Hi:
I want to:
capture the word under my cursor and translate it into Chinese. A
normal way to achieve this is injecting a DLL and hooking the
systemcall 'ExTextOut'; when my cursor move over the word, an 'invalidatedata'
will be invoked and 'ExTextOut' too, then I
can get the word.
Since the 'render' process of Chrome has no window, I create a remote
thread into the 'render' process's address
space and call 'LoadLibrary' to load my DLL. The result is the 'OpenFile'
failed.
I found:
'OpenFile' is hooked and the address is 'ErrorXXX'(or sth like that).
So I manually changed the address to the real adress of 'OpenFile' in 'NTDLL',
but it still failed
with error: 'permission denied'.
My question is:
what the sandbox does is just hooking a subset of the systemcall such
as 'OpenFile' 'CreateProcess' etc ?
what is the correct way to achieve my goal?
thanks a lot.
ZHOU Xiaobo
--
Chromium Developers mailing list: [email protected]
View archives, change email options, or unsubscribe:
http://groups.google.com/group/chromium-dev
