I've just made a simple testcase to verify that LPP-8151 can be closed:
---
<canvas debug="true">
    <handler name="oninit">
        if (void 0 instanceof String) {}
    </handler>
</canvas>
---

The debugger intercepted the runtime error, but the file and line 
information aren't really accurate:
---
ERROR @events/LaszloEvents.lzs#513: TypeError: Error #1010: A term is 
undefined and has no properties.
---

This could lead to some confusion and make it a bit difficult to find 
the error in the source code. Should we worry about this?


On 9/27/2009 7:51 PM, P T Withington wrote:
> [UPDATED to address André's review comments and LPP-8513]
>
>
> Change 20090925-ptw-s by [email protected] on 2009-09-25 11:41:42 EDT
>     in /Users/ptw/OpenLaszlo/trunk
>     for http://svn.openlaszlo.org/openlaszlo/trunk
>
> Summary: Eliminate Debug.evalCarefully, now that it is no longer needed
>
> Bugs Fixed:
> LPP-8479 Debug.evalCarefully not necessary after LPP-8222?
> LPP-8513 Improved debugging of runtime errors injects unnecessarily 
> "with(this)"
>
> Technical Reviewer: [email protected] (pending)
> QA Reviewer: hminsky (pending)
>
> Release Notes:
>     When debugging is on, we will catch, report, and ignore errors
>     that are not specifically declared with `#pragma
>     "throwsError=true".
>
> Details:
>     test/lpp-8222:  Update to reflect that turning on debugging is
>     essentially the same as turning on catcherrors.
>
>     JavascriptGenerator: Add a few more well-known globals to reduce
>     checking overhead.  Replace the concept of a 'checked node' with
>     just making sure the source annotations are correct, in case the
>     node causes a runtime error.  Restrict variable analysis to
>     unannotated function body, except add in annotation variables for
>     renaming.  Simplify the error-catching logic.  In the error
>     handler, if backtracing is on, use the more accurate runtime line
>     information.
>
>     Compiler.java:  Add a subclass of PassthroughNode, AnnotatedNode,
>     so we can be a little more efficient and not double-annotate.
>     (This unfortunately doesn't prevent all redundant settings of the
>     backtrace line number.)
>
> Tests:
>     Verify that test case in bug displays error (with correct line
>     number, when backtrace is enabled).
>
>     Examine output of compiler in DHTML to see that line annotations
>     are inserted for "checked" nodes.
>
>     smokecheck x {swf10, dhtml}
>
> Files:
> M      test/lpp-8222.lzx
> M      WEB-INF/lps/server/src/org/openlaszlo/sc/JavascriptGenerator.java
> M      WEB-INF/lps/server/src/org/openlaszlo/sc/Compiler.java
>
> Changeset: 
> http://svn.openlaszlo.org/openlaszlo/patches/20090925-ptw-s.tar
>
_______________________________________________
Laszlo-reviews mailing list
[email protected]
http://www.openlaszlo.org/mailman/listinfo/laszlo-reviews

Reply via email to