hi all, I'm working on Pytave - a Python<->Octave bridge that can either use NumPy or the older Numeric to map onto Octave's arrays. I would like to implement support for NumPy's string arrays, but I'm a little confused about how the data is stored internally. Does PyArrayObject::data point to a single contiguous char[] buffer, like with the old Numeric char arrays, with PyArrayObject::descr->elsize being the maximum length? If yes, how are string lengths determined (I noticed that the strings stored can have differing lengths when displayed from the interpreter)?
Finally, is there any way to create an array in NumPy (from within the interpreter) that would have type == PyArray_CHAR? (There will be no need for that when I implement the above, I'm just wondering whether I overlooked something). Many thanks in advance. -- RNDr. Jaroslav Hajek computing expert & GNU Octave developer Aeronautical Research and Test Institute (VZLU) Prague, Czech Republic url: www.highegg.matfyz.cz _______________________________________________ NumPy-Discussion mailing list [email protected] http://mail.scipy.org/mailman/listinfo/numpy-discussion
