functions as arguments to functions.
- To: mathgroup at smc.vnet.net
- Subject: [mg13962] functions as arguments to functions.
- From: sean_ross_at_pl-04m3 at smtpgw1.plk.af.mil
- Date: Fri, 11 Sep 1998 15:06:51 -0400
- Sender: owner-wri-mathgroup at wolfram.com
Lets consider a function 'newfunction' which takes an expression as
an
argument. If I wish to work with the expression as a function
inside
newfunction, I have some options as to what syntax to use as
illustrated below in the private functions g,g2,g3,g4. Note that I
wish to shield the user of the function from pure function
constructs
like most built-in mathematica functions do.
newfunction[f_,x_,x1_]:=Module[{g,g2,g3,g4},
g=(f/.x->#)&;
g2[xx_]=Hold[f]/.x->xx;
g3=Function[x,f];
g4[xx_]:=f/.x->xx;
{g[x1],ReleaseHold[g2[x1]],g3[x1],g4[x1]}]
Here is the expected behavior: All of the above forms are
equivalent
to each other and to the original function when evaluated
numerically
as illustrated below. It turns out that the syntaxes evaluate with
different speeds, but they all arrive at the same place.
In[2]:=
newfunction[Sin[x]^3-x,x,2.5]
Out[2]=
{-2.28565,-2.28565,-2.28565,-2.28565}
In[9]:=
Sin[2.5]^3-2.5
Out[9]=
-2.28565
I am currently working with a function which itself calls functions
which use Solve and other things. For some reason, this
complicated
function does not behave as expected when given to newfunction as
an
argument.
In[7]:=
newfunction[DkKTPsfgII[63 Degree,0. Degree,1.064,x],x,1.32]
Out[7]=
{-0.0466874,-0.0466874,-0.0466874,-0.0466874}
In[8]:=
DkKTPsfgII[63 Degree,0. Degree,1.064,1.32]
Out[8]=
-0.00034538
Notice that here, the four syntaxes agree with each other, but they
are all wrong in the sense that they don't evaluate to the same
number
the original function did. The really weird thing is that the
syntaxes for g2 and g3 work correctly when not placed inside a
Module
construct. The syntaxes for g1 and g4 don't work correctly at all.
I
am attaching a notebook which contains the definition of the
DkKTPsfgII function.
I would appreciate any suggestions to:
--alternate syntaxes for passing functions as arguments.
--any insights as to why this particular function(DkKTPsfgII) is
pathological to mathematica and how I might change it.
--any workarounds.
Please respond directly to me: rosss at plk.af.mil or
seanross at worldnet.att.net as I don't subscribe to the mathgroup any
more. Also, please bear in mind that simple examples are
irrelevant
to this issue. All four syntaxes work for simple functions. They
don't work for really complicated ones. I have a dozen more
similarly
pathological functions, but am not including them so as not to
clutter
up the issue.
Thanks, Sean Ross
(***********************************************************************
Mathematica-Compatible Notebook
This notebook can be used on any computer system with Mathematica 3.0,
MathReader 3.0, or any compatible application. The data for the
notebook starts with the line of stars above.
To get the notebook into a Mathematica-compatible application, do one of
the following:
* Save the data starting with the line of stars above into a file
with a name ending in .nb, then open the file inside the application;
* Copy the data starting with the line of stars above to the
clipboard, then use the Paste menu command inside the application.
Data for notebooks contains only printable 7-bit ASCII and can be sent
directly in email or through ftp in text mode. Newlines can be CR, LF
or CRLF (Unix, Macintosh or MS-DOS style).
NOTE: If you modify the data for this notebook not in a Mathematica-
compatible application, you must delete the line below containing the
word CacheID, otherwise Mathematica-compatible applications may try to
use invalid cache data.
For more information on notebooks and Mathematica-compatible
applications, contact Wolfram Research:
web: http://www.wolfram.com
email: info at wolfram.com
phone: +1-217-398-0700 (U.S.)
Notebook reader applications are available free of charge from Wolfram
Research.
***********************************************************************)
(*CacheID: 232*)
(*NotebookFileLineBreakTest
NotebookFileLineBreakTest*)
(*NotebookOptionsPosition[ 7945, 172]*)
(*NotebookOutlinePosition[ 8594, 195]*) (*
CellTagsIndexPosition[ 8550, 191]*) (*WindowFrame->Normal*)
Notebook[{
Cell[TextData[{
"KTP is a biaxial crystal, so I first must work out the general form
for \ the percieved index of refraction in a biaxial crystal. I start
with Yariv, \ chapter 4. First, we solve the vector equation ",
Cell[BoxData[
\(TraditionalForm
\`\(k\&\[RightVector]\) x\ \(k\&\[RightVector]\) x\
E\&\[RightVector] =
\(-\[Omega]\^2\) \(\[Mu]\_0\) D\&\[RightVector]\)]],
" (1). It can be written as a sort of eigenvector problem and a
quadratic \ solution for the effective index of refraction obtained.
This is contained \ in the nextKTP function, which calls on the next
function for the algebraic \ form of the determinant of (1) and on
nxKTP etc for the indices along the \ principle axes. The solution of
the determinant of (1) is quadratic in ",
Cell[BoxData[
\(TraditionalForm\`n\^2\)]],
", so there are always two answers returned. These correspond to
preferred \ orthogonal axes for the polarization direction." }],
"Text"],
Cell[CellGroupData[{
Cell["\<\
index of refraction as a function of propagation direction in a biaxial
\ crystal (KTP)\
\>", "Section"],
Cell[BoxData[
RowBox[{
RowBox[{
\(Clear[kmatrix, nmatrix, nextKTP, nxKTP, nyKTP, nzKTP, abs,
unitv]\),
";", "\n", \(abs[v : {__}] := Sqrt[v . v]\), ";", "\n",
\(unitv[v : {__}] := v/abs[v]\), ";", "\n",
RowBox[{\(kmatrix[\[Theta]_, \[Phi]_, \[Lambda]_]\), ":=",
RowBox[{"(",
TagBox[
RowBox[{
RowBox[{"(", GridBox[{
{
\(\(-n\^2\)\
\((Cos[\[Theta]]\^2 +
Sin[\[Theta]]\^2\ Sin[\[Phi]]\^2)\) +
n\_x\%2\),
\(n\^2\ Cos[\[Phi]]\ Sin[\[Theta]]\^2\
Sin[\[Phi]]\),
\(n\^2\ Cos[\[Theta]]\ Cos[\[Phi]]\
Sin[\[Theta]]\)},
{\(n\^2\ Cos[\[Phi]]\ Sin[\[Theta]]\^2\
Sin[\[Phi]]\),
\(\(-n\^2\)\
\((Cos[\[Theta]]\^2 +
Cos[\[Phi]]\^2\ Sin[\[Theta]]\^2)\) +
n\_y\%2\),
\(n\^2\ Cos[\[Theta]]\ Sin[\[Theta]]\
Sin[\[Phi]]\)},
{\(n\^2\ Cos[\[Theta]]\ Cos[\[Phi]]\
Sin[\[Theta]]\),
\(n\^2\ Cos[\[Theta]]\ Sin[\[Theta]]\
Sin[\[Phi]]\),
\(\(-n\^2\)\ Sin[\[Theta]]\^2 + n\_z\%2\)}
}], ")"}], "/.",
\({n\^2 -> nn, n\^4 -> nn\^2, n\_x -> nxKTP[\[Lambda]],
n\_y -> nyKTP[\[Lambda]], n\_z ->
nzKTP[\[Lambda]]}\)}],
(MatrixForm[ #]&)], ")"}]}], ";", "\n",
\( (*a\ symbolic\ determinant\ is\ needed\ here . \ \ If\ you\
take\ a
\ numerical\ one, \
small\ cubic\ terms\ creep\ in\ for\ values\ of\ \[Theta]\
and\
\[Phi]\ near, \
but\ on\ on\ the\ principle\ axes\ and\ give\ bad\ \(answers .
\)*)
\), "\n", "\n",
\(kdeterminant[\[Theta]_, \[Phi]_, \[Lambda]_] :=
\((n\_x\%2\ n\_y\%2\ n\_z\%2 +
n\^4\ \((
Cos[\[Theta]]\^2\ Cos[\[Phi]]\^2\ Sin[\[Theta]]\^2\
n\_x\%2 + Cos[\[Phi]]\^4\ Sin[\[Theta]]\^4\
n\_x\%2 +
Cos[\[Phi]]\^2\ Sin[\[Theta]]\^4\ Sin[\[Phi]]\^2\
n\_x\%2 +
Cos[\[Theta]]\^2\ Sin[\[Theta]]\^2\
Sin[\[Phi]]\^2\
n\_y\%2 +
Cos[\[Phi]]\^2\ Sin[\[Theta]]\^4\ Sin[\[Phi]]\^2\
n\_y\%2 + Sin[\[Theta]]\^4\ Sin[\[Phi]]\^4\
n\_y\%2 +
Cos[\[Theta]]\^4\ n\_z\%2 +
Cos[\[Theta]]\^2\ Cos[\[Phi]]\^2\
Sin[\[Theta]]\^2\
n\_z\%2 +
Cos[\[Theta]]\^2\ Sin[\[Theta]]\^2\
Sin[\[Phi]]\^2\
n\_z\%2)\) +
n\^2\ \((
\(-Cos[\[Phi]]\^2\)\ Sin[\[Theta]]\^2\ n\_x\%2\
n\_y\%2 -
Sin[\[Theta]]\^2\ Sin[\[Phi]]\^2\ n\_x\%2\ n\_y\%2
-
Cos[\[Theta]]\^2\ n\_x\%2\ n\_z\%2 -
Cos[\[Phi]]\^2\ Sin[\[Theta]]\^2\ n\_x\%2\ n\_z\%2
-
Cos[\[Theta]]\^2\ n\_y\%2\ n\_z\%2 -
Sin[\[Theta]]\^2\ Sin[\[Phi]]\^2\ n\_y\%2\
n\_z\%2)\))
\) /. {n\^2 -> nn, n\^4 -> nn\^2, n\_x ->
nxKTP[\[Lambda]],
n\_y -> nyKTP[\[Lambda]], n\_z -> nzKTP[\[Lambda]]}\),
";",
"\n", "\n",
\(nmatrix[\[Theta]_, \[Phi]_, \[Lambda]_, j_: 1] :=
Module[{mat},
mat = kmatrix[\[Theta], \[Phi], \[Lambda]] /.
nn -> \(nextKTP[\[Theta], \[Phi],
\[Lambda]]\)[\([j]\)]^2; \n
\tChop[mat/Max[Chop[mat]]]]\), ";", "\n", "\n", "\n",
\(nextKTP[\[Theta]_, \[Phi]_, \[Lambda]_] := \t\t\n\t\t
Sqrt[Flatten[
nn /. Solve[
kdeterminant[\[Theta], \[Phi], \[Lambda]] == 0,
{nn}]]]\),
";", "\n", "\n",
\(nxKTP[\[Lambda]_] :=
Module[{aa = 2.11460, bb = 0.89188, cc = 0.20861, dd =
0.01320},
Sqrt[aa + bb\/\(1 - \((cc/\[Lambda])\)\^2\) - dd\
\[Lambda]\^2]]
\), ";", "\n",
\(nyKTP[\[Lambda]_] :=
Module[{aa = 2.15180, bb = 0.87862, cc = 0.21801, dd =
0.01327},
Sqrt[aa + bb\/\(1 - \((cc/\[Lambda])\)\^2\) - dd\
\[Lambda]\^2]]
\), ";", "\n",
\(nzKTP[\[Lambda]_] :=
Module[{aa = 2.31360, bb = 1.00012, cc = 0.23831, dd =
0.01679},
Sqrt[aa + bb\/\(1 - \((cc/\[Lambda])\)\^2\) - dd\
\[Lambda]\^2]]
\), ";"}], "\n"}]], "Input"] }, Open ]],
Cell[BoxData[
\(\(\[CapitalDelta]kKTPsfgII[\[Theta]_, \[Phi]_, \[Lambda]1_,
\[Lambda]2_] :=
Module[{\[Lambda]sum},
\[Lambda]sum = \((1\/\[Lambda]1 + 1\/\[Lambda]2)\)\^\(-1\);
\(nextKTP[\[Theta], \[Phi],
\[Lambda]sum]\)[\([1]\)]\/\[Lambda]sum -
\(nextKTP[\[Theta], \[Phi], \[Lambda]1]\)[\([2]\)]\/\[Lambda]1
-
\(nextKTP[\[Theta], \[Phi],
\[Lambda]2]\)[\([1]\)]\/\[Lambda]2];
\)\)], "Input"]
},
FrontEndVersion->"Microsoft Windows 3.0", ScreenRectangle->{{0, 800},
{0, 572}}, WindowSize->{651, 434},
WindowMargins->{{Automatic, 46}, {Automatic, 5}} ]
(***********************************************************************
Cached data follows. If you edit this Notebook file directly, not
using Mathematica, you must remove the line containing CacheID at the
top of the file. The cache data will then be recreated when you save
this file from within Mathematica.
***********************************************************************)
(*CellTagsOutline
CellTagsIndex->{}
*)
(*CellTagsIndex
CellTagsIndex->{}
*)
(*NotebookFileOutline
Notebook[{
Cell[1709, 49, 980, 17, 128, "Text"],
Cell[CellGroupData[{
Cell[2714, 70, 114, 3, 74, "Section"], Cell[2831, 75, 4641, 84, 890,
"Input"] }, Open ]],
Cell[7487, 162, 454, 8, 77, "Input"] }
]
*)
(***********************************************************************
End of Mathematica Notebook file.
***********************************************************************)