321021 - Unable to get exceptions from JS_EvaluateScript and friends

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

[clkao]

User-Agent:       Mozilla/5.0 (Macintosh; U; PPC Mac OS X; zh-tw) AppleWebKit/416.12 (KHTML, like Gecko) Safari/416.13
Build Identifier: 

One should be able to use JS_IsExceptionPending() to check if the context contains uncaught exceptions after JS_EvaluateScript is called.  However, the uncaught exceptions are reported as error by the LAST_FRAME_CHECKS macro.  Being reported as error means the embedder has no way accessing the raw exception object.  Attached is an example.


Reproducible: Always

Steps to Reproduce:
Compile the test script and run.  Expected to see "returning false 1", rahter than "returning false 0"

Comment 1

[clkao]

Attachment: test case

Comment 2

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Yeah, this sucks, and we think it's biting others as well.

I proposed adding a context option to suppress LAST_FRAME_CHECKS' exception-to-error conversion.

Anyone want to throw another option on the table?

Comment 3

[Brendan Eich [:brendan]]

Yeah, JSOPTION_DONT_REPORT_UNCAUGHT, unless someone has a better idea.  Shaver, you want to field this one?

/be

Comment 4

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

My other idea was to have LAST_FRAME_CHECKS still call the error reporter (if set?) but leave the exception state intact.  I don't like that idea as much as JSOPTION_DONT_REPORT_UNCAUGHT, though the double semantic negative in that name gives me mild pause.

Comment 5

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Attachment: Add JSOPTION_DONT_REPORT_UNCAUGHT

This does it for me.  I added

    JS_SetOptions(cx, JSOPTION_DONT_REPORT_UNCAUGHT);

after the JS_SetGlobalObject call in the aforementioned testcase.

Comment 6

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 206743 [details] [diff] [review]
Add JSOPTION_DONT_REPORT_UNCAUGHT

API compatibility always seems to make things uglier. I should remember to use this instead of a fake fp in evalInSandbox!

Comment 7

[Brendan Eich [:brendan]]

Comment on attachment 206743 [details] [diff] [review]
Add JSOPTION_DONT_REPORT_UNCAUGHT

>Index: jsapi.c
>===================================================================
>RCS file: /cvsroot/mozilla/js/src/jsapi.c,v
>retrieving revision 3.229
>diff -u -p -r3.229 jsapi.c
>--- jsapi.c	5 Dec 2005 21:35:15 -0000	3.229
>+++ jsapi.c	24 Dec 2005 01:30:59 -0000
>@@ -3646,7 +3646,7 @@ JS_CompileUCScript(JSContext *cx, JSObje
> #if JS_HAS_EXCEPTIONS
> # define LAST_FRAME_EXCEPTION_CHECK(cx,result)                                \
>     JS_BEGIN_MACRO                                                            \
>-        if (!(result))                                                        \
>+        if (!(result) && (!(cx)->options & JSOPTION_DONT_REPORT_UNCAUGHT))    \

Oops, you meant

        if (!(result) && !((cx)->options & JSOPTION_DONT_REPORT_UNCAUGHT))

>+#define JSOPTION_DONT_REPORT_UNCAUGHT \
>+                                JS_BIT(8)       /* prevent exceptions left
>+                                                   uncaught, when returning from
>+                                                   the outermost frame or API
>+                                                   call, from being converted
>+                                                   to an error report */

Couple of English nits:
- Use plural consistently: "prevent exceptions... from being converted into error reports".
- Transpose the dependent clause: "When returning from the outermost API call, prevent ...".

/be

Comment 8

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Attachment: deponents and negations transposed, nouns enpluralificated

I am so getting "transpose the deponent clause" on a T-shirt.

Comment 9

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 206745 [details] [diff] [review]
deponents and negations transposed, nouns enpluralificated

Your jsapi.c changes seem unchanged from above.

Comment 10

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Attachment: you have no idea how late it is here, in my brain

Thanks.

Comment 11

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 206746 [details] [diff] [review]
you have no idea how late it is here, in my brain

I could guess, but I'll give you the benefit of the doubt.

Comment 12

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Comment on attachment 206746 [details] [diff] [review]
you have no idea how late it is here, in my brain

The double "prevent" in the jsapi.h comment is fixed my tree, natch.

Comment 13

[Brendan Eich [:brendan]]

Comment on attachment 206746 [details] [diff] [review]
you have no idea how late it is here, in my brain

sr=me with undoubled "prevent".

/be

Comment 14

[Brendan Eich [:brendan]]

shaver, you gonna check this in soon?

/be

Comment 15

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Fixed.