Add another test on debug global evaluate

Add another test on debug global evaluate

Committed: http://code.google.com/p/v8/source/detail?r=3639

[Peter Rybin, 2010-01-16 23:50:50]

Hi Søren

I developed this test when I thought that supporting "global evaluate" will take
more efforts than it actually took.

Now there are no changes that actually requires this test.

However, this test seems to be pretty accurate and it also improves some shared
function. I don't know whether I should dispose it.

How about submitting it?

Peter

[Søren Thygesen Gjesse, 2010-01-18 10:22:52]

LGTM

http://codereview.chromium.org/553009/diff/1/2
File test/cctest/test-debug.cc (right):

http://codereview.chromium.org/553009/diff/1/2#newcode2166
test/cctest/test-debug.cc:2166: int Utf16ToAscii(const uint16_t* input_buffer,
Just the function definition to the top of the file.

http://codereview.chromium.org/553009/diff/1/2#newcode2170
test/cctest/test-debug.cc:2170: bool GetEvaluateStringResult(char *message,
char* buffer, int buffer_size);
Move the function definition up here.

http://codereview.chromium.org/553009/diff/1/2#newcode2190
test/cctest/test-debug.cc:2190: static void DebugProcessDebugMessagesHanlder(
Hanlder -> Handler

http://codereview.chromium.org/553009/diff/1/2#newcode2200
test/cctest/test-debug.cc:2200: % DebugProcessDebugMessagesData::kArraySize;
Please move this logic into DebugProcessDebugMessagesData, e.g. method current()
which returns a EvaluateResult*.

http://codereview.chromium.org/553009/diff/1/2#newcode2206
test/cctest/test-debug.cc:2206: process_debug_messages_data.counter++;
Please move this logic into DebugProcessDebugMessagesData, e.g. method next().
This method can also have a flag for whether the counter has wrapped, and that
value can be checked in the test.

http://codereview.chromium.org/553009/diff/1/2#newcode2210
test/cctest/test-debug.cc:2210: int AsciiToUtf16(const char* input_buffer,
uint16_t* output_buffer);
Just the function definition to the top of the file.

http://codereview.chromium.org/553009/diff/1/2#newcode2225
test/cctest/test-debug.cc:2225: v8::Script::Compile(v8::String::New("void(0)"));
I don't think the void0_script is used.

[Peter Rybin, 2010-01-18 19:15:09]

http://codereview.chromium.org/553009/diff/1/2
File test/cctest/test-debug.cc (right):

http://codereview.chromium.org/553009/diff/1/2#newcode2166
test/cctest/test-debug.cc:2166: int Utf16ToAscii(const uint16_t* input_buffer,
On 2010/01/18 10:22:52, Søren Gjesse wrote:
> Just the function definition to the top of the file.

Done.

http://codereview.chromium.org/553009/diff/1/2#newcode2170
test/cctest/test-debug.cc:2170: bool GetEvaluateStringResult(char *message,
char* buffer, int buffer_size);
On 2010/01/18 10:22:52, Søren Gjesse wrote:
> Move the function definition up here.

Done.

http://codereview.chromium.org/553009/diff/1/2#newcode2190
test/cctest/test-debug.cc:2190: static void DebugProcessDebugMessagesHanlder(
On 2010/01/18 10:22:52, Søren Gjesse wrote:
> Hanlder -> Handler

Done.

http://codereview.chromium.org/553009/diff/1/2#newcode2200
test/cctest/test-debug.cc:2200: % DebugProcessDebugMessagesData::kArraySize;
On 2010/01/18 10:22:52, Søren Gjesse wrote:
> Please move this logic into DebugProcessDebugMessagesData, e.g. method
current()
> which returns a EvaluateResult*.

Done.

http://codereview.chromium.org/553009/diff/1/2#newcode2206
test/cctest/test-debug.cc:2206: process_debug_messages_data.counter++;
On 2010/01/18 10:22:52, Søren Gjesse wrote:
> Please move this logic into DebugProcessDebugMessagesData, e.g. method next().
> This method can also have a flag for whether the counter has wrapped, and that
> value can be checked in the test.

Done.

http://codereview.chromium.org/553009/diff/1/2#newcode2210
test/cctest/test-debug.cc:2210: int AsciiToUtf16(const char* input_buffer,
uint16_t* output_buffer);
On 2010/01/18 10:22:52, Søren Gjesse wrote:
> Just the function definition to the top of the file.

Done.

http://codereview.chromium.org/553009/diff/1/2#newcode2225
test/cctest/test-debug.cc:2225: v8::Script::Compile(v8::String::New("void(0)"));
On 2010/01/18 10:22:52, Søren Gjesse wrote:
> I don't think the void0_script is used.

Done.