Release 1.54.0 beta1

[SVN r84619]

doc/projects.html

@@ -396,27 +396,33 @@
       </blockquote>
     </dd>
 
-    <dt><b><a href="http://www.rationaldiscovery.com">Rational Discovery
-    LLC</a></b></dt>
+    <dt><b><a href="http://pyrap.googlecode.com/">Pyrap</a></b></dt>
+    <dd>
+      <p><a href="diepen@astron.nl"
+        >Ger van Diepen</a> writes:</p>
+
+      <blockquote>
+        <p>Pyrap is the python interface to the Radio-Astronomical Package
+        casacore (<a href="http://casacore.googlecode.com/"
+        >casacore.googlecode.com</a>). Astronomers love pyrap because
+        it makes it easily possible to get their data (observed with
+        radio-astronomical telescopes like LOFAR, ASKAP, and eVLA) in numpy
+        arrays and do basic data inspection and manipulation using the many
+        python packages that are available.</p>
+
+        <p>Boost.Python made it quite easily possible to create converters for
+        the various data types, also for numpy arrays and individual elements
+        of a numpy array. It's nice they work fully recursively. Mapping C++
+        functions to Python was straightforward.</p>
+      </blockquote>
+    </dd>
+
+    <dt><b><a href="http://www.rdkit.org/"
+           >RDKit: Cheminformatics and Machine Learning Software</a></b></dt>
 
     <dd>
-      Rational Discovery provides computational modeling, combinatorial
-      library design and custom software development services to the
-      pharmaceutical, biotech and chemical industries. We do a substantial
-      amount of internal research to develop new approaches for applying
-      machine-learning techniques to solve chemical problems. Because we're a
-      small organization and chemistry is a large and complex field, it is
-      essential that we be able to quickly and easily prototype and test new
-      algorithms.
-
-      <p>For our internal software, we implement core data structures in C
-      and expose them to Python using Boost.Python. Algorithm development is
-      done in Python and then translated to C if required (often it's not).
-      This hybrid development approach not only greatly increases our
-      productivity, but it also allows "non-developers" (people without C
-      experience) to take part in method development. Learning C is a
-      daunting task, but "Python fits your brain." (Thanks to Bruce Eckel for
-      the quote.)</p>
+      A collection of cheminformatics and machine-learning software
+      written in C++ and Python.
     </dd>
   </dl>
 

src/converter/builtin_converters.cpp

@@ -377,7 +377,8 @@ namespace
       static unaryfunc* get_slot(PyObject* obj)
       {
 #if PY_VERSION_HEX >= 0x03000000
-          return (PyUnicode_Check(obj)) ? &py_unicode_as_string_unaryfunc : 0;
+          return (PyUnicode_Check(obj)) ? &py_unicode_as_string_unaryfunc : 
+                  PyBytes_Check(obj) ? &py_object_identity : 0;
 #else
           return (PyString_Check(obj)) ? &obj->ob_type->tp_str : 0;