monitor console command implemented.

Source/core/inspector/InjectedScriptHost.idl

Index: Source/core/inspector/InjectedScriptHost.idl
diff --git a/Source/core/inspector/InjectedScriptHost.idl b/Source/core/inspector/InjectedScriptHost.idl
index 952884dc0327bfcd1e50ea38fbdbdd10af3e6d6e..33092abe58bcdcfb384936caa0b1ea6742acca6a 100644
--- a/Source/core/inspector/InjectedScriptHost.idl
+++ b/Source/core/inspector/InjectedScriptHost.idl
@@ -47,8 +47,8 @@
     [Custom] any evaluate(DOMString text);
     [Custom] DOMString databaseId(any database);
     [Custom] DOMString storageId(any storage);
-    [Custom] boolean setBreakpoint(any fn);
-    [Custom] boolean removeBreakpoint(any fn);
+    void setBreakpoint(DOMString scriptId, long line, long column, DOMString source, DOMString condition);

[yurys, 2013/06/13 09:46:51]

I don't think it is a good idea to let inspected page pass some arbitrary JS
code that would be evaluated by the debugger. It is a potential security
problem. Can we avoid constructing this condition in the injected script and do
it somewhere the inspected code doesn't have access to?

[SeRya, 2013/06/13 10:45:14]

Moved to C++.

+    void removeBreakpoint(DOMString scriptId, long line, long column, DOMString source);
 
     // Only declarative scope (local, with and catch) is accepted. Returns undefined. 
     [Custom] any setFunctionVariableValue(any functionObject, int scopeIndex, DOMString variableName, any newValue);