Hi Karl, H Ina, ... This is the big problem! How to hide the data while in fact still allowing external access to it. This can be done by making things private...
... I support this idea. It is more difficult to modify private data by accident. This is what we want I suppose. Using an accessor supposes you know what you...
Karl Meerbergen
Karl.Meerbergen@...
Sep 1, 2004 9:17 am
1896
Hi, Since functor_type has become private in matrix<T,F,A>, I have modified the bindings so that now orientation_category is used instead. I have committed the...
Karl Meerbergen
Karl.Meerbergen@...
Sep 1, 2004 2:31 pm
1897
Dear All, Every time I make commits I always worry about what I will break for people using the current CVS version. The trouble is I need to commit to get ...
Dear All, Now that uBLAS is compiling (and hopefully) working so well on so many compilers I have made another round of changes! Firstly I have introduced STL...
Hi, Doing a cvs update of the Boost CVS I encountered a compiler bug in ublas: ...boost/boost/numeric/ublas/matrix_sparse.hpp:2556: error: no matching ...
Hi Matthias, My mistake when adding Allocators. I had temporarily removed to default constructor from unbounded_array. I have now given it a new one! This...
Dear Patrick, ... That is bad! ... I think this should be the same problem as Matthias reported and should be fixed now. ... These are the proxy shortcuts! The...
Hi all, I compiled a file with the following lines: #include <boost/numeric/ublas/matrix_sparse.hpp> #include <iostream> int main() { using namespace...
Karl Meerbergen
Karl.Meerbergen@...
Sep 2, 2004 10:44 am
1904
... Ublas uses rather advanced C++ functionalities and thus sometimes we hit trouble with different compilers. This situation will improve because 1) compilers...
Michael, attached is a simple example that runs perfectly when compiled against Boost 1.31 but aborts at run time when compiled against currect Boost CVS. Hope...
... (Sorry for the long delay). Currently none of the regression tests compile on vacpp. But I'm going over them in following days trying to find the causes of...
... How did you come up with BOOST_UBLAS_ENABLE_PROXY ? I have grepped all of ublas but found nothing. Could I add following to ublas/config.hpp #ifndef...
Hi, I'm looking for a ublas function which accesses some vector elements as a matrix, e.g. something like reshape() in FORTRAN or Matlab. There exists...
Toon, ... Or we could simply go back to the previous default to keep people happy! I think we should collectively decided either way so that user code does not...
Angus, ... Fun fun fun! ... Sure did! It is fixed and committed. I thought it would be a nasty problem with the very hard proxy or sparse matrix code. But that...
Ralf, ... uBLAS has an intelligent resize. It's default behaviour is to preserve the common sub vector or sub matrix of the previous and new size. New elements...
... Perhaps, we should first have a discussion on the interface and user friendly-ness of ublas objects before making decisions on whether to drop or develop...
Karl Meerbergen
Karl.Meerbergen@...
Sep 2, 2004 1:29 pm
1916
Hi Michael, ... I now introduced ctors and = in my classes. This works fine for me. For the range things, see other posting. Patrick...
Hello, ... Appreciated it is long! ... This could be a good compromise! The shortcuts are only "syntactic sugar" so the they don't need full generality. For...
Toon, ... Hmmm. My new information was from Generic Programming and the STL. This explictly excludes size==0 in Allocator concept. Stroupstrop did mention it...
... This is a completely different syntax from what we have now. It suggests that the classes range and slice are no longer required. This is not necessarily a...
Karl Meerbergen
Karl.Meerbergen@...
Sep 2, 2004 1:53 pm
1921
... preserve the ... elements ... create a new ... Not really. Let's say I have a vector v with v.size() = 20. Now I want to access this as a matrix m with...