Hi, I have made changes, sending them soon. Dimitry: In your first suggestion, i think using a new variable is not a good idea. It has been done like that for all options and previously for the same options. next time i will think of something valid. In your third suggestion, you asked me not to set to 0, if set to 0xdeadbeef, that means i have to check for 0xdeadbeef. :(. when does API set its error to 0, what might be the cases generally. In your Last suggestion if i initialize the length variable in the tests, the native simply gives garbage, thats inorder to make the test pass. Thanks for your suggestions, Vijay On 11/9/05, Robert Shearman <rob(a)codeweavers.com> wrote:
Vijay Kiran Kamuju wrote:
+ case INTERNET_OPTION_ASYNC: + case INTERNET_OPTION_ASYNC_ID: + case INTERNET_OPTION_ASYNC_PRIORITY: + case INTERNET_OPTION_CALLBACK_FILTER: + case INTERNET_OPTION_CODEPAGE: + case INTERNET_OPTION_CONNECT_BACKOFF: + case INTERNET_OPTION_CONNECT_TIME: + case INTERNET_OPTION_DATA_RECEIVE_TIMEOUT: + case INTERNET_OPTION_DATA_SEND_TIMEOUT: + case INTERNET_OPTION_DISABLE_AUTODIAL: + case INTERNET_OPTION_DISCONNECTED_TIMEOUT: + case INTERNET_OPTION_IDLE_STATE: + case INTERNET_OPTION_KEEP_CONNECTION: + case INTERNET_OPTION_LISTEN_TIMEOUT: + case INTERNET_OPTION_OFFLINE_MODE: + case INTERNET_OPTION_OFFLINE_SEMANTICS: + case INTERNET_OPTION_POLICY: + case INTERNET_OPTION_RECEIVE_THROUGHPUT: + case INTERNET_OPTION_SEND_THROUGHPUT: + WARN("This request is not supported by native either\n");
You should print out the option being requested, otherwise the "This" in "This request" is meaningless. Just printing the option number is fine.
-- Rob Shearman