I am using pywin32 scripting as a client to my C++ COM application. Many of my
functions in C++ are returning errors such as E_FAIL, E_POINTER, E_INVALIDARG,
etc...
However, the error message in Python is often cryptic such as a simple
"Exception occurred" with the error code.
I suppose I could h
This problem seems to be related to a post from a while back
http://markmail.org/message/moyolyvs5i26vbwr
It all comes down to the getevents(clsid) function defined in
win32com/client/__init__.py
This function expects an incoming clsid, but it never is one. The incoming clsid
is _actually_ the II
Mark Hammond skippinet.com.au> writes:
>
> Eeek. Have you actually imported the other library, or simply copied the
> interface definition. I'd be surprised to find the same interface defined
> in 2 typelibs (but not surprised to see cross-typelib references).
>
I think I was a bit vague in m
Brad Johnson ballardtech.com> writes:
Just as a follow-up, this didn't even end up being a complete solution. With my
particular combination of type libraries in my application, I can't rely on the
gencache to hold the correct mapping in the mapCLSIDToClass dict for reasons
state
Mark Hammond skippinet.com.au> writes:
>
> While you have obviously struck a bug, I need to understand it more before I
> can comment further. I'll try and reproduce it using the C++ test object
> and IDL file - if you could help do that it would be greatly appreciated and
> help a timely solut
I have embedded the Python interpreter into a COM application. In this
application, I create worker threads that execute Python code.
I marshal some COM objects that live in the UI thread of the application over to
the worker thread and wrap them with EnsureDispatch() for use in the new thread.
M
> > However, while it is waiting for the
> > GIL in the UI thread (when the worker thread has it), it is locked into a
> > WaitForSingleObject call (some layers below PyGILState_Ensure).
>
> Obviously the GIL can only be held by one thread at a time. The
> question then is why your worker thread
> So one your thread has acquired the GIL, one of 3 things happens:
>
> * It returns quickly and the GIL is released.
> * It executes many many instructions: every 'n' instructions Python will
> temporarily release the GIL for you.
> * It calls some other blocking function. This other blocking f
I have a C++ application that creates a collection of COM objects. I am
embedding a Python interpreter that will have access to Automation objects
written in C++.
I would like to give the Python interpreter access to the IDispatch interface on
these objects.
Stated another way, how can I have Pyt
>
> PyCom_PyObjectFromIUnknown can be used to wrap an IDispatch pointer in a
> PyIDispatch object.
>
> Roger
>
Thanks a ton, Roger! I spend a good amount of time trying to find an answer to
my question, but now that I have that function name there is a bunch of info
through Google to ge
Brad Johnson ballardtech.com> writes:
I have successfully used the PyCom_PyObjectFromIUnknown function to get a
PyIDispatch object from a raw IDispatch pointer.
My next goal is to get an early bound class instance from that pointer (using a
class generated from makepy).
I can take
I am embedding a Python interpreter in a C++ app for Windows.
The main thread in my application calls PyEval_InitThreads();
At some point, the main thread creates a new thread to do some concurrent work
using the Python interpreter.
The trouble is that in my worker thread I am never able to comp
Brad Johnson ballardtech.com> writes:
>
> I am embedding a Python interpreter in a C++ app for Windows.
>
> The main thread in my application calls PyEval_InitThreads();
>
> At some point, the main thread creates a new thread to do some concurrent work
> using the Pyth
I have implemented an object that supports connection points in COM.
In ATL, there is a #define that limits the number of parameters supported for a
connection point client:
#define _ATL_MAX_VARTYPES 8
I have one function that has 11 parameters, so for my C++ client (using ATL) I
had to redefine
Mark Hammond skippinet.com.au> writes:
> There is no arbitrary limit that I recall. Can you show us a snippet of
> failing code and the traceback?
>
> Thanks,
>
> Mark
>
Hi Mark,
I've gotten right down to the problem. Has nothing to do with the number of
parameters. One of the parameters is
Brad Johnson ballardtech.com> writes:
> So I guess my question becomes how can I change the parameter type so everyone
> is happy?
>
Never mind. I managed to make it a reference to a SAFEARRAY of USHORTS without a
hiccup from either client. I guess I need to brush up a little b
Mark Hammond skippinet.com.au> writes:
> Pass it as a variant holding the actual type instead of adding a kind of
> "indirection".
>
> Cheers,
>
> Mark
>
Thanks.
That's exactly what I ended up doing. I guess I was just a little too green to
do it right the first time. I didn't know how to pas
Mark Hammond skippinet.com.au> writes:
> but the code fails to gracefully handle the non-byref variant case, as you
> found. Oh well, some other day (or if someone is keen, a bug couldn't
> hurt...)
>
> Cheers,
>
> Mark
>
I added a bug to the SourceForge bug tracker :-).
___
Hello everyone,
I am attempting to create an MSI of pywin32 so I can have it silently install
with another application.
I am running into this error and traceback when executing "python setup.py
bdist_msi"
running install_egg_info
Writing build\bdist.win32\msi\Lib\site-packages\pywin32-210-py2.5
Mark Hammond skippinet.com.au> writes:
>
> Hi all,
> FYI, I'm off on a 2 week vacation to China (wh!)
>
Hope it is an enjoyable break for you, you deserve it :-)
___
python-win32 mailing list
python-win32@python.org
http://mail.python.org/mai
Brad Johnson ballardtech.com> writes:
Note, I can fix this by hard coding a valid version in
\distutils\command\bdist_msi.py by changing
sversion = "%d.%d.%d" % StrictVersion(version).version
to
sversion = "210"
(where 210 is the build number)
But, obviously this just
21 matches
Mail list logo
