Add asynchronous ('async', 'media' and conditional 'defer' attributes) loading of remote javascript
and css files within Cantera 2.4.0 Sphinx Documentation to avoid too long render of pages
in case of offline usage of documentation.
This removes Python 2.7 support from the "full" Python module. The
"minimal" Python module (the ctml_writer and ck2cti scripts) and the
SCons build system still support Python 2.7.
Remove 'from __future__' imports, specification of base class
'object', and arguments to 'super()'.
Make use of 'str' being the unicode string type.
Eliminate special cases which require checking the version.
Fix some deprecation warnings issued by legacy capabilities.
Use the Cython module from the Python installation specified by
'python_cmd', rather than the Python installation that is running
SCons. This allows complilation of Cantera for Python versions that
aren't supported by SCons (e.g. Python 3.4).
This only works because the "dot" function is always called with std::vector<T>::iterator as input so that "argument dependent lookup" introduces the std namespace to the function. If the dot function is called like this "dot(v.data(), v.data()+v.size(), v.data())", where "v" is a std::vector and the input are plain pointers, the compiler will not find "inner_product". The same is true for the use of "accumulate".
Recent versions of clang++ warn that instantiation of a templated variable is
required at a certain point where no definition is available. Declaring such a
definition to be available is fine with older versions of clang++ as well, but
causes linker errors with g++, so this change is only applied when using
clang++.
When the nonreactant_orders option was enabled, specifying reactant orders for
species which were not present in the phase previously resulted in out-of-bounds
memory access.
Caused most Matlab flame simulations to fail, e.g. those using
CounterFlowDiffusionFlame.m or flame.m. Fixes regression introduced in c1067aa.
Fixes#554
In this example, a time-dependent mass flow rate function is used to inject a
specific fuel mass into a reactor. This is a more practical use case for this
capability than the fictitious hydrogen radical igniter used in combustor.py.
Discontinuities in heat capacity, in contrast to enthalpy and entropy, are less
of a problem and are not always indicative of problems with a mechanism. As
such, a looser tolerance on this quantity is reasonable.
This threshold eliminates warnings from the nDodecane_Reitz.cti input file included with Cantera.
