|
From: SourceForge.net <no...@so...> - 2007-07-18 05:25:06
|
Feature Requests item #1755527, was opened at 2007-07-17 10:33 Message generated for change (Comment added) made by rupole You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=551957&aid=1755527&group_id=78018 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: win32 Group: None Status: Open Resolution: None Priority: 5 Private: No Submitted By: ken (kchmbrln) Assigned to: Nobody/Anonymous (nobody) Summary: Default to buffer for win32ts.WTSQuerySessionInformation Initial Comment: Currently win32ts.WTSQuerySessionInformation raises a PyExc_NotImplementedError exception in those cases where no specific interpretation of the results has been provided. It would be more useful if, instead of an exception, it would return a buffer object with the raw contents of the buffer that the api function returned copied into it. Then the user could use something like struct.unpack_from(fmt, buffer [,offset = 0]) to interpret the results as needed. Something like (untested): switch (WTSInfoClass){ ... default: void *newbuf ret=PyBuffer_New(bufsize); ret->ob_type->tp_as_buffer->bf_getwritebuffer(ret, 0, &newbuf); memcpy(newbuf, buf, bufsize); } ---------------------------------------------------------------------- >Comment By: Roger Upole (rupole) Date: 2007-07-18 00:25 Message: Logged In: YES user_id=771074 Originator: NO Another issue with copying the buffer is that the buffer could contain pointers to addresses inside the original buffer, which would become invalid after copying and freeing the original buffer. There are only 2 classes of information not supported yet (WTSClientAddress and WTSClientDisplay), so I'll go ahead and implement them. ---------------------------------------------------------------------- Comment By: Mark Hammond (mhammond) Date: 2007-07-17 16:46 Message: Logged In: YES user_id=14198 Originator: NO If we do that, it would prevent us upgrading the code in the future (as it would then break code which assumed the buffer). An alternative might be an extra param to indicate if a buffer is wanted or not, which would then be used in all cases. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=551957&aid=1755527&group_id=78018 |
