[PATCH v10 1/4] jscript: Get rid of the variable_obj parameter to exec_source.

[Gabriel Ivăncescu]

On 17/03/2020 20:22, Jacek Caban wrote:
> On 17.03.2020 13:59, Gabriel Ivăncescu wrote:
>> Hi Jacek,
>>
>> On 16/03/2020 20:32, Jacek Caban wrote:
>>> On 12.03.2020 18:11, Gabriel Ivăncescu wrote:
>>>> Signed-off-by: Gabriel Ivăncescu <gabrielopcode at gmail.com>
>>>> ---
>>>>   dlls/jscript/engine.c   | 39 +++++++++++++++++++++++++++++----------
>>>>   dlls/jscript/engine.h   |  2 +-
>>>>   dlls/jscript/function.c |  6 ++----
>>>>   dlls/jscript/global.c   |  2 +-
>>>>   dlls/jscript/jscript.c  |  6 +++---
>>>>   5 files changed, 36 insertions(+), 19 deletions(-)
>>>
>>>
>>> This patch breaks tests.
>>
>> Sorry for delay, but what tests does it break exactly? I'm not getting 
>> any failures. I tried mshtml_test script too.
> 
> 
> All of them with my build. You used uninitialized hres in function.c. I 
> fixed that and sent a new version.
> 

Oh yeah that was dumb of me, thanks. I don't know what the usual policy 
is now, but: should I resend the rest of the patches, or are they still 
under review and don't need to be resent?

> 
>>> Also, what was the exact problem with handling EXEC_RETURN_TO_INTERP?
>>>
>>
>> If I add a check for that flag on top of EXEC_EVAL, I get failures in 
>> jscript/run.c test_eval:
>>
>> run.c:2931: Test failed: GetDispID(x) returned 00000000
>> run.c:2931: Test failed: GetDispID(x) returned 00000000 
> 
> 
> I see, that part is fine then.
> 
> 
> Thanks,
> 
> Jacek
>