Added some missing initialization checks to the debugger API....

Added some missing initialization checks to the debugger API.

BUG=3723
(Chromium)

Committed: http://code.google.com/p/v8/source/detail?r=648

[Kasper Lund, 2008-10-30 10:09:34]

LGTM. Two comments:

http://codereview.chromium.org/8909/diff/1/2
File src/api.cc (right):

http://codereview.chromium.org/8909/diff/1/2#newcode2890
Line 2890: if (!i::V8::HasBeenSetup()) return;
Does the command sender expect an answer? If so, this seems a bit scary.

http://codereview.chromium.org/8909/diff/1/3
File test/cctest/test-debug.cc (right):

http://codereview.chromium.org/8909/diff/1/3#newcode3148
Line 3148: static void DummyMessageHandler(const uint16_t* message, int length,
void *data) {
Does this lint?

[Søren Thygesen Gjesse, 2008-10-30 10:27:02]

http://codereview.chromium.org/8909/diff/1/2
File src/api.cc (right):

http://codereview.chromium.org/8909/diff/1/2#newcode2890
Line 2890: if (!i::V8::HasBeenSetup()) return;
On 2008/10/30 10:09:34, Kasper Lund wrote:
> Does the command sender expect an answer? If so, this seems a bit scary.

All command responses are send through a callback the the function registered
with SetMessageHandler which ensures initialization.

http://codereview.chromium.org/8909/diff/1/3
File test/cctest/test-debug.cc (right):

http://codereview.chromium.org/8909/diff/1/3#newcode3148
Line 3148: static void DummyMessageHandler(const uint16_t* message, int length,
void *data) {
On 2008/10/30 10:09:34, Kasper Lund wrote:
> Does this lint?

Sorry, broke the line.

[Erik does not do reviews, 2008-10-30 16:37:52]

LGTM

From your reply to Kasper, it seems like it's unexpected for the debugger to be
uninitialized in some of these cases.  If that's the case, should these be
DCHECKs instead of early returns?