Hello Alex,
On 10/13/2014 08:19 AM, Alex Henrie wrote:
The only difference from the previous version of this patch is the correction of a typo in one of the comments.
dlls/kernel32/tests/codepage.c | 751 ++++++++++++++++++++++++++++++++++ +++++++ 1 file changed, 751 insertions(+)
while more tests are good that's a lot of lines to have for just one function. So splitting the WCHAR => UTF-7 from the UTF-7 => WCHAR tests should help make it more manageable.
Also some of the strings are huge. Most of those seem to be for the WCHAR => UTF-7 and could be probably generated by just using the directly_encodable_table[] from patch 2. If the tests cannot be generated from the table then please split the strings into more manageable chunks, as is they are unreadable / unmaintainable.
Also the due to the size of the patch it is hard to discern what the difference between the different test structs are. Especially as the names are basically equivalent in this context: - utf16_to_utf7_tests and wcstombs_tests - utf7_to_utf16_tests and mbstowcs_tests
bye michael
-
Michael Stefaniuc