Files
MeanField_SERiF/docs/man/man3/serif_equations_Contribution.3

81 lines
5.2 KiB
Groff
Raw Permalink Normal View History

.TH "serif::equations::Contribution< TestOperandT, SupportT >" 3 "Version v0.0.1a" "SERiF" \" -*- nroff -*-
.ad l
.nh
.SH NAME
serif::equations::Contribution< TestOperandT, SupportT > \- A helper struct for authoring equations with fewer lines of boiler-plate\&. The idea here is that every equation must report certain members so that the backend can inspect at compile time\&. An author could certainly write these members themselves, but the \fBContribution\fP struct auto generates those members based on its template parameters\&. Therefore, by inheriting from the \fBContribution\fP struct an equation author can avoid writing those members themselves\&. Note that while this is inheritance it is still a compile-time relation\&. \fBContribution\fP does not add any runtime overhead to the equation\&.
.SH SYNOPSIS
.br
.PP
.PP
\fR#include <contributions\&.hpp>\fP
.PP
Inherited by \fBserif::equations::BarycenterConstraint< Axis, Row >\fP, \fBserif::operators::NodalRequest< Potential >\fP, and \fBserif::equations::BarycenterConstraint< Axis, Row >\fP\&.
.SS "Public Types"
.in +1c
.ti -1c
.RI "using \fBTest\fP = TestOperandT::term_type"
.br
.RI "The test type of the contribution\&. "
.ti -1c
.RI "using \fBTestOperand\fP = TestOperandT"
.br
.RI "The operand type of the contribution (associating a test function and a term) "
.ti -1c
.RI "using \fBSupport\fP = SupportT"
.br
.RI "The domain support of the contribution (which then boils down to the dofs which are supported by the term on that test function) "
.in -1c
.SH "Detailed Description"
.PP
.SS "template<discretization::forms::IsOperand TestOperandT, typename SupportT = discretization::blocks::SupportOf<typename TestOperandT::term_type>>
.br
struct serif::equations::Contribution< TestOperandT, SupportT >"A helper struct for authoring equations with fewer lines of boiler-plate\&. The idea here is that every equation must report certain members so that the backend can inspect at compile time\&. An author could certainly write these members themselves, but the \fBContribution\fP struct auto generates those members based on its template parameters\&. Therefore, by inheriting from the \fBContribution\fP struct an equation author can avoid writing those members themselves\&. Note that while this is inheritance it is still a compile-time relation\&. \fBContribution\fP does not add any runtime overhead to the equation\&.
.PP
\fBTemplate Parameters\fP
.RS 4
\fITestOperandT\fP The test operation type, which must be a valid operand (\fBIsOperand\fP)\&.
.br
\fISupportT\fP The support type, which must be a valid support (\fBIsSupport\fP)\&. This defaults to the support of the test term\&.
.RE
.PP
\fBNote\fP
.RS 4
The support type is, by default, read from the test term\&. There are very few situations where you should need to override this default\&.
.PP
Read this as "residuals are contributed to the row(s) defined by the test operation\&. e\&.g Some density constraint function might contribute to the residuals of the overall system through some density row\&.
.RE
.PP
.SH "Member Typedef Documentation"
.PP
.SS "template<discretization::forms::IsOperand TestOperandT, typename SupportT = discretization::blocks::SupportOf<typename TestOperandT::term_type>> using \fBserif::equations::Contribution\fP< TestOperandT, SupportT >::Support = SupportT"
.PP
The domain support of the contribution (which then boils down to the dofs which are supported by the term on that test function)
.SS "template<discretization::forms::IsOperand TestOperandT, typename SupportT = discretization::blocks::SupportOf<typename TestOperandT::term_type>> using \fBserif::equations::Contribution\fP< TestOperandT, SupportT >::Test = TestOperandT::term_type"
.PP
The test type of the contribution\&. The terminology \fBTest\fP here comes from finite element analysis\&. Specifically, the test type is the type of the test function used in the weak formulation of the equation\&. More concretely, the test type is the type for which the residuals are computed\&. Therefore think of this as the type of the residual row that this contribution contributes to\&. There are a lot of repetitions of the word type here, which I can recognize is a bit confusing\&. Effectively, the way that we have structured much of this code is that we embed particular functions into the type system\&. Therefore when I say "the type of the test function" what I mean is the c++ type that we have defined to represent the test function\&.
.PP
\fBNote\fP
.RS 4
The above comment would probably make any actual writer shake in their boots\&. I'm sure there is a better way to phrase that; however, I am at a loss for the moment\&. The TL;DR here is that since we encode so much information into the type system we use the word type a lot\&. If this makes you rationally upset please email me\&.\&.\&.
.RE
.PP
.SS "template<discretization::forms::IsOperand TestOperandT, typename SupportT = discretization::blocks::SupportOf<typename TestOperandT::term_type>> using \fBserif::equations::Contribution\fP< TestOperandT, SupportT >::TestOperand = TestOperandT"
.PP
The operand type of the contribution (associating a test function and a term)
.SH "Author"
.PP
Generated automatically by Doxygen for SERiF from the source code\&.