Tucker - There's still something wrong with this.
I saw that Max had lzpix running on SWF9 and I tried it with this
change,
and it's not really functional. I'll try to figure out what's going
on tomorrow.
On Jul 16, 2008, at 11:10 PM, Donald Anderson wrote:
Change 20080716-dda-V by [EMAIL PROTECTED] on
2008-07-16 21:26:41 EDT
in /Users/dda/laszlo/src/svn/openlaszlo/trunk-a
for http://svn.openlaszlo.org/openlaszlo/trunk
Summary: Fix remapLocals and withThis in SWF9
New Features:
Bugs Fixed: LPP-6637
Technical Reviewer: ptw (pending)
QA Reviewer: (pending)
Doc Reviewer: (pending)
Documentation:
Release Notes:
Details:
The SWF9 code generator has had a long standing bug that prevented
the withThis pragma from working. This change set fixes the
fundamental bug and
other smaller problems in order to allow withThis to function
properly.
withThis is necessary for the new approach to handling states.
The code that disables the withThis from being operation has
been removed.
The bug was an extra visit of the function - the second visit
sees some
artifacts left by the first visit, and gets confused. This has
been fixed.
Mixins in SWF9 are handled by keeping the AST for the mixin and
visiting
it whenever it is mixed in. The visited AST was not a complete
copy,
only a shallow copy, and this exposed the same sort of extra
visit bug.
A deep copy function for SimpleNode (the basis for AST) was added.
This required minor changes to all existing handcoded AST
classes to
support deep copy. Fortunately there are not many of these
classes,
but if we add any, we'll need to implement the deepCopy()
function.
Remapped formal arguments, (e.g. function foo(a, b) => function
foo($1, $2)),
must retain any ellipsis from the original declaration (e.g.
function foo(a, ...b) => function foo($1, ...$2))
I am also retaining type info, (e.g. function foo(a:*, b) =>
function foo($1:*, $2))
Tests:
Passes test case given by ptw in LPP-6637.
Builds LFC without errors with withThis code enabled.
Regression tests: smokecheck for swf8/dhtml.
Unfortunately SWF9 hello seems broken (not due to this change),
so unable to test that.
Files:
M WEB-INF/lps/server/src/org/openlaszlo/sc/
JavascriptGenerator.java
M WEB-INF/lps/server/src/org/openlaszlo/sc/SWF9Generator.java
M WEB-INF/lps/server/src/org/openlaszlo/sc/Compiler.java
M WEB-INF/lps/server/sc/src/org/openlaszlo/sc/parser/
SimpleNode.java
M WEB-INF/lps/server/sc/src/org/openlaszlo/sc/parser/
ASTModifiedDefinition.java
M WEB-INF/lps/server/sc/src/org/openlaszlo/sc/parser/
ASTOperator.java
M WEB-INF/lps/server/sc/src/org/openlaszlo/sc/parser/
ASTLiteral.java
M WEB-INF/lps/server/sc/src/org/openlaszlo/sc/parser/
ASTFormalParameterList.java
M WEB-INF/lps/server/sc/src/org/openlaszlo/sc/parser/
ASTIdentifier.java
M WEB-INF/lps/server/sc/src/org/openlaszlo/sc/parser/
ASTPassthroughDirective.java
Changeset: http://svn.openlaszlo.org/openlaszlo/patches/20080716-dda-V.tar
--
Don Anderson
Java/C/C++, Berkeley DB, systems consultant
voice: 617-547-7881
email: [EMAIL PROTECTED]
www: http://www.ddanderson.com
--
Don Anderson
Java/C/C++, Berkeley DB, systems consultant
voice: 617-547-7881
email: [EMAIL PROTECTED]
www: http://www.ddanderson.com
