2018, at 6:34 AM, aha...@apache.org  
>>> wrote:
>>> 
>>> This is an automated email from the ASF dual-hosted git repository.
>>> 
>>> aharui pushed a commit to branch develop
>>> in repository 
>>> https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitbox.apache.org%2Frepos%2Fasf%2Froyale-compiler.gitdata=02%7C01%7Caharui%40adobe.com%7Cd1f87f436fbd44ebc97508d628933155%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636741009996972974sdata=59bTRbYx0R8V%2FOtR3BTuTFW3u1oxfMOca8yJNUaZy3w%3Dreserved=0
>>>  
>>> 
>>>  
>>> >>  
>>> >
>>> 
>>> commit 86bfe0a6ef8789e8ce065c856ce6f888f7562776
>>> Author: Alex Harui 
>>> AuthorDate: Tue Sep 25 12:19:02 2018 -0700
>>> 
>>>  fix bracket access of XML/XMLList
>>> ---
>>> .../codegen/js/jx/DynamicAccessEmitter.java| 34 
>>> +-
>>> .../codegen/js/royale/TestRoyaleGlobalClasses.java | 30 ++-
>>> 2 files changed, 62 insertions(+), 2 deletions(-)
>>> 
>>> diff --git 
>>> a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>>>  
>>> b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>>> index ae977f9..0d94fac 100644
>>> --- 
>>> a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>>> +++ 
>>>

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-02 Thread Alex Harui

The recommended practice is to add a test case and make sure it fails 
appropriately.  Then set @ignore on it before committing so it doesn't break 
the build.  Whoever works on the fix can remove the @ignore locally when 
working on a fix

I thought there were already test cases for handling number/int appropriately, 
but maybe not.

It may be that there is additional code and tests needed to handle setting a 
value via bracket access as well.  Feel free to add test cases for that and 
make the necessary changes in compiler.

-Alex

On 10/2/18, 11:16 AM, "Harbs"  wrote:

I guess I can add a test case, but that’ll make the build fail.

I’m not sure what to propose for a fix though.

Any indexed (i.e. Number, int or uint type) access should always remain 
bracketed access. .child() seems wrong for assignment in any case. If anything 
it should be .setChild(), but I’m not sure how well that works for XMLList. It 
only does anything for single-item XMLList instances (which behave like XML 
instances).

Thanks,
Harbs

> On Oct 2, 2018, at 7:22 PM, Alex Harui  wrote:
> 
> Add a test case to the test suite, propose the fix or ask for help on it.
> 
> -Alex
> 
> On 10/2/18, 1:03 AM, "Harbs" mailto:harbs.li...@gmail.com>> wrote:
> 
>I don’t know the reason for this change, but this broke bracket access 
in XMLList.
> 
>myList[myList.length()] = xml should be output unchanged.
> 
>It’s now output as:
> 
>myList.child(myList.length()) = xml
> 
>That’s wrong…
> 
>Harbs
> 
>> On Sep 28, 2018, at 6:34 AM, aha...@apache.org wrote:
>> 
>> This is an automated email from the ASF dual-hosted git repository.
>> 
>> aharui pushed a commit to branch develop
>> in repository 
https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitbox.apache.org%2Frepos%2Fasf%2Froyale-compiler.gitdata=02%7C01%7Caharui%40adobe.com%7Cd1f87f436fbd44ebc97508d628933155%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636741009996972974sdata=59bTRbYx0R8V%2FOtR3BTuTFW3u1oxfMOca8yJNUaZy3w%3Dreserved=0
 

>> 
>> commit 86bfe0a6ef8789e8ce065c856ce6f888f7562776
>> Author: Alex Harui 
>> AuthorDate: Tue Sep 25 12:19:02 2018 -0700
>> 
>>   fix bracket access of XML/XMLList
>> ---
>> .../codegen/js/jx/DynamicAccessEmitter.java| 34 
+-
>> .../codegen/js/royale/TestRoyaleGlobalClasses.java | 30 
++-
>> 2 files changed, 62 insertions(+), 2 deletions(-)
>> 
>> diff --git 
a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
 
b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>> index ae977f9..0d94fac 100644
>> --- 
a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>> +++ 
b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>> @@ -21,8 +21,12 @@ package 
org.apache.royale.compiler.internal.codegen.js.jx;
>> 
>> import org.apache.royale.compiler.codegen.ISubEmitter;
>> import org.apache.royale.compiler.codegen.js.IJSEmitter;
>> +import org.apache.royale.compiler.definitions.ITypeDefinition;
>> import org.apache.royale.compiler.internal.codegen.as.ASEmitterTokens;
>> import org.apache.royale.compiler.internal.codegen.js.JSSubEmitter;
>> +import 
org.apache.royale.compiler.internal.codegen.js.royale.JSRoyaleEmitter;
>> +import 
org.apache.royale.compiler.internal.tree.as.MemberAccessExpressionNode;
>> +import org.apache.royale.compiler.internal.tree.as.NumericLiteralNode;
>> import org.apache.royale.compiler.tree.ASTNodeID;
>> import org.apache.royale.compiler.tree.as.IDynamicAccessNode;
>> import org.apache.royale.compiler.tree.as.IExpressionNode;
>> @@ -43,11 +47,39 @@ public class DynamicAccessEmitter extends 
JSSubEmitter implements
>>if (leftOperandNode.getNodeID() == ASTNodeID.Op_AtID)
>>  return;
>> 
>> +IExpressionNode rightOperandNode = node.getRightOperandNode();
>> +IJSEmitter ijs = getEmitter();
>> +JSRoyaleEmitter fjs = (ijs instanceof JSRoyaleEmitter) ? 
>> +
(JSRoyaleEmitter)ijs : null;
>> +if (fjs != null)
>> +{
>> +boolean isXML = false;
>> +if (leftOperandNode instanceof 
MemberAccessExpressionNode)
>> +

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-02 Thread Harbs

I guess I can add a test case, but that’ll make the build fail.

I’m not sure what to propose for a fix though.

Any indexed (i.e. Number, int or uint type) access should always remain 
bracketed access. .child() seems wrong for assignment in any case. If anything 
it should be .setChild(), but I’m not sure how well that works for XMLList. It 
only does anything for single-item XMLList instances (which behave like XML 
instances).

Thanks,
Harbs

> On Oct 2, 2018, at 7:22 PM, Alex Harui  wrote:
> 
> Add a test case to the test suite, propose the fix or ask for help on it.
> 
> -Alex
> 
> On 10/2/18, 1:03 AM, "Harbs"  > wrote:
> 
>I don’t know the reason for this change, but this broke bracket access in 
> XMLList.
> 
>myList[myList.length()] = xml should be output unchanged.
> 
>It’s now output as:
> 
>myList.child(myList.length()) = xml
> 
>That’s wrong…
> 
>Harbs
> 
>> On Sep 28, 2018, at 6:34 AM, aha...@apache.org wrote:
>> 
>> This is an automated email from the ASF dual-hosted git repository.
>> 
>> aharui pushed a commit to branch develop
>> in repository 
>> https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitbox.apache.org%2Frepos%2Fasf%2Froyale-compiler.gitdata=02%7C01%7Caharui%40adobe.com%7Cea3bf6b27b1d4c681a5d08d6283d7b9a%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636740641870109874sdata=6nGbwmtGYJKnYRUUPdFKixN%2BVUpEG0Ri9VY8Lp8Ib2k%3Dreserved=0
>>  
>> 
>> 
>> commit 86bfe0a6ef8789e8ce065c856ce6f888f7562776
>> Author: Alex Harui 
>> AuthorDate: Tue Sep 25 12:19:02 2018 -0700
>> 
>>   fix bracket access of XML/XMLList
>> ---
>> .../codegen/js/jx/DynamicAccessEmitter.java| 34 
>> +-
>> .../codegen/js/royale/TestRoyaleGlobalClasses.java | 30 ++-
>> 2 files changed, 62 insertions(+), 2 deletions(-)
>> 
>> diff --git 
>> a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>>  
>> b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>> index ae977f9..0d94fac 100644
>> --- 
>> a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>> +++ 
>> b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>> @@ -21,8 +21,12 @@ package org.apache.royale.compiler.internal.codegen.js.jx;
>> 
>> import org.apache.royale.compiler.codegen.ISubEmitter;
>> import org.apache.royale.compiler.codegen.js.IJSEmitter;
>> +import org.apache.royale.compiler.definitions.ITypeDefinition;
>> import org.apache.royale.compiler.internal.codegen.as.ASEmitterTokens;
>> import org.apache.royale.compiler.internal.codegen.js.JSSubEmitter;
>> +import 
>> org.apache.royale.compiler.internal.codegen.js.royale.JSRoyaleEmitter;
>> +import 
>> org.apache.royale.compiler.internal.tree.as.MemberAccessExpressionNode;
>> +import org.apache.royale.compiler.internal.tree.as.NumericLiteralNode;
>> import org.apache.royale.compiler.tree.ASTNodeID;
>> import org.apache.royale.compiler.tree.as.IDynamicAccessNode;
>> import org.apache.royale.compiler.tree.as.IExpressionNode;
>> @@ -43,11 +47,39 @@ public class DynamicAccessEmitter extends JSSubEmitter 
>> implements
>>if (leftOperandNode.getNodeID() == ASTNodeID.Op_AtID)
>>  return;
>> 
>> +IExpressionNode rightOperandNode = node.getRightOperandNode();
>> +IJSEmitter ijs = getEmitter();
>> +JSRoyaleEmitter fjs = (ijs instanceof JSRoyaleEmitter) ? 
>> +
>> (JSRoyaleEmitter)ijs : null;
>> +if (fjs != null)
>> +{
>> +boolean isXML = false;
>> +if (leftOperandNode instanceof MemberAccessExpressionNode)
>> +isXML = 
>> fjs.isLeftNodeXMLish((MemberAccessExpressionNode)leftOperandNode);
>> +else if (leftOperandNode instanceof IExpressionNode)
>> +isXML = fjs.isXML((IExpressionNode)leftOperandNode);
>> +if (isXML)
>> +{
>> +ITypeDefinition type = 
>> rightOperandNode.resolveType(getProject());
>> +if (!type.isInstanceOf("int", getProject()) && 
>> !type.isInstanceOf("uint", getProject()) && !type.isInstanceOf("Number", 
>> getProject()) )
>> +{
>> +String field = 
>> fjs.stringifyNode(rightOperandNode);
>> +if (field.startsWith("\"@"))
>> +{
>> +

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-02 Thread Alex Harui

Add a test case to the test suite, propose the fix or ask for help on it.

-Alex

On 10/2/18, 1:03 AM, "Harbs"  wrote:

I don’t know the reason for this change, but this broke bracket access in 
XMLList.

myList[myList.length()] = xml should be output unchanged.

It’s now output as:

myList.child(myList.length()) = xml

That’s wrong…

Harbs

> On Sep 28, 2018, at 6:34 AM, aha...@apache.org wrote:
> 
> This is an automated email from the ASF dual-hosted git repository.
> 
> aharui pushed a commit to branch develop
> in repository 
https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgitbox.apache.org%2Frepos%2Fasf%2Froyale-compiler.gitdata=02%7C01%7Caharui%40adobe.com%7Cea3bf6b27b1d4c681a5d08d6283d7b9a%7Cfa7b1b5a7b34438794aed2c178decee1%7C0%7C0%7C636740641870109874sdata=6nGbwmtGYJKnYRUUPdFKixN%2BVUpEG0Ri9VY8Lp8Ib2k%3Dreserved=0
> 
> commit 86bfe0a6ef8789e8ce065c856ce6f888f7562776
> Author: Alex Harui 
> AuthorDate: Tue Sep 25 12:19:02 2018 -0700
> 
>fix bracket access of XML/XMLList
> ---
> .../codegen/js/jx/DynamicAccessEmitter.java| 34 
+-
> .../codegen/js/royale/TestRoyaleGlobalClasses.java | 30 
++-
> 2 files changed, 62 insertions(+), 2 deletions(-)
> 
> diff --git 
a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
 
b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
> index ae977f9..0d94fac 100644
> --- 
a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
> +++ 
b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
> @@ -21,8 +21,12 @@ package 
org.apache.royale.compiler.internal.codegen.js.jx;
> 
> import org.apache.royale.compiler.codegen.ISubEmitter;
> import org.apache.royale.compiler.codegen.js.IJSEmitter;
> +import org.apache.royale.compiler.definitions.ITypeDefinition;
> import org.apache.royale.compiler.internal.codegen.as.ASEmitterTokens;
> import org.apache.royale.compiler.internal.codegen.js.JSSubEmitter;
> +import 
org.apache.royale.compiler.internal.codegen.js.royale.JSRoyaleEmitter;
> +import 
org.apache.royale.compiler.internal.tree.as.MemberAccessExpressionNode;
> +import org.apache.royale.compiler.internal.tree.as.NumericLiteralNode;
> import org.apache.royale.compiler.tree.ASTNodeID;
> import org.apache.royale.compiler.tree.as.IDynamicAccessNode;
> import org.apache.royale.compiler.tree.as.IExpressionNode;
> @@ -43,11 +47,39 @@ public class DynamicAccessEmitter extends 
JSSubEmitter implements
> if (leftOperandNode.getNodeID() == ASTNodeID.Op_AtID)
>   return;
> 
> +IExpressionNode rightOperandNode = node.getRightOperandNode();
> +IJSEmitter ijs = getEmitter();
> + JSRoyaleEmitter fjs = (ijs instanceof JSRoyaleEmitter) ? 
> + 
(JSRoyaleEmitter)ijs : null;
> + if (fjs != null)
> + {
> + boolean isXML = false;
> + if (leftOperandNode instanceof MemberAccessExpressionNode)
> + isXML = 
fjs.isLeftNodeXMLish((MemberAccessExpressionNode)leftOperandNode);
> + else if (leftOperandNode instanceof IExpressionNode)
> + isXML = fjs.isXML((IExpressionNode)leftOperandNode);
> + if (isXML)
> + {
> + ITypeDefinition type = 
rightOperandNode.resolveType(getProject());
> + if (!type.isInstanceOf("int", getProject()) && 
!type.isInstanceOf("uint", getProject()) && !type.isInstanceOf("Number", 
getProject()) )
> + {
> + String field = 
fjs.stringifyNode(rightOperandNode);
> + if (field.startsWith("\"@"))
> + {
> + field = field.replace("@", "");
> + write(".attribute(" + field + 
")");
> + }
> + else
> + write(".child(" + field + ")");
> + return;
> + }   
> + }
> + }
> + 
> startMapping(node, leftOperandNode);
> write(ASEmitterTokens.SQUARE_OPEN);
> endMapping(node);
> 
> -IExpressionNode rightOperandNode = node.getRightOperandNode();
> getWalker().walk(rightOperandNode);
> 
> startMapping(node, rightOperandNode);
> diff --git

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

2018-10-02 Thread Harbs

I don’t know the reason for this change, but this broke bracket access in 
XMLList.

myList[myList.length()] = xml should be output unchanged.

It’s now output as:

myList.child(myList.length()) = xml

That’s wrong…

Harbs

> On Sep 28, 2018, at 6:34 AM, aha...@apache.org wrote:
> 
> This is an automated email from the ASF dual-hosted git repository.
> 
> aharui pushed a commit to branch develop
> in repository https://gitbox.apache.org/repos/asf/royale-compiler.git
> 
> commit 86bfe0a6ef8789e8ce065c856ce6f888f7562776
> Author: Alex Harui 
> AuthorDate: Tue Sep 25 12:19:02 2018 -0700
> 
>fix bracket access of XML/XMLList
> ---
> .../codegen/js/jx/DynamicAccessEmitter.java| 34 +-
> .../codegen/js/royale/TestRoyaleGlobalClasses.java | 30 ++-
> 2 files changed, 62 insertions(+), 2 deletions(-)
> 
> diff --git 
> a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
>  
> b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
> index ae977f9..0d94fac 100644
> --- 
> a/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
> +++ 
> b/compiler-jx/src/main/java/org/apache/royale/compiler/internal/codegen/js/jx/DynamicAccessEmitter.java
> @@ -21,8 +21,12 @@ package org.apache.royale.compiler.internal.codegen.js.jx;
> 
> import org.apache.royale.compiler.codegen.ISubEmitter;
> import org.apache.royale.compiler.codegen.js.IJSEmitter;
> +import org.apache.royale.compiler.definitions.ITypeDefinition;
> import org.apache.royale.compiler.internal.codegen.as.ASEmitterTokens;
> import org.apache.royale.compiler.internal.codegen.js.JSSubEmitter;
> +import org.apache.royale.compiler.internal.codegen.js.royale.JSRoyaleEmitter;
> +import 
> org.apache.royale.compiler.internal.tree.as.MemberAccessExpressionNode;
> +import org.apache.royale.compiler.internal.tree.as.NumericLiteralNode;
> import org.apache.royale.compiler.tree.ASTNodeID;
> import org.apache.royale.compiler.tree.as.IDynamicAccessNode;
> import org.apache.royale.compiler.tree.as.IExpressionNode;
> @@ -43,11 +47,39 @@ public class DynamicAccessEmitter extends JSSubEmitter 
> implements
> if (leftOperandNode.getNodeID() == ASTNodeID.Op_AtID)
>   return;
> 
> +IExpressionNode rightOperandNode = node.getRightOperandNode();
> +IJSEmitter ijs = getEmitter();
> + JSRoyaleEmitter fjs = (ijs instanceof JSRoyaleEmitter) ? 
> + (JSRoyaleEmitter)ijs : 
> null;
> + if (fjs != null)
> + {
> + boolean isXML = false;
> + if (leftOperandNode instanceof MemberAccessExpressionNode)
> + isXML = 
> fjs.isLeftNodeXMLish((MemberAccessExpressionNode)leftOperandNode);
> + else if (leftOperandNode instanceof IExpressionNode)
> + isXML = fjs.isXML((IExpressionNode)leftOperandNode);
> + if (isXML)
> + {
> + ITypeDefinition type = 
> rightOperandNode.resolveType(getProject());
> + if (!type.isInstanceOf("int", getProject()) && 
> !type.isInstanceOf("uint", getProject()) && !type.isInstanceOf("Number", 
> getProject()) )
> + {
> + String field = 
> fjs.stringifyNode(rightOperandNode);
> + if (field.startsWith("\"@"))
> + {
> + field = field.replace("@", "");
> + write(".attribute(" + field + 
> ")");
> + }
> + else
> + write(".child(" + field + ")");
> + return;
> + }   
> + }
> + }
> + 
> startMapping(node, leftOperandNode);
> write(ASEmitterTokens.SQUARE_OPEN);
> endMapping(node);
> 
> -IExpressionNode rightOperandNode = node.getRightOperandNode();
> getWalker().walk(rightOperandNode);
> 
> startMapping(node, rightOperandNode);
> diff --git 
> a/compiler-jx/src/test/java/org/apache/royale/compiler/internal/codegen/js/royale/TestRoyaleGlobalClasses.java
>  
> b/compiler-jx/src/test/java/org/apache/royale/compiler/internal/codegen/js/royale/TestRoyaleGlobalClasses.java
> index 85b3a15..b2c8a17 100644
> --- 
> a/compiler-jx/src/test/java/org/apache/royale/compiler/internal/codegen/js/royale/TestRoyaleGlobalClasses.java
> +++ 
> b/compiler-jx/src/test/java/org/apache/royale/compiler/internal/codegen/js/royale/TestRoyaleGlobalClasses.java
> @@ -482,7 +482,17 @@ public class TestRoyaleGlobalClasses extends 
> TestGoogGlobalClasses
> asBlockWalker.visitVariable(node);
>

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

Re: [royale-compiler] 01/03: fix bracket access of XML/XMLList

8 matches

Site Navigation

Mail list logo

Footer information