Mercurial > hg > octave-lyh
annotate doc/interpreter/dynamic.txi @ 10791:3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
interpreter/doccheck: New directory for spelling/grammar scripts.
interpreter/doccheck/README: Instructions for using scripts.
interpreter/doccheck/spellcheck: Script to spellcheck a Texinfo file.
interpreter/doccheck/aspell.conf: GNU Aspell configuration file for
Octave documentation.
interpreter/doccheck/aspell-octave.en.pws: Private Aspell dictionary.
interpreter/doccheck/add_to_aspell_dict: Script to add new
Octave-specific words to
private Aspell dictionary.
interpreter/octave.texi: New @nospell macro which forces Aspell
to ignore the word marked by the macro.
interpreter/mk_doc_cache.m: Skip new @nospell macro when building
doc_cache.
| author | Rik <octave@nomad.inbox5.com> |
|---|---|
| date | Sat, 17 Jul 2010 19:53:01 -0700 |
| parents | 8d20fb66a0dc |
| children | 322f43e0e170 |
| rev | line source |
|---|---|
| 8920 | 1 @c Copyright (C) 2007, 2008, 2009 John W. Eaton and David Bateman |
| 7018 | 2 @c Copyright (C) 2007 Paul Thomas and Christoph Spiel |
| 3 @c | |
| 4 @c This file is part of Octave. | |
| 5 @c | |
| 6 @c Octave is free software; you can redistribute it and/or modify it | |
| 7 @c under the terms of the GNU General Public License as published by the | |
| 8 @c Free Software Foundation; either version 3 of the License, or (at | |
| 9 @c your option) any later version. | |
| 10 @c | |
| 11 @c Octave is distributed in the hope that it will be useful, but WITHOUT | |
| 12 @c ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or | |
| 13 @c FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License | |
| 14 @c for more details. | |
| 15 @c | |
| 16 @c You should have received a copy of the GNU General Public License | |
| 17 @c along with Octave; see the file COPYING. If not, see | |
| 18 @c <http://www.gnu.org/licenses/>. | |
| 6578 | 19 |
| 6569 | 20 @node Dynamically Linked Functions |
| 21 @appendix Dynamically Linked Functions | |
| 22 @cindex dynamic-linking | |
| 23 | |
| 24 Octave has the possibility of including compiled code as dynamically | |
| 25 linked extensions and then using these extensions as if they were part | |
| 8828 | 26 of Octave itself. Octave can call C++ code |
| 6569 | 27 through its native oct-file interface or C code through its mex |
| 6571 | 28 interface. It can also indirectly call functions written in any other |
| 29 language through a simple wrapper. The reasons to write code in a | |
| 6569 | 30 compiled language might be either to link to an existing piece of code |
| 31 and allow it to be used within Octave, or to allow improved performance | |
| 32 for key pieces of code. | |
| 33 | |
| 34 Before going further, you should first determine if you really need to | |
| 6571 | 35 use dynamically linked functions at all. Before proceeding with writing |
| 6569 | 36 any dynamically linked function to improve performance you should |
| 37 address ask yourself | |
| 38 | |
| 39 @itemize @bullet | |
| 40 @item | |
| 6939 | 41 Can I get the same functionality using the Octave scripting language only? |
| 6569 | 42 @item |
| 6572 | 43 Is it thoroughly optimized Octave code? Vectorization of Octave code, |
| 6569 | 44 doesn't just make it concise, it generally significantly improves its |
| 6571 | 45 performance. Above all, if loops must be used, make sure that the |
| 6569 | 46 allocation of space for variables takes place outside the loops using an |
| 8828 | 47 assignment to a matrix of the right size, or zeros. |
| 6569 | 48 @item |
| 49 Does it make as much use as possible of existing built-in library | |
| 6572 | 50 routines? These are highly optimized and many do not carry the overhead |
| 6569 | 51 of being interpreted. |
| 52 @item | |
| 53 Does writing a dynamically linked function represent useful investment | |
| 54 of your time, relative to staying in Octave? | |
| 55 @end itemize | |
| 56 | |
| 8475 | 57 Also, as oct- and mex-files are dynamically linked to Octave, they |
|
8347
fa78cb8d8a5c
corrections for typos
Brian Gough<bjg@network-theory.co.uk>
parents:
8097
diff
changeset
|
58 introduce the possibility of Octave crashing due to errors in |
| 7001 | 59 the user code. For example a segmentation violation in the user's code |
| 6569 | 60 will cause Octave to abort. |
| 61 | |
| 62 @menu | |
| 6572 | 63 * Oct-Files:: |
| 64 * Mex-Files:: | |
| 65 * Standalone Programs:: | |
| 6569 | 66 @end menu |
| 67 | |
| 68 @node Oct-Files | |
| 69 @section Oct-Files | |
| 70 @cindex oct-files | |
| 71 @cindex mkoctfile | |
| 72 @cindex oct | |
| 73 | |
| 74 @menu | |
| 6572 | 75 * Getting Started with Oct-Files:: |
| 76 * Matrices and Arrays in Oct-Files:: | |
| 77 * Character Strings in Oct-Files:: | |
| 78 * Cell Arrays in Oct-Files:: | |
| 79 * Structures in Oct-Files:: | |
| 80 * Sparse Matrices in Oct-Files:: | |
| 81 * Accessing Global Variables in Oct-Files:: | |
| 82 * Calling Octave Functions from Oct-Files:: | |
| 83 * Calling External Code from Oct-Files:: | |
| 84 * Allocating Local Memory in Oct-Files:: | |
| 85 * Input Parameter Checking in Oct-Files:: | |
| 86 * Exception and Error Handling in Oct-Files:: | |
| 87 * Documentation and Test of Oct-Files:: | |
| 6593 | 88 @c * Application Programming Interface for Oct-Files:: |
| 6569 | 89 @end menu |
| 90 | |
| 91 @node Getting Started with Oct-Files | |
| 92 @subsection Getting Started with Oct-Files | |
| 93 | |
| 94 The basic command to build oct-files is @code{mkoctfile} and it can be | |
| 95 call from within octave or from the command line. | |
| 96 | |
| 97 @DOCSTRING(mkoctfile) | |
| 98 | |
| 99 Consider the short example | |
| 100 | |
| 9906 | 101 @example |
| 102 @group | |
| 103 @EXAMPLEFILE(helloworld.cc) | |
| 104 @end group | |
| 105 @end example | |
| 6569 | 106 |
| 107 This example although short introduces the basics of writing a C++ | |
| 6571 | 108 function that can be dynamically linked to Octave. The easiest way to |
| 6569 | 109 make available most of the definitions that might be necessary for an |
| 110 oct-file in Octave is to use the @code{#include <octave/oct.h>} | |
| 6571 | 111 header. |
| 6569 | 112 |
| 113 The macro that defines the entry point into the dynamically loaded | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
114 function is @w{@code{DEFUN_DLD}}. This macro takes four arguments, these being |
| 6569 | 115 |
| 116 @enumerate 1 | |
| 6571 | 117 @item The function name as it will be seen in Octave, |
| 6572 | 118 @item The list of arguments to the function of type @code{octave_value_list}, |
| 6569 | 119 @item The number of output arguments, which can and often is omitted if |
| 120 not used, and | |
| 121 @item The string that will be seen as the help text of the function. | |
| 122 @end enumerate | |
| 123 | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
124 The return type of functions defined with @w{@code{DEFUN_DLD}} is always |
| 6572 | 125 @code{octave_value_list}. |
| 6569 | 126 |
| 127 There are a couple of important considerations in the choice of function | |
| 6571 | 128 name. Firstly, it must be a valid Octave function name and so must be a |
| 6569 | 129 sequence of letters, digits and underscores, not starting with a |
| 6571 | 130 digit. Secondly, as Octave uses the function name to define the filename |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
131 it attempts to find the function in, the function name in the @w{@code{DEFUN_DLD}} |
| 6571 | 132 macro must match the filename of the oct-file. Therefore, the above |
| 6572 | 133 function should be in a file @file{helloworld.cc}, and it should be |
| 134 compiled to an oct-file using the command | |
| 6569 | 135 |
| 136 @example | |
| 137 mkoctfile helloworld.cc | |
| 138 @end example | |
| 139 | |
| 8828 | 140 This will create a file called @file{helloworld.oct}, that is the compiled |
| 6571 | 141 version of the function. It should be noted that it is perfectly |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
142 acceptable to have more than one @w{@code{DEFUN_DLD}} function in a source |
| 6571 | 143 file. However, there must either be a symbolic link to the oct-file for |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
144 each of the functions defined in the source code with the @w{@code{DEFUN_DLD}} |
| 6569 | 145 macro or the autoload (@ref{Function Files}) function should be used. |
| 146 | |
| 147 The rest of this function then shows how to find the number of input | |
| 148 arguments, how to print through the octave pager, and return from the | |
| 6571 | 149 function. After compiling this function as above, an example of its use |
| 6569 | 150 is |
| 151 | |
| 152 @example | |
| 153 @group | |
| 6572 | 154 helloworld (1, 2, 3) |
| 155 @print{} Hello World has 3 input arguments and 0 output arguments. | |
| 6569 | 156 @end group |
| 157 @end example | |
| 158 | |
| 159 @node Matrices and Arrays in Oct-Files | |
| 160 @subsection Matrices and Arrays in Oct-Files | |
| 161 | |
| 162 Octave supports a number of different array and matrix classes, the | |
| 6571 | 163 majority of which are based on the Array class. The exception is the |
| 164 sparse matrix types discussed separately below. There are three basic | |
| 165 matrix types | |
| 6569 | 166 |
| 6572 | 167 @table @code |
| 6569 | 168 @item Matrix |
| 169 A double precision matrix class defined in dMatrix.h, | |
| 170 @item ComplexMatrix | |
| 171 A complex matrix class defined in CMatrix.h, and | |
| 172 @item BoolMatrix | |
| 173 A boolean matrix class defined in boolMatrix.h. | |
| 174 @end table | |
| 175 | |
| 6571 | 176 These are the basic two-dimensional matrix types of octave. In |
| 6569 | 177 additional there are a number of multi-dimensional array types, these |
| 178 being | |
| 179 | |
| 6572 | 180 @table @code |
| 6569 | 181 @item NDArray |
| 6572 | 182 A double precision array class defined in @file{dNDArray.h} |
| 6569 | 183 @item ComplexNDarray |
| 6572 | 184 A complex array class defined in @file{CNDArray.h} |
| 6569 | 185 @item boolNDArray |
| 6572 | 186 A boolean array class defined in @file{boolNDArray.h} |
| 187 @item int8NDArray | |
| 188 @itemx int16NDArray | |
| 189 @itemx int32NDArray | |
| 190 @itemx int64NDArray | |
| 191 8, 16, 32 and 64-bit signed array classes defined in | |
| 192 @file{int8NDArray.h}, @file{int16NDArray.h}, etc. | |
| 193 @item uint8NDArray | |
| 194 @itemx uint16NDArray | |
| 195 @itemx uint32NDArray | |
| 196 @itemx uint64NDArray | |
| 197 8, 16, 32 and 64-bit unsigned array classes defined in | |
| 198 @file{uint8NDArray.h}, @file{uint16NDArray.h}, etc. | |
| 6569 | 199 @end table |
| 200 | |
| 201 There are several basic means of constructing matrices of | |
| 6572 | 202 multi-dimensional arrays. Considering the @code{Matrix} type as an |
| 203 example | |
| 6569 | 204 |
| 205 @itemize @bullet | |
| 6571 | 206 @item |
| 207 We can create an empty matrix or array with the empty constructor. For | |
| 6569 | 208 example |
| 209 | |
| 210 @example | |
| 211 Matrix a; | |
| 212 @end example | |
| 213 | |
| 214 This can be used on all matrix and array types | |
| 6571 | 215 @item |
| 216 Define the dimensions of the matrix or array with a dim_vector. For | |
| 6569 | 217 example |
| 218 | |
| 219 @example | |
| 220 @group | |
| 6572 | 221 dim_vector dv (2); |
| 6569 | 222 dv(0) = 2; dv(1) = 2; |
| 6572 | 223 Matrix a (dv); |
| 6569 | 224 @end group |
| 225 @end example | |
| 226 | |
| 227 This can be used on all matrix and array types | |
| 228 @item | |
| 6572 | 229 Define the number of rows and columns in the matrix. For example |
| 6569 | 230 |
| 231 @example | |
| 6572 | 232 Matrix a (2, 2) |
| 6569 | 233 @end example |
| 234 | |
| 235 However, this constructor can only be used with the matrix types. | |
| 236 @end itemize | |
| 237 | |
| 238 These types all share a number of basic methods and operators, a | |
| 239 selection of which include | |
| 240 | |
| 6572 | 241 @deftypefn Method T& {operator ()} (octave_idx_type) |
| 242 @deftypefnx Method T& elem (octave_idx_type) | |
| 243 The @code{()} operator or @code{elem} method allow the values of the | |
| 244 matrix or array to be read or set. These can take a single argument, | |
| 245 which is of type @code{octave_idx_type}, that is the index into the matrix or | |
| 6571 | 246 array. Additionally, the matrix type allows two argument versions of the |
| 6572 | 247 @code{()} operator and elem method, giving the row and column index of the |
| 6569 | 248 value to obtain or set. |
| 6572 | 249 @end deftypefn |
| 6569 | 250 |
| 7001 | 251 Note that these functions do significant error checking and so in some |
| 252 circumstances the user might prefer to access the data of the array or | |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
253 matrix directly through the @nospell{fortran_vec} method discussed below. |
| 6572 | 254 |
| 255 @deftypefn Method octave_idx_type nelem (void) const | |
| 6569 | 256 The total number of elements in the matrix or array. |
| 6572 | 257 @end deftypefn |
| 258 | |
| 259 @deftypefn Method size_t byte_size (void) const | |
| 6569 | 260 The number of bytes used to store the matrix or array. |
| 6572 | 261 @end deftypefn |
| 262 | |
| 263 @deftypefn Method dim_vector dims (void) const | |
| 6569 | 264 The dimensions of the matrix or array in value of type dim_vector. |
| 6572 | 265 @end deftypefn |
| 266 | |
| 267 @deftypefn Method void resize (const dim_vector&) | |
| 268 A method taking either an argument of type @code{dim_vector}, or in the | |
| 269 case of a matrix two arguments of type @code{octave_idx_type} defining | |
| 270 the number of rows and columns in the matrix. | |
| 271 @end deftypefn | |
| 272 | |
| 273 @deftypefn Method T* fortran_vec (void) | |
| 6569 | 274 This method returns a pointer to the underlying data of the matrix or a |
| 275 array so that it can be manipulated directly, either within Octave or by | |
| 276 an external library. | |
| 6572 | 277 @end deftypefn |
| 6569 | 278 |
| 6572 | 279 Operators such an @code{+}, @code{-}, or @code{*} can be used on the |
| 280 majority of the above types. In addition there are a number of methods | |
| 281 that are of interest only for matrices such as @code{transpose}, | |
| 282 @code{hermitian}, @code{solve}, etc. | |
| 6569 | 283 |
| 284 The typical way to extract a matrix or array from the input arguments of | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
285 @w{@code{DEFUN_DLD}} function is as follows |
| 6569 | 286 |
| 9906 | 287 @example |
| 288 @group | |
| 289 @EXAMPLEFILE(addtwomatrices.cc) | |
| 290 @end group | |
| 291 @end example | |
| 6569 | 292 |
| 293 To avoid segmentation faults causing Octave to abort, this function | |
| 294 explicitly checks that there are sufficient arguments available before | |
| 6571 | 295 accessing these arguments. It then obtains two multi-dimensional arrays |
| 6572 | 296 of type @code{NDArray} and adds these together. Note that the array_value |
| 297 method is called without using the @code{is_matrix_type} type, and instead the | |
| 6571 | 298 error_state is checked before returning @code{A + B}. The reason to |
| 6569 | 299 prefer this is that the arguments might be a type that is not an |
| 6572 | 300 @code{NDArray}, but it would make sense to convert it to one. The |
| 301 @code{array_value} method allows this conversion to be performed | |
| 302 transparently if possible, and sets @code{error_state} if it is not. | |
| 6569 | 303 |
| 6572 | 304 @code{A + B}, operating on two @code{NDArray}'s returns an |
| 305 @code{NDArray}, which is cast to an @code{octave_value} on the return | |
| 306 from the function. An example of the use of this demonstration function | |
| 307 is | |
| 6569 | 308 |
| 309 @example | |
| 310 @group | |
| 6572 | 311 addtwomatrices (ones (2, 2), ones (2, 2)) |
| 6569 | 312 @result{} 2 2 |
| 313 2 2 | |
| 314 @end group | |
| 315 @end example | |
| 316 | |
| 6572 | 317 A list of the basic @code{Matrix} and @code{Array} types, the methods to |
| 318 extract these from an @code{octave_value} and the associated header is | |
| 319 listed below. | |
| 6569 | 320 |
| 321 @multitable @columnfractions .3 .4 .3 | |
| 6572 | 322 @item @code{RowVector} @tab @code{row_vector_value} @tab @file{dRowVector.h} |
| 323 @item @code{ComplexRowVector} @tab @code{complex_row_vector_value} @tab @file{CRowVector.h} | |
| 324 @item @code{ColumnVector} @tab @code{column_vector_value} @tab @file{dColVector.h} | |
| 325 @item @code{ComplexColumnVector} @tab @code{complex_column_vector_value} @tab @file{CColVector.h} | |
| 326 @item @code{Matrix} @tab @code{matrix_value} @tab @file{dMatrix.h} | |
| 327 @item @code{ComplexMatrix} @tab @code{complex_matrix_value} @tab @file{CMatrix.h} | |
| 328 @item @code{boolMatrix} @tab @code{bool_matrix_value} @tab @file{boolMatrix.h} | |
| 329 @item @code{charMatrix} @tab @code{char_matrix_value} @tab @file{chMatrix.h} | |
| 330 @item @code{NDArray} @tab @code{array_value} @tab @file{dNDArray.h} | |
| 331 @item @code{ComplexNDArray} @tab @code{complex_array_value} @tab @file{CNDArray.h} | |
| 332 @item @code{boolNDArray} @tab @code{bool_array_value} @tab @file{boolNDArray.h} | |
| 333 @item @code{charNDArray} @tab @code{char_array_value} @tab @file{charNDArray.h} | |
| 334 @item @code{int8NDArray} @tab @code{int8_array_value} @tab @file{int8NDArray.h} | |
| 335 @item @code{int16NDArray} @tab @code{int16_array_value} @tab @file{int16NDArray.h} | |
| 336 @item @code{int32NDArray} @tab @code{int32_array_value} @tab @file{int32NDArray.h} | |
| 337 @item @code{int64NDArray} @tab @code{int64_array_value} @tab @file{int64NDArray.h} | |
| 338 @item @code{uint8NDArray} @tab @code{uint8_array_value} @tab @file{uint8NDArray.h} | |
| 339 @item @code{uint16NDArray} @tab @code{uint16_array_value} @tab @file{uint16NDArray.h} | |
| 340 @item @code{uint32NDArray} @tab @code{uint32_array_value} @tab @file{uint32NDArray.h} | |
| 341 @item @code{uint64NDArray} @tab @code{uint64_array_value} @tab @file{uint64NDArray.h} | |
| 6569 | 342 @end multitable |
| 343 | |
| 6572 | 344 @node Character Strings in Oct-Files |
| 345 @subsection Character Strings in Oct-Files | |
| 346 | |
| 347 In Octave a character string is just a special @code{Array} class. | |
| 348 Consider the example | |
| 349 | |
| 9906 | 350 @example |
| 351 @EXAMPLEFILE(stringdemo.cc) | |
| 352 @end example | |
| 6572 | 353 |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
354 An example of the use of this function is |
| 6572 | 355 |
| 356 @example | |
| 357 @group | |
| 358 s0 = ["First String"; "Second String"]; | |
| 359 [s1,s2] = stringdemo (s0) | |
| 360 @result{} s1 = Second String | |
| 361 First String | |
| 362 | |
| 363 @result{} s2 = First String | |
| 364 Second String | |
| 365 | |
| 366 typeinfo (s2) | |
| 367 @result{} sq_string | |
| 368 typeinfo (s1) | |
| 369 @result{} string | |
| 370 @end group | |
| 371 @end example | |
| 372 | |
| 373 One additional complication of strings in Octave is the difference | |
| 374 between single quoted and double quoted strings. To find out if an | |
| 375 @code{octave_value} contains a single or double quoted string an example is | |
| 376 | |
| 377 @example | |
| 378 @group | |
| 379 if (args(0).is_sq_string ()) | |
| 7081 | 380 octave_stdout << |
| 381 "First argument is a singularly quoted string\n"; | |
| 6572 | 382 else if (args(0).is_dq_string ()) |
| 7081 | 383 octave_stdout << |
| 384 "First argument is a doubly quoted string\n"; | |
| 6572 | 385 @end group |
| 386 @end example | |
| 387 | |
| 388 Note however, that both types of strings are represented by the | |
| 389 @code{charNDArray} type, and so when assigning to an | |
| 390 @code{octave_value}, the type of string should be specified. For example | |
| 391 | |
| 392 @example | |
| 393 @group | |
| 394 octave_value_list retval; | |
| 395 charNDArray c; | |
| 396 @dots{} | |
| 6577 | 397 // Create single quoted string |
| 398 retval(1) = octave_value (ch, true, '\''); | |
| 399 | |
| 400 // Create a double quoted string | |
| 401 retval(0) = octave_value (ch, true); | |
| 6572 | 402 @end group |
| 403 @end example | |
| 404 | |
| 405 @node Cell Arrays in Oct-Files | |
| 406 @subsection Cell Arrays in Oct-Files | |
| 407 | |
| 7001 | 408 Octave's cell type is equally accessible within oct-files. A cell |
| 6572 | 409 array is just an array of @code{octave_value}s, and so each element of the cell |
| 410 array can then be treated just like any other @code{octave_value}. A simple | |
| 411 example is | |
| 412 | |
| 9906 | 413 @example |
| 414 @group | |
| 415 @EXAMPLEFILE(celldemo.cc) | |
| 416 @end group | |
| 417 @end example | |
| 6572 | 418 |
| 419 Note that cell arrays are used less often in standard oct-files and so | |
| 420 the @file{Cell.h} header file must be explicitly included. The rest of this | |
| 421 example extracts the @code{octave_value}s one by one from the cell array and | |
| 422 returns be as individual return arguments. For example consider | |
| 423 | |
| 424 @example | |
| 425 @group | |
| 426 [b1, b2, b3] = celldemo (@{1, [1, 2], "test"@}) | |
| 427 @result{} | |
| 428 b1 = 1 | |
| 429 b2 = | |
| 430 | |
| 431 1 2 | |
| 432 | |
| 433 b3 = test | |
| 434 @end group | |
| 435 @end example | |
| 436 | |
| 437 @node Structures in Oct-Files | |
| 438 @subsection Structures in Oct-Files | |
| 439 | |
| 440 A structure in Octave is map between a number of fields represented and | |
| 441 their values. The Standard Template Library @code{map} class is used, | |
| 442 with the pair consisting of a @code{std::string} and an octave | |
| 443 @code{Cell} variable. | |
| 444 | |
| 445 A simple example demonstrating the use of structures within oct-files is | |
| 446 | |
| 9906 | 447 @example |
| 448 @EXAMPLEFILE(structdemo.cc) | |
| 449 @end example | |
| 6572 | 450 |
| 451 An example of its use is | |
| 452 | |
| 453 @example | |
| 454 @group | |
| 455 x.a = 1; x.b = "test"; x.c = [1, 2]; | |
| 456 structdemo (x, "b") | |
| 457 @result{} selected = test | |
| 458 @end group | |
| 459 @end example | |
| 460 | |
| 7001 | 461 The commented code above demonstrates how to iterate over all of the |
| 6572 | 462 fields of the structure, where as the following code demonstrates finding |
| 463 a particular field in a more concise manner. | |
| 464 | |
| 465 As can be seen the @code{contents} method of the @code{Octave_map} class | |
| 466 returns a @code{Cell} which allows structure arrays to be represented. | |
| 467 Therefore, to obtain the underlying @code{octave_value} we write | |
| 468 | |
| 469 @example | |
| 470 octave_value tmp = arg0.contents (p1) (0); | |
| 471 @end example | |
| 472 | |
| 6593 | 473 where the trailing (0) is the () operator on the @code{Cell} object. We |
| 474 can equally iterate of the elements of the Cell array to address the | |
| 475 elements of the structure array. | |
| 6572 | 476 |
| 477 @node Sparse Matrices in Oct-Files | |
| 478 @subsection Sparse Matrices in Oct-Files | |
| 6569 | 479 |
| 480 There are three classes of sparse objects that are of interest to the | |
| 481 user. | |
| 482 | |
| 6572 | 483 @table @code |
| 6569 | 484 @item SparseMatrix |
| 485 A double precision sparse matrix class | |
| 486 @item SparseComplexMatrix | |
| 487 A complex sparse matrix class | |
| 488 @item SparseBoolMatrix | |
| 489 A boolean sparse matrix class | |
| 490 @end table | |
| 491 | |
| 492 All of these classes inherit from the @code{Sparse<T>} template class, | |
| 6571 | 493 and so all have similar capabilities and usage. The @code{Sparse<T>} |
| 6569 | 494 class was based on Octave @code{Array<T>} class, and so users familiar |
| 6572 | 495 with Octave's @code{Array} classes will be comfortable with the use of |
| 6569 | 496 the sparse classes. |
| 497 | |
| 498 The sparse classes will not be entirely described in this section, due | |
| 6572 | 499 to their similarity with the existing @code{Array} classes. However, |
| 500 there are a few differences due the different nature of sparse objects, | |
| 501 and these will be described. Firstly, although it is fundamentally | |
| 502 possible to have N-dimensional sparse objects, the Octave sparse classes do | |
| 6571 | 503 not allow them at this time. So all operations of the sparse classes |
| 6569 | 504 must be 2-dimensional. This means that in fact @code{SparseMatrix} is |
| 505 similar to Octave's @code{Matrix} class rather than its | |
| 506 @code{NDArray} class. | |
| 507 | |
| 508 @menu | |
| 6572 | 509 * Array and Sparse Differences:: |
| 510 * Creating Sparse Matrices in Oct-Files:: | |
| 511 * Using Sparse Matrices in Oct-Files:: | |
| 6569 | 512 @end menu |
| 513 | |
| 6572 | 514 @node Array and Sparse Differences |
| 6569 | 515 @subsubsection The Differences between the Array and Sparse Classes |
| 516 | |
| 517 The number of elements in a sparse matrix is considered to be the number | |
| 6571 | 518 of non-zero elements rather than the product of the dimensions. Therefore |
| 6569 | 519 |
| 520 @example | |
| 6577 | 521 @group |
| 522 SparseMatrix sm; | |
| 523 @dots{} | |
| 524 int nel = sm.nelem (); | |
| 525 @end group | |
| 6569 | 526 @end example |
| 527 | |
| 6571 | 528 returns the number of non-zero elements. If the user really requires the |
| 6569 | 529 number of elements in the matrix, including the non-zero elements, they |
| 6571 | 530 should use @code{numel} rather than @code{nelem}. Note that for very |
| 7001 | 531 large matrices, where the product of the two dimensions is larger than |
| 532 the representation of an unsigned int, then @code{numel} can overflow. | |
| 6569 | 533 An example is @code{speye(1e6)} which will create a matrix with a million |
| 6571 | 534 rows and columns, but only a million non-zero elements. Therefore the |
| 6569 | 535 number of rows by the number of columns in this case is more than two |
| 536 hundred times the maximum value that can be represented by an unsigned int. | |
| 537 The use of @code{numel} should therefore be avoided useless it is known | |
| 538 it won't overflow. | |
| 539 | |
| 540 Extreme care must be take with the elem method and the "()" operator, | |
| 6571 | 541 which perform basically the same function. The reason is that if a |
| 6569 | 542 sparse object is non-const, then Octave will assume that a |
| 6571 | 543 request for a zero element in a sparse matrix is in fact a request |
| 544 to create this element so it can be filled. Therefore a piece of | |
| 6569 | 545 code like |
| 546 | |
| 547 @example | |
| 6577 | 548 @group |
| 549 SparseMatrix sm; | |
| 550 @dots{} | |
| 551 for (int j = 0; j < nc; j++) | |
| 552 for (int i = 0; i < nr; i++) | |
| 553 std::cerr << " (" << i << "," << j << "): " << sm(i,j) | |
| 554 << std::endl; | |
| 555 @end group | |
| 6569 | 556 @end example |
| 557 | |
| 558 is a great way of turning the sparse matrix into a dense one, and a | |
| 559 very slow way at that since it reallocates the sparse object at each | |
| 560 zero element in the matrix. | |
| 561 | |
| 562 An easy way of preventing the above from happening is to create a temporary | |
| 6571 | 563 constant version of the sparse matrix. Note that only the container for |
| 6569 | 564 the sparse matrix will be copied, while the actual representation of the |
| 6571 | 565 data will be shared between the two versions of the sparse matrix. So this |
| 566 is not a costly operation. For example, the above would become | |
| 6569 | 567 |
| 568 @example | |
| 6577 | 569 @group |
| 570 SparseMatrix sm; | |
| 571 @dots{} | |
| 572 const SparseMatrix tmp (sm); | |
| 573 for (int j = 0; j < nc; j++) | |
| 574 for (int i = 0; i < nr; i++) | |
| 575 std::cerr << " (" << i << "," << j << "): " << tmp(i,j) | |
| 576 << std::endl; | |
| 577 @end group | |
| 6569 | 578 @end example |
| 579 | |
| 580 Finally, as the sparse types aren't just represented as a contiguous | |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
581 block of memory, the @nospell{@code{fortran_vec}} method of the @code{Array<T>} |
| 6571 | 582 is not available. It is however replaced by three separate methods |
| 6569 | 583 @code{ridx}, @code{cidx} and @code{data}, that access the raw compressed |
| 584 column format that the Octave sparse matrices are stored in. | |
| 585 Additionally, these methods can be used in a manner similar to @code{elem}, | |
| 6571 | 586 to allow the matrix to be accessed or filled. However, in that case it is |
| 6569 | 587 up to the user to respect the sparse matrix compressed column format |
| 588 discussed previous. | |
| 589 | |
| 6572 | 590 @node Creating Sparse Matrices in Oct-Files |
| 591 @subsubsection Creating Sparse Matrices in Oct-Files | |
| 6569 | 592 |
| 6572 | 593 You have several alternatives for creating a sparse matrix. |
| 594 You can first create the data as three vectors representing the | |
| 6569 | 595 row and column indexes and the data, and from those create the matrix. |
| 6572 | 596 Or alternatively, you can create a sparse matrix with the appropriate |
| 6571 | 597 amount of space and then fill in the values. Both techniques have their |
| 6569 | 598 advantages and disadvantages. |
| 599 | |
| 6572 | 600 Here is an example of how to create a small sparse matrix with the first |
| 601 technique | |
| 6569 | 602 |
| 603 @example | |
| 6577 | 604 @group |
| 605 int nz = 4, nr = 3, nc = 4; | |
| 606 | |
| 607 ColumnVector ridx (nz); | |
| 608 ColumnVector cidx (nz); | |
| 609 ColumnVector data (nz); | |
| 6569 | 610 |
| 6577 | 611 ridx(0) = 0; ridx(1) = 0; ridx(2) = 1; ridx(3) = 2; |
| 612 cidx(0) = 0; cidx(1) = 1; cidx(2) = 3; cidx(3) = 3; | |
| 613 data(0) = 1; data(1) = 2; data(2) = 3; data(3) = 4; | |
| 6569 | 614 |
| 6577 | 615 SparseMatrix sm (data, ridx, cidx, nr, nc); |
| 616 @end group | |
| 6569 | 617 @end example |
| 618 | |
| 6572 | 619 @noindent |
|
8817
03b7f618ab3d
include docstrings for new functions in the manual
John W. Eaton <jwe@octave.org>
parents:
8486
diff
changeset
|
620 which creates the matrix given in section |
|
03b7f618ab3d
include docstrings for new functions in the manual
John W. Eaton <jwe@octave.org>
parents:
8486
diff
changeset
|
621 @ref{Storage of Sparse Matrices}. Note that the compressed matrix |
|
03b7f618ab3d
include docstrings for new functions in the manual
John W. Eaton <jwe@octave.org>
parents:
8486
diff
changeset
|
622 format is not used at the time of the creation of the matrix itself, |
|
03b7f618ab3d
include docstrings for new functions in the manual
John W. Eaton <jwe@octave.org>
parents:
8486
diff
changeset
|
623 however it is used internally. |
| 6569 | 624 |
| 625 As previously mentioned, the values of the sparse matrix are stored | |
| 6571 | 626 in increasing column-major ordering. Although the data passed by the |
| 6569 | 627 user does not need to respect this requirement, the pre-sorting the |
| 628 data significantly speeds up the creation of the sparse matrix. | |
| 629 | |
| 630 The disadvantage of this technique of creating a sparse matrix is | |
| 6571 | 631 that there is a brief time where two copies of the data exists. Therefore |
| 6569 | 632 for extremely memory constrained problems this might not be the right |
| 633 technique to create the sparse matrix. | |
| 634 | |
| 635 The alternative is to first create the sparse matrix with the desired | |
| 6571 | 636 number of non-zero elements and then later fill those elements in. The |
| 637 easiest way to do this is | |
| 6569 | 638 |
| 6571 | 639 @example |
| 6577 | 640 @group |
| 641 int nz = 4, nr = 3, nc = 4; | |
| 642 SparseMatrix sm (nr, nc, nz); | |
| 643 sm(0,0) = 1; sm(0,1) = 2; sm(1,3) = 3; sm(2,3) = 4; | |
| 644 @end group | |
| 6569 | 645 @end example |
| 646 | |
| 6571 | 647 That creates the same matrix as previously. Again, although it is not |
| 6569 | 648 strictly necessary, it is significantly faster if the sparse matrix is |
| 649 created in this manner that the elements are added in column-major | |
| 6571 | 650 ordering. The reason for this is that if the elements are inserted |
| 6569 | 651 at the end of the current list of known elements then no element |
| 652 in the matrix needs to be moved to allow the new element to be | |
| 6571 | 653 inserted. Only the column indexes need to be updated. |
| 6569 | 654 |
| 655 There are a few further points to note about this technique of creating | |
| 6572 | 656 a sparse matrix. Firstly, it is possible to create a sparse matrix |
| 6571 | 657 with fewer elements than are actually inserted in the matrix. Therefore |
| 6569 | 658 |
| 6571 | 659 @example |
| 6577 | 660 @group |
| 661 int nz = 4, nr = 3, nc = 4; | |
| 662 SparseMatrix sm (nr, nc, 0); | |
| 663 sm(0,0) = 1; sm(0,1) = 2; sm(1,3) = 3; sm(2,3) = 4; | |
| 664 @end group | |
| 6569 | 665 @end example |
| 666 | |
| 6572 | 667 @noindent |
| 668 is perfectly valid. However it is a very bad idea. The reason is that | |
| 6569 | 669 as each new element is added to the sparse matrix the space allocated |
| 6571 | 670 to it is increased by reallocating the memory. This is an expensive |
| 6569 | 671 operation, that will significantly slow this means of creating a sparse |
| 6572 | 672 matrix. Furthermore, it is possible to create a sparse matrix with |
| 673 too much storage, so having @var{nz} above equaling 6 is also valid. | |
| 6569 | 674 The disadvantage is that the matrix occupies more memory than strictly |
| 675 needed. | |
| 676 | |
| 6572 | 677 It is not always easy to know the number of non-zero elements prior |
| 6571 | 678 to filling a matrix. For this reason the additional storage for the |
| 6569 | 679 sparse matrix can be removed after its creation with the |
| 6571 | 680 @dfn{maybe_compress} function. Furthermore, the maybe_compress can |
| 6569 | 681 deallocate the unused storage, but it can equally remove zero elements |
| 682 from the matrix. The removal of zero elements from the matrix is | |
| 683 controlled by setting the argument of the @dfn{maybe_compress} function | |
| 6572 | 684 to be @samp{true}. However, the cost of removing the zeros is high because it |
| 6571 | 685 implies resorting the elements. Therefore, if possible it is better |
| 686 is the user doesn't add the zeros in the first place. An example of | |
| 6569 | 687 the use of @dfn{maybe_compress} is |
| 688 | |
| 689 @example | |
| 6577 | 690 @group |
| 6569 | 691 int nz = 6, nr = 3, nc = 4; |
| 6577 | 692 |
| 6569 | 693 SparseMatrix sm1 (nr, nc, nz); |
| 694 sm1(0,0) = 1; sm1(0,1) = 2; sm1(1,3) = 3; sm1(2,3) = 4; | |
| 695 sm1.maybe_compress (); // No zero elements were added | |
| 696 | |
| 697 SparseMatrix sm2 (nr, nc, nz); | |
| 6571 | 698 sm2(0,0) = 1; sm2(0,1) = 2; sm(0,2) = 0; sm(1,2) = 0; |
| 6569 | 699 sm1(1,3) = 3; sm1(2,3) = 4; |
| 700 sm2.maybe_compress (true); // Zero elements were added | |
| 6577 | 701 @end group |
| 6569 | 702 @end example |
| 703 | |
| 704 The use of the @dfn{maybe_compress} function should be avoided if | |
| 705 possible, as it will slow the creation of the matrices. | |
| 706 | |
| 707 A third means of creating a sparse matrix is to work directly with | |
| 6571 | 708 the data in compressed row format. An example of this technique might |
| 6569 | 709 be |
| 710 | |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
711 @c Note the @verbatim environment is a relatively new addition to Texinfo. |
| 6571 | 712 @c Therefore use the @example environment and replace @, with @@, |
| 6569 | 713 @c { with @{, etc |
| 714 | |
| 715 @example | |
| 6577 | 716 octave_value arg; |
| 717 @dots{} | |
| 718 int nz = 6, nr = 3, nc = 4; // Assume we know the max no nz | |
| 719 SparseMatrix sm (nr, nc, nz); | |
| 720 Matrix m = arg.matrix_value (); | |
| 6569 | 721 |
| 6577 | 722 int ii = 0; |
| 723 sm.cidx (0) = 0; | |
| 724 for (int j = 1; j < nc; j++) | |
| 725 @{ | |
| 726 for (int i = 0; i < nr; i++) | |
| 727 @{ | |
| 728 double tmp = foo (m(i,j)); | |
| 729 if (tmp != 0.) | |
| 730 @{ | |
| 731 sm.data(ii) = tmp; | |
| 732 sm.ridx(ii) = i; | |
| 733 ii++; | |
| 734 @} | |
| 735 @} | |
| 736 sm.cidx(j+1) = ii; | |
| 737 @} | |
| 7081 | 738 sm.maybe_compress (); // If don't know a-priori |
| 739 // the final no of nz. | |
| 6569 | 740 @end example |
| 741 | |
| 6572 | 742 @noindent |
| 6569 | 743 which is probably the most efficient means of creating the sparse matrix. |
| 744 | |
| 745 Finally, it might sometimes arise that the amount of storage initially | |
| 6571 | 746 created is insufficient to completely store the sparse matrix. Therefore, |
| 6569 | 747 the method @code{change_capacity} exists to reallocate the sparse memory. |
| 6571 | 748 The above example would then be modified as |
| 6569 | 749 |
| 750 @example | |
| 6577 | 751 octave_value arg; |
| 752 @dots{} | |
| 753 int nz = 6, nr = 3, nc = 4; // Assume we know the max no nz | |
| 754 SparseMatrix sm (nr, nc, nz); | |
| 755 Matrix m = arg.matrix_value (); | |
| 6569 | 756 |
| 6577 | 757 int ii = 0; |
| 758 sm.cidx (0) = 0; | |
| 759 for (int j = 1; j < nc; j++) | |
| 760 @{ | |
| 761 for (int i = 0; i < nr; i++) | |
| 762 @{ | |
| 763 double tmp = foo (m(i,j)); | |
| 764 if (tmp != 0.) | |
| 765 @{ | |
| 766 if (ii == nz) | |
| 767 @{ | |
| 768 nz += 2; // Add 2 more elements | |
| 769 sm.change_capacity (nz); | |
| 770 @} | |
| 771 sm.data(ii) = tmp; | |
| 772 sm.ridx(ii) = i; | |
| 773 ii++; | |
| 774 @} | |
| 775 @} | |
| 776 sm.cidx(j+1) = ii; | |
| 777 @} | |
| 7081 | 778 sm.maybe_mutate (); // If don't know a-priori |
| 779 // the final no of nz. | |
| 6569 | 780 @end example |
| 781 | |
| 782 Note that both increasing and decreasing the number of non-zero elements in | |
| 6571 | 783 a sparse matrix is expensive, as it involves memory reallocation. Also as |
| 6569 | 784 parts of the matrix, though not its entirety, exist as the old and new copy |
| 6571 | 785 at the same time, additional memory is needed. Therefore if possible this |
| 6569 | 786 should be avoided. |
| 787 | |
| 6572 | 788 @node Using Sparse Matrices in Oct-Files |
| 6569 | 789 @subsubsection Using Sparse Matrices in Oct-Files |
| 790 | |
| 791 Most of the same operators and functions on sparse matrices that are | |
| 792 available from the Octave are equally available with oct-files. | |
| 793 The basic means of extracting a sparse matrix from an @code{octave_value} | |
| 794 and returning them as an @code{octave_value}, can be seen in the | |
| 795 following example | |
| 796 | |
| 797 @example | |
| 6577 | 798 @group |
| 799 octave_value_list retval; | |
| 6569 | 800 |
| 6577 | 801 SparseMatrix sm = args(0).sparse_matrix_value (); |
| 7081 | 802 SparseComplexMatrix scm = |
| 803 args(1).sparse_complex_matrix_value (); | |
| 6577 | 804 SparseBoolMatrix sbm = args(2).sparse_bool_matrix_value (); |
| 805 @dots{} | |
| 806 retval(2) = sbm; | |
| 807 retval(1) = scm; | |
| 808 retval(0) = sm; | |
| 809 @end group | |
| 6569 | 810 @end example |
| 811 | |
| 812 The conversion to an octave-value is handled by the sparse | |
| 813 @code{octave_value} constructors, and so no special care is needed. | |
| 814 | |
| 815 @node Accessing Global Variables in Oct-Files | |
| 816 @subsection Accessing Global Variables in Oct-Files | |
| 817 | |
| 818 Global variables allow variables in the global scope to be | |
| 6571 | 819 accessed. Global variables can easily be accessed with oct-files using |
| 6569 | 820 the support functions @code{get_global_value} and |
| 6571 | 821 @code{set_global_value}. @code{get_global_value} takes two arguments, |
| 822 the first is a string representing the variable name to obtain. The | |
| 6569 | 823 second argument is a boolean argument specifying what to do in the case |
| 6571 | 824 that no global variable of the desired name is found. An example of the |
| 6569 | 825 use of these two functions is |
| 826 | |
| 9906 | 827 @example |
| 828 @EXAMPLEFILE(globaldemo.cc) | |
| 829 @end example | |
| 6569 | 830 |
| 831 An example of its use is | |
| 832 | |
| 833 @example | |
| 834 @group | |
| 835 global a b | |
| 836 b = 10; | |
| 837 globaldemo ("b") | |
| 838 @result{} 10 | |
| 839 globaldemo ("c") | |
| 840 @result{} "Global variable not found" | |
| 841 num2str (a) | |
| 842 @result{} 42 | |
| 843 @end group | |
| 844 @end example | |
| 845 | |
| 846 @node Calling Octave Functions from Oct-Files | |
| 847 @subsection Calling Octave Functions from Oct-Files | |
| 848 | |
| 849 There is often a need to be able to call another octave function from | |
| 850 within an oct-file, and there are many examples of such within octave | |
| 6571 | 851 itself. For example the @code{quad} function is an oct-file that |
| 6569 | 852 calculates the definite integral by quadrature over a user supplied |
| 853 function. | |
| 854 | |
| 6571 | 855 There are also many ways in which a function might be passed. It might |
| 856 be passed as one of | |
| 6569 | 857 |
| 858 @enumerate 1 | |
| 859 @item Function Handle | |
| 860 @item Anonymous Function Handle | |
| 861 @item Inline Function | |
| 862 @item String | |
| 863 @end enumerate | |
| 864 | |
| 865 The example below demonstrates an example that accepts all four means of | |
| 6571 | 866 passing a function to an oct-file. |
| 6569 | 867 |
| 9906 | 868 @example |
| 869 @EXAMPLEFILE(funcdemo.cc) | |
| 870 @end example | |
| 6569 | 871 |
| 872 The first argument to this demonstration is the user supplied function | |
| 873 and the following arguments are all passed to the user function. | |
| 874 | |
| 875 @example | |
| 876 @group | |
| 6572 | 877 funcdemo (@@sin,1) |
| 6569 | 878 @result{} 0.84147 |
| 6572 | 879 funcdemo (@@(x) sin(x), 1) |
| 6569 | 880 @result{} 0.84147 |
| 6572 | 881 funcdemo (inline ("sin(x)"), 1) |
| 6569 | 882 @result{} 0.84147 |
| 6572 | 883 funcdemo ("sin",1) |
| 6569 | 884 @result{} 0.84147 |
| 885 funcdemo (@@atan2, 1, 1) | |
| 886 @result{} 0.78540 | |
| 887 @end group | |
| 888 @end example | |
| 889 | |
| 890 When the user function is passed as a string, the treatment of the | |
| 6571 | 891 function is different. In some cases it is necessary to always have the |
| 6572 | 892 user supplied function as an @code{octave_function} object. In that |
| 893 case the string argument can be used to create a temporary function like | |
| 6569 | 894 |
| 895 @example | |
| 896 @group | |
| 6577 | 897 std::octave fcn_name = unique_symbol_name ("__fcn__"); |
| 898 std::string fname = "function y = "; | |
| 899 fname.append (fcn_name); | |
| 900 fname.append ("(x) y = "); | |
| 901 fcn = extract_function (args(0), "funcdemo", fcn_name, | |
| 902 fname, "; endfunction"); | |
| 903 @dots{} | |
| 904 if (fcn_name.length ()) | |
| 905 clear_function (fcn_name); | |
| 6569 | 906 @end group |
| 907 @end example | |
| 908 | |
| 6571 | 909 There are two important things to know in this case. The number of input |
| 6569 | 910 arguments to the user function is fixed, and in the above is a single |
| 911 argument, and secondly to avoid leaving the temporary function in the | |
| 912 Octave symbol table it should be cleared after use. | |
| 913 | |
| 914 @node Calling External Code from Oct-Files | |
| 915 @subsection Calling External Code from Oct-Files | |
| 916 | |
| 917 Linking external C code to Octave is relatively simple, as the C | |
| 6571 | 918 functions can easily be called directly from C++. One possible issue is |
| 6569 | 919 the declarations of the external C functions might need to be explicitly |
| 6571 | 920 defined as C functions to the compiler. If the declarations of the |
| 6569 | 921 external C functions are in the header @code{foo.h}, then the manner in |
| 922 which to ensure that the C++ compiler treats these declarations as C | |
| 923 code is | |
| 924 | |
| 925 @example | |
| 926 @group | |
| 927 #ifdef __cplusplus | |
| 6571 | 928 extern "C" |
| 6569 | 929 @{ |
| 930 #endif | |
| 931 #include "foo.h" | |
| 932 #ifdef __cplusplus | |
| 933 @} /* end extern "C" */ | |
| 934 #endif | |
| 935 @end group | |
| 936 @end example | |
| 937 | |
| 6571 | 938 Calling Fortran code however can pose some difficulties. This is due to |
| 6569 | 939 differences in the manner in compilers treat the linking of Fortran code |
| 6571 | 940 with C or C++ code. Octave supplies a number of macros that allow |
| 6569 | 941 consistent behavior across a number of compilers. |
| 942 | |
| 943 The underlying Fortran code should use the @code{XSTOPX} function to | |
| 6571 | 944 replace the Fortran @code{STOP} function. @code{XSTOPX} uses the Octave |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
945 exception handler to treat failing cases in the Fortran code |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
946 explicitly. Note that Octave supplies its own replacement @sc{blas} |
| 6569 | 947 @code{XERBLA} function, which uses @code{XSTOPX}. |
| 948 | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
949 If the underlying code calls @code{XSTOPX}, then the @w{@code{F77_XFCN}} |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
950 macro should be used to call the underlying Fortran function. The Fortran |
| 6569 | 951 exception state can then be checked with the global variable |
| 6572 | 952 @code{f77_exception_encountered}. If @code{XSTOPX} will not be called, |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
953 then the @w{@code{F77_FCN}} macro should be used instead to call the Fortran |
| 6569 | 954 code. |
| 955 | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
956 There is no harm in using @w{@code{F77_XFCN}} in all cases, except that for |
| 6569 | 957 Fortran code that is short running and executes a large number of times, |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
958 there is potentially an overhead in doing so. However, if @w{@code{F77_FCN}} |
| 6569 | 959 is used with code that calls @code{XSTOP}, Octave can generate a |
| 960 segmentation fault. | |
| 961 | |
| 962 An example of the inclusion of a Fortran function in an oct-file is | |
| 963 given in the following example, where the C++ wrapper is | |
| 964 | |
| 9906 | 965 @example |
| 966 @EXAMPLEFILE(fortdemo.cc) | |
| 967 @end example | |
| 6569 | 968 |
| 6572 | 969 @noindent |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
970 and the Fortran function is |
| 6569 | 971 |
| 9906 | 972 @example |
| 973 @EXAMPLEFILE(fortsub.f) | |
| 974 @end example | |
| 6569 | 975 |
| 976 This example demonstrates most of the features needed to link to an | |
| 977 external Fortran function, including passing arrays and strings, as well | |
| 6571 | 978 as exception handling. An example of the behavior of this function is |
| 6569 | 979 |
| 980 @example | |
| 981 @group | |
| 6572 | 982 [b, s] = fortdemo (1:3) |
| 6569 | 983 @result{} |
| 984 b = 1.00000 0.50000 0.33333 | |
| 985 s = There are 3 values in the input vector | |
| 986 [b, s] = fortdemo(0:3) | |
| 987 error: fortsub:divide by zero | |
| 988 error: exception encountered in Fortran subroutine fortsub_ | |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
989 error: fortdemo: error in Fortran |
| 6569 | 990 @end group |
| 991 @end example | |
| 992 | |
| 993 @node Allocating Local Memory in Oct-Files | |
| 994 @subsection Allocating Local Memory in Oct-Files | |
| 995 | |
| 996 Allocating memory within an oct-file might seem easy as the C++ | |
| 6571 | 997 new/delete operators can be used. However, in that case care must be |
| 998 taken to avoid memory leaks. The preferred manner in which to allocate | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
999 memory for use locally is to use the @w{@code{OCTAVE_LOCAL_BUFFER}} macro. |
| 6572 | 1000 An example of its use is |
| 6569 | 1001 |
| 1002 @example | |
| 1003 OCTAVE_LOCAL_BUFFER (double, tmp, len) | |
| 1004 @end example | |
| 1005 | |
| 1006 that returns a pointer @code{tmp} of type @code{double *} of length | |
| 1007 @code{len}. | |
| 1008 | |
| 1009 @node Input Parameter Checking in Oct-Files | |
| 1010 @subsection Input Parameter Checking in Oct-Files | |
| 1011 | |
| 6580 | 1012 As oct-files are compiled functions they have the possibility of causing |
| 7001 | 1013 Octave to abort abnormally. It is therefore important that |
| 1014 each and every function has the minimum of parameter | |
| 6580 | 1015 checking needed to ensure that Octave behaves well. |
| 1016 | |
| 1017 The minimum requirement, as previously discussed, is to check the number | |
| 1018 of input arguments before using them to avoid referencing a non existent | |
| 1019 argument. However, it some case this might not be sufficient as the | |
| 6593 | 1020 underlying code imposes further constraints. For example an external |
| 6580 | 1021 function call might be undefined if the input arguments are not |
| 6593 | 1022 integers, or if one of the arguments is zero. Therefore, oct-files often |
| 6580 | 1023 need additional input parameter checking. |
| 1024 | |
| 1025 There are several functions within Octave that might be useful for the | |
| 6593 | 1026 purposes of parameter checking. These include the methods of the |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1027 octave_value class like @code{is_real_matrix}, etc., but equally include |
| 6593 | 1028 more specialized functions. Some of the more common ones are |
| 6580 | 1029 demonstrated in the following example |
| 1030 | |
| 9906 | 1031 @example |
| 1032 @EXAMPLEFILE(paramdemo.cc) | |
| 1033 @end example | |
| 6580 | 1034 |
| 1035 @noindent | |
| 1036 and an example of its use is | |
| 1037 | |
| 1038 @example | |
| 1039 @group | |
| 1040 paramdemo ([1, 2, NaN, Inf]) | |
| 1041 @result{} Properties of input array: | |
| 1042 includes Inf or NaN values | |
| 1043 includes other values than 1 and 0 | |
| 1044 includes only int, Inf or NaN values | |
| 1045 @end group | |
| 1046 @end example | |
| 6569 | 1047 |
| 1048 @node Exception and Error Handling in Oct-Files | |
| 1049 @subsection Exception and Error Handling in Oct-Files | |
| 1050 | |
| 1051 Another important feature of Octave is its ability to react to the user | |
| 6571 | 1052 typing @kbd{Control-C} even during calculations. This ability is based on the |
| 6569 | 1053 C++ exception handler, where memory allocated by the C++ new/delete |
| 6571 | 1054 methods are automatically released when the exception is treated. When |
| 6569 | 1055 writing an oct-file, to allow Octave to treat the user typing @kbd{Control-C}, |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
1056 the @w{@code{OCTAVE_QUIT}} macro is supplied. For example |
| 6569 | 1057 |
| 1058 @example | |
| 1059 @group | |
| 6577 | 1060 for (octave_idx_type i = 0; i < a.nelem (); i++) |
| 1061 @{ | |
| 1062 OCTAVE_QUIT; | |
| 1063 b.elem(i) = 2. * a.elem(i); | |
| 1064 @} | |
| 6569 | 1065 @end group |
| 1066 @end example | |
| 1067 | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
1068 The presence of the @w{@code{OCTAVE_QUIT}} macro in the inner loop allows Octave to |
| 6571 | 1069 treat the user request with the @kbd{Control-C}. Without this macro, the user |
| 6569 | 1070 must either wait for the function to return before the interrupt is |
| 1071 processed, or press @kbd{Control-C} three times to force Octave to exit. | |
| 1072 | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
1073 The @w{@code{OCTAVE_QUIT}} macro does impose a very small speed penalty, and so for |
| 6569 | 1074 loops that are known to be small it might not make sense to include |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
1075 @w{@code{OCTAVE_QUIT}}. |
| 6569 | 1076 |
| 1077 When creating an oct-file that uses an external libraries, the function | |
| 1078 might spend a significant portion of its time in the external | |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
1079 library. It is not generally possible to use the @w{@code{OCTAVE_QUIT}} macro in |
| 6571 | 1080 this case. The alternative in this case is |
| 6569 | 1081 |
| 1082 @example | |
| 1083 @group | |
| 6577 | 1084 BEGIN_INTERRUPT_IMMEDIATELY_IN_FOREIGN_CODE; |
| 1085 @dots{} some code that calls a "foreign" function @dots{} | |
| 1086 END_INTERRUPT_IMMEDIATELY_IN_FOREIGN_CODE; | |
| 6569 | 1087 @end group |
| 1088 @end example | |
| 1089 | |
| 1090 The disadvantage of this is that if the foreign code allocates any | |
| 1091 memory internally, then this memory might be lost during an interrupt, | |
| 6571 | 1092 without being deallocated. Therefore, ideally Octave itself should |
| 6569 | 1093 allocate any memory that is needed by the foreign code, with either the |
|
10791
3140cb7a05a1
Add spellchecker scripts for Octave and run spellcheck of documentation
Rik <octave@nomad.inbox5.com>
parents:
9906
diff
changeset
|
1094 @nospell{fortran_vec} method or the @w{@code{OCTAVE_LOCAL_BUFFER}} macro. |
| 6569 | 1095 |
|
9038
fca0dc2fb042
Cleanup documentation files stmt.texi and func.texi
Rik <rdrider0-list@yahoo.com>
parents:
8920
diff
changeset
|
1096 The Octave unwind_protect mechanism (@ref{The @code{unwind_protect} Statement}) |
| 6571 | 1097 can also be used in oct-files. In conjunction with the exception |
| 6569 | 1098 handling of Octave, it is important to enforce that certain code is run |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1099 to allow variables, etc. to be restored even if an exception occurs. An |
| 6569 | 1100 example of the use of this mechanism is |
| 1101 | |
| 9906 | 1102 @example |
| 1103 @EXAMPLEFILE(unwinddemo.cc) | |
| 1104 @end example | |
| 6569 | 1105 |
| 1106 As can be seen in the example | |
| 1107 | |
| 1108 @example | |
| 1109 @group | |
| 6572 | 1110 unwinddemo (1, 0) |
| 6569 | 1111 @result{} Inf |
| 1112 1 / 0 | |
| 1113 @result{} warning: division by zero | |
| 6593 | 1114 Inf |
| 6569 | 1115 @end group |
| 1116 @end example | |
| 1117 | |
| 1118 The division by zero (and in fact all warnings) is disabled in the | |
| 1119 @code{unwinddemo} function. | |
| 1120 | |
| 1121 @node Documentation and Test of Oct-Files | |
| 1122 @subsection Documentation and Test of Oct-Files | |
| 1123 | |
| 6580 | 1124 The documentation of an oct-file is the fourth string parameter of the |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
1125 @w{@code{DEFUN_DLD}} macro. This string can be formatted in the same manner |
| 6580 | 1126 as the help strings for user functions (@ref{Documentation Tips}), |
| 1127 however there are some issue that are particular to the formatting of | |
| 1128 help strings within oct-files. | |
| 1129 | |
| 1130 The major issue is that the help string will typically be longer than a | |
| 1131 single line of text, and so the formatting of long help strings need to | |
| 7001 | 1132 be taken into account. There are several manners in which to treat this |
| 6580 | 1133 issue, but the most common is illustrated in the following example |
| 1134 | |
| 1135 @example | |
| 1136 @group | |
| 1137 DEFUN_DLD (do_what_i_want, args, nargout, | |
| 1138 "-*- texinfo -*-\n\ | |
| 1139 @@deftypefn @{Function File@} @{@} do_what_i_say (@@var@{n@})\n\ | |
| 7081 | 1140 A function that does what the user actually wants rather\n\ |
| 1141 than what they requested.\n\ | |
| 6580 | 1142 @@end deftypefn") |
| 1143 @{ | |
| 1144 @dots{} | |
| 1145 @} | |
| 1146 @end group | |
| 1147 @end example | |
| 1148 | |
| 1149 @noindent | |
| 1150 where, as can be seen, end line of text within the help string is | |
|
8347
fa78cb8d8a5c
corrections for typos
Brian Gough<bjg@network-theory.co.uk>
parents:
8097
diff
changeset
|
1151 terminated by @code{\n\} which is an embedded new-line in the string |
| 6580 | 1152 together with a C++ string continuation character. Note that the final |
| 1153 @code{\} must be the last character on the line. | |
| 1154 | |
| 1155 Octave also includes the ability to embed the test and demonstration | |
| 1156 code for a function within the code itself (@ref{Test and Demo Functions}). | |
| 1157 This can be used from within oct-files (or in fact any file) with | |
| 1158 certain provisos. Firstly, the test and demo functions of Octave look | |
| 1159 for a @code{%!} as the first characters on a new-line to identify test | |
| 1160 and demonstration code. This is equally a requirement for | |
| 1161 oct-files. Furthermore the test and demonstration code must be included | |
| 1162 in a comment block of the compiled code to avoid it being interpreted by | |
| 6606 | 1163 the compiler. Finally, the Octave test and demonstration code must have |
| 6580 | 1164 access to the source code of the oct-file and not just the compiled code |
| 6606 | 1165 as the tests are stripped from the compiled code. An example in an |
| 6580 | 1166 oct-file might be |
| 1167 | |
| 1168 @example | |
| 1169 @group | |
| 1170 /* | |
| 1171 | |
| 1172 %!error (sin()) | |
| 1173 %!error (sin(1,1)) | |
| 1174 %!assert (sin([1,2]),[sin(1),sin(2)]) | |
| 1175 | |
| 1176 */ | |
| 1177 @end group | |
| 1178 @end example | |
| 6569 | 1179 |
| 6593 | 1180 @c @node Application Programming Interface for Oct-Files |
| 1181 @c @subsection Application Programming Interface for Oct-Files | |
| 1182 @c | |
| 1183 @c WRITE ME, using Coda section 1.3 as a starting point. | |
| 6569 | 1184 |
| 1185 @node Mex-Files | |
| 1186 @section Mex-Files | |
| 1187 @cindex mex-files | |
| 1188 @cindex mex | |
| 1189 | |
| 1190 Octave includes an interface to allow legacy mex-files to be compiled | |
| 6571 | 1191 and used with Octave. This interface can also be used to share code |
| 1192 between Octave and non Octave users. However, as mex-files expose the | |
| 6593 | 1193 internal API of an alternative product to Octave, and the internal |
| 6569 | 1194 structure of Octave is different to this product, a mex-file can never |
| 6571 | 1195 have the same performance in Octave as the equivalent oct-file. In |
| 6569 | 1196 particular to support the manner in which mex-files access the variables |
| 1197 passed to mex functions, there are a significant number of additional | |
| 6593 | 1198 copies of memory when calling or returning from a mex function. For |
| 1199 this reason, new code should be written using the oct-file interface | |
| 6569 | 1200 discussed above if possible. |
| 1201 | |
| 1202 @menu | |
| 6572 | 1203 * Getting Started with Mex-Files:: |
| 6580 | 1204 * Working with Matrices and Arrays in Mex-Files:: |
| 1205 * Character Strings in Mex-Files:: | |
| 1206 * Cell Arrays with Mex-Files:: | |
| 6572 | 1207 * Structures with Mex-Files:: |
| 1208 * Sparse Matrices with Mex-Files:: | |
| 6580 | 1209 * Calling Other Functions in Mex-Files:: |
| 6593 | 1210 @c * Application Programming Interface for Mex-Files:: |
| 6569 | 1211 @end menu |
| 1212 | |
| 1213 @node Getting Started with Mex-Files | |
| 1214 @subsection Getting Started with Mex-Files | |
| 1215 | |
| 1216 The basic command to build a mex-file is either @code{mkoctfile --mex} or | |
| 6571 | 1217 @code{mex}. The first can either be used from within Octave or from the |
| 8486 | 1218 command line. However, to avoid issues with the installation of other |
| 6569 | 1219 products, the use of the command @code{mex} is limited to within Octave. |
| 1220 | |
| 1221 @DOCSTRING(mex) | |
| 1222 | |
| 1223 @DOCSTRING(mexext) | |
| 1224 | |
| 1225 One important difference between the use of mex with other products and | |
| 1226 with Octave is that the header file "matrix.h" is implicitly included | |
| 6571 | 1227 through the inclusion of "mex.h". This is to avoid a conflict with the |
| 6569 | 1228 Octave file "Matrix.h" with operating systems and compilers that don't |
| 1229 distinguish between filenames in upper and lower case | |
| 1230 | |
| 1231 Consider the short example | |
| 1232 | |
| 9906 | 1233 @example |
| 1234 @group | |
| 1235 @EXAMPLEFILE(firstmexdemo.c) | |
| 1236 @end group | |
| 1237 @end example | |
| 6569 | 1238 |
| 6593 | 1239 This simple example demonstrates the basics of writing a mex-file. The |
| 1240 entry point into the mex-file is defined by @code{mexFunction}. Note | |
| 6580 | 1241 that the function name is not explicitly included in the |
| 1242 @code{mexFunction} and so there can only be a single @code{mexFunction} | |
| 1243 entry point per-file. Also the name of the function is determined by the | |
| 1244 name of the mex-file itself. Therefore if the above function is in the | |
| 1245 file @file{firstmexdemo.c}, it can be compiled with | |
| 1246 | |
| 1247 @example | |
| 1248 mkoctfile --mex firstmexdemo.c | |
| 1249 @end example | |
| 1250 | |
| 1251 @noindent | |
| 1252 which creates a file @file{firstmexdemo.mex}. The function can then be run | |
| 1253 from Octave as | |
| 1254 | |
| 1255 @example | |
| 1256 @group | |
| 1257 firstmexdemo() | |
| 1258 @result{} 1.2346 | |
| 1259 @end group | |
| 1260 @end example | |
| 1261 | |
| 1262 It should be noted that the mex-file contains no help string for the | |
| 6593 | 1263 functions it contains. To document mex-files, there should exist an |
| 1264 m-file in the same directory as the mex-file itself. Taking the above as | |
| 6580 | 1265 an example, we would therefore have a file @file{firstmexdemo.m} that might |
| 1266 contain the text | |
| 1267 | |
| 1268 @example | |
| 1269 %FIRSTMEXDEMO Simple test of the functionality of a mex-file. | |
| 1270 @end example | |
| 1271 | |
| 1272 In this case, the function that will be executed within Octave will be | |
| 1273 given by the mex-file, while the help string will come from the | |
| 6593 | 1274 m-file. This can also be useful to allow a sample implementation of the |
| 6580 | 1275 mex-file within the Octave language itself for testing purposes. |
| 1276 | |
|
8347
fa78cb8d8a5c
corrections for typos
Brian Gough<bjg@network-theory.co.uk>
parents:
8097
diff
changeset
|
1277 Although we cannot have multiple entry points into a single mex-file, |
| 6580 | 1278 we can use the @code{mexFunctionName} function to determine what name |
| 6593 | 1279 the mex-file was called with. This can be used to alter the behavior of |
| 1280 the mex-file based on the function name. For example if | |
| 6580 | 1281 |
| 9906 | 1282 @example |
| 1283 @group | |
| 1284 @EXAMPLEFILE(myfunc.c) | |
| 1285 @end group | |
| 1286 @end example | |
| 6580 | 1287 |
| 1288 @noindent | |
| 1289 is in file @file{myfunc.c}, and it is compiled with | |
| 1290 | |
| 1291 @example | |
| 1292 @group | |
| 1293 mkoctfile --mex myfunc.c | |
| 1294 ln -s myfunc.mex myfunc2.mex | |
| 1295 @end group | |
| 1296 @end example | |
| 1297 | |
| 1298 Then as can be seen by | |
| 1299 | |
| 1300 @example | |
| 1301 @group | |
| 1302 myfunc() | |
| 1303 @result{} You called function: myfunc | |
| 1304 This is the principal function | |
| 1305 myfunc2() | |
| 1306 @result{} You called function: myfunc2 | |
| 1307 @end group | |
| 1308 @end example | |
| 1309 | |
| 1310 @noindent | |
| 1311 the behavior of the mex-file can be altered depending on the functions | |
| 1312 name. | |
| 1313 | |
| 6593 | 1314 Allow the user should only include @code{mex.h} in their code, Octave |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1315 declares additional functions, typedefs, etc., available to the user to |
| 6593 | 1316 write mex-files in the headers @code{mexproto.h} and @code{mxarray.h}. |
| 1317 | |
| 6580 | 1318 @node Working with Matrices and Arrays in Mex-Files |
| 1319 @subsection Working with Matrices and Arrays in Mex-Files | |
| 1320 | |
| 6593 | 1321 The basic mex type of all variables is @code{mxArray}. All variables, |
| 1322 such as matrices, cell arrays or structures are all stored in this basic | |
| 6580 | 1323 type, and this type serves basically the same purpose as the |
| 6593 | 1324 octave_value class in oct-files. That is it acts as a container for the |
| 6580 | 1325 more specialized types. |
| 1326 | |
| 6593 | 1327 The @code{mxArray} structure contains at a minimum, the variable it |
| 1328 represents name, its dimensions, its type and whether the variable is | |
| 1329 real or complex. It can however contain a number of additional fields | |
| 1330 depending on the type of the @code{mxArray}. There are a number of | |
| 1331 functions to create @code{mxArray} structures, including | |
| 1332 @code{mxCreateCellArray}, @code{mxCreateSparse} and the generic | |
| 1333 @code{mxCreateNumericArray}. | |
| 1334 | |
| 1335 The basic functions to access the data contained in an array is | |
| 1336 @code{mxGetPr}. As the mex interface assumes that the real and imaginary | |
| 6939 | 1337 parts of a complex array are stored separately, there is an equivalent |
| 6593 | 1338 function @code{mxGetPi} that get the imaginary part. Both of these |
| 1339 functions are for use only with double precision matrices. There also | |
| 1340 exists the generic function @code{mxGetData} and @code{mxGetImagData} | |
| 1341 that perform the same operation on all matrix types. For example | |
| 1342 | |
| 1343 @example | |
| 1344 @group | |
| 1345 mxArray *m; | |
| 6686 | 1346 mwSize *dims; |
| 6593 | 1347 UINT32_T *pr; |
| 1348 | |
| 6686 | 1349 dims = (mwSize *) mxMalloc (2 * sizeof(mwSize)); |
| 6593 | 1350 dims[0] = 2; |
| 1351 dims[1] = 2; | |
| 1352 m = mxCreateNumericArray (2, dims, mxUINT32_CLASS, mxREAL); | |
| 1353 pr = = (UINT32_T *) mxGetData (m); | |
| 1354 @end group | |
| 1355 @end example | |
| 1356 | |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1357 There are also the functions @code{mxSetPr}, etc., that perform the |
| 6593 | 1358 inverse, and set the data of an Array to use the block of memory pointed |
| 1359 to by the argument of @code{mxSetPr}. | |
| 1360 | |
| 6686 | 1361 Note the type @code{mwSize} used above, and @code{mwIndex} are defined |
| 1362 as the native precision of the indexing in Octave on the platform on | |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1363 which the mex-file is built. This allows both 32- and 64-bit platforms |
|
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1364 to support mex-files. @code{mwSize} is used to define array dimension |
| 6686 | 1365 and maximum number or elements, while @code{mwIndex} is used to define |
| 1366 indexing into arrays. | |
| 1367 | |
| 6593 | 1368 An example that demonstration how to work with arbitrary real or complex |
| 1369 double precision arrays is given by the file @file{mypow2.c} as given | |
| 1370 below. | |
| 1371 | |
| 9906 | 1372 @example |
| 1373 @EXAMPLEFILE(mypow2.c) | |
| 1374 @end example | |
| 6593 | 1375 |
| 1376 @noindent | |
| 1377 with an example of its use | |
| 1378 | |
| 1379 @example | |
| 1380 @group | |
| 1381 b = randn(4,1) + 1i * randn(4,1); | |
| 1382 all(b.^2 == mypow2(b)) | |
| 1383 @result{} 1 | |
| 1384 @end group | |
| 1385 @end example | |
| 1386 | |
| 1387 | |
| 7096 | 1388 The example above uses the functions @code{mxGetDimensions}, |
| 1389 @code{mxGetNumberOfElements}, and @code{mxGetNumberOfDimensions} to work | |
| 1390 with the dimensions of multi-dimensional arrays. The functions | |
| 1391 @code{mxGetM}, and @code{mxGetN} are also available to find the number | |
| 1392 of rows and columns in a matrix. | |
| 6580 | 1393 |
| 1394 @node Character Strings in Mex-Files | |
| 1395 @subsection Character Strings in Mex-Files | |
| 1396 | |
| 6593 | 1397 As mex-files do not make the distinction between single and double |
| 1398 quoted strings within Octave, there is perhaps less complexity in the | |
| 1399 use of strings and character matrices in mex-files. An example of their | |
| 1400 use, that parallels the demo in @file{stringdemo.cc}, is given in the | |
| 1401 file @file{mystring.c}, as seen below. | |
| 1402 | |
| 9906 | 1403 @example |
| 1404 @EXAMPLEFILE(mystring.c) | |
| 1405 @end example | |
| 6593 | 1406 |
| 1407 @noindent | |
| 1408 An example of its expected output is | |
| 1409 | |
| 1410 @example | |
| 1411 @group | |
| 1412 mystring(["First String"; "Second String"]) | |
| 1413 @result{} s1 = Second String | |
| 1414 First String | |
| 1415 @end group | |
| 1416 @end example | |
| 1417 | |
| 7096 | 1418 Other functions in the mex interface for handling character strings are |
| 1419 @code{mxCreateString}, @code{mxArrayToString}, and | |
| 1420 @code{mxCreateCharMatrixFromStrings}. In a mex-file, a character string | |
| 1421 is considered to be a vector rather than a matrix. This is perhaps an | |
| 1422 arbitrary distinction as the data in the mxArray for the matrix is | |
| 1423 consecutive in any case. | |
| 6580 | 1424 |
| 1425 @node Cell Arrays with Mex-Files | |
| 1426 @subsection Cell Arrays with Mex-Files | |
| 6569 | 1427 |
| 6593 | 1428 We can perform exactly the same operations in Cell arrays in mex-files |
| 1429 as we can in oct-files. An example that reduplicates the functional of | |
| 1430 the @file{celldemo.cc} oct-file in a mex-file is given by | |
| 1431 @file{mycell.c} as below | |
| 1432 | |
| 9906 | 1433 @example |
| 1434 @group | |
| 1435 @EXAMPLEFILE(mycell.c) | |
| 1436 @end group | |
| 1437 @end example | |
| 6593 | 1438 |
| 1439 @noindent | |
| 1440 which as can be seen below has exactly the same behavior as the oct-file | |
| 1441 version. | |
| 1442 | |
| 1443 @example | |
| 1444 @group | |
| 1445 [b1, b2, b3] = mycell (@{1, [1, 2], "test"@}) | |
| 1446 @result{} | |
| 1447 b1 = 1 | |
| 1448 b2 = | |
| 1449 | |
| 1450 1 2 | |
| 1451 | |
| 1452 b3 = test | |
| 1453 @end group | |
| 1454 @end example | |
| 1455 | |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1456 Note in the example the use of the @code{mxDuplicateArray} function. This |
| 6593 | 1457 is needed as the @code{mxArray} pointer returned by @code{mxGetCell} |
| 1458 might be deallocated. The inverse function to @code{mxGetCell} is | |
| 1459 @code{mcSetCell} and is defined as | |
| 1460 | |
| 1461 @example | |
| 1462 void mxSetCell (mxArray *ptr, int idx, mxArray *val); | |
| 1463 @end example | |
| 1464 | |
| 7007 | 1465 Finally, to create a cell array or matrix, the appropriate functions are |
| 6593 | 1466 |
| 1467 @example | |
| 1468 @group | |
| 1469 mxArray *mxCreateCellArray (int ndims, const int *dims); | |
| 1470 mxArray *mxCreateCellMatrix (int m, int n); | |
| 1471 @end group | |
| 1472 @end example | |
| 6569 | 1473 |
| 6572 | 1474 @node Structures with Mex-Files |
| 1475 @subsection Structures with Mex-Files | |
| 6569 | 1476 |
| 6593 | 1477 The basic function to create a structure in a mex-file is |
| 1478 @code{mxCreateStructMatrix}, which creates a structure array with a two | |
| 1479 dimensional matrix, or @code{mxCreateStructArray}. | |
| 1480 | |
| 1481 @example | |
| 1482 @group | |
| 7081 | 1483 mxArray *mxCreateStructArray (int ndims, int *dims, |
| 1484 int num_keys, | |
| 6593 | 1485 const char **keys); |
| 7081 | 1486 mxArray *mxCreateStructMatrix (int rows, int cols, |
| 1487 int num_keys, | |
| 6593 | 1488 const char **keys); |
| 1489 @end group | |
| 1490 @end example | |
| 1491 | |
| 1492 Accessing the fields of the structure can then be performed with the | |
| 1493 @code{mxGetField} and @code{mxSetField} or alternatively with the | |
| 1494 @code{mxGetFieldByNumber} and @code{mxSetFieldByNumber} functions. | |
| 1495 | |
| 1496 @example | |
| 1497 @group | |
| 7081 | 1498 mxArray *mxGetField (const mxArray *ptr, mwIndex index, |
| 1499 const char *key); | |
| 6593 | 1500 mxArray *mxGetFieldByNumber (const mxArray *ptr, |
| 6686 | 1501 mwIndex index, int key_num); |
| 1502 void mxSetField (mxArray *ptr, mwIndex index, | |
| 6593 | 1503 const char *key, mxArray *val); |
| 6686 | 1504 void mxSetFieldByNumber (mxArray *ptr, mwIndex index, |
| 6593 | 1505 int key_num, mxArray *val); |
| 1506 @end group | |
| 1507 @end example | |
| 1508 | |
| 1509 A difference between the oct-file interface to structures and the | |
| 1510 mex-file version is that the functions to operate on structures in | |
| 1511 mex-files directly include an @code{index} over the elements of the | |
| 1512 arrays of elements per @code{field}. Whereas the oct-file structure | |
| 1513 includes a Cell Array per field of the structure. | |
| 1514 | |
| 1515 An example that demonstrates the use of structures in mex-file can be | |
| 1516 found in the file @file{mystruct.c}, as seen below | |
| 6580 | 1517 |
| 9906 | 1518 @example |
| 1519 @EXAMPLEFILE(mystruct.c) | |
| 1520 @end example | |
| 6580 | 1521 |
| 6593 | 1522 An example of the behavior of this function within Octave is then |
| 1523 | |
| 1524 @example | |
| 7081 | 1525 a(1).f1 = "f11"; a(1).f2 = "f12"; |
| 1526 a(2).f1 = "f21"; a(2).f2 = "f22"; | |
| 6593 | 1527 b = mystruct(a) |
| 1528 @result{} field f1(0) = f11 | |
| 1529 field f1(1) = f21 | |
| 1530 field f2(0) = f12 | |
| 1531 field f2(1) = f22 | |
| 1532 b = | |
| 1533 @{ | |
| 1534 this = | |
| 1535 | |
| 1536 (, | |
| 1537 [1] = this1 | |
| 1538 [2] = this2 | |
| 1539 [3] = this3 | |
| 1540 [4] = this4 | |
| 1541 ,) | |
| 1542 | |
| 1543 that = | |
| 1544 | |
| 1545 (, | |
| 1546 [1] = that1 | |
| 1547 [2] = that2 | |
| 1548 [3] = that3 | |
| 1549 [4] = that4 | |
| 1550 ,) | |
| 1551 | |
| 1552 @} | |
| 1553 @end example | |
| 6569 | 1554 |
| 1555 @node Sparse Matrices with Mex-Files | |
| 1556 @subsection Sparse Matrices with Mex-Files | |
| 1557 | |
| 6593 | 1558 The Octave format for sparse matrices is identical to the mex format in |
| 7001 | 1559 that it is a compressed column sparse format. Also in both, sparse |
|
9209
923c7cb7f13f
Simplify TeXinfo files by eliminating redundant @iftex followed by @tex construction.
Rik <rdrider0-list@yahoo.com>
parents:
9080
diff
changeset
|
1560 matrices are required to be two-dimensional. The only difference is that |
| 6939 | 1561 the real and imaginary parts of the matrix are stored separately. |
| 6593 | 1562 |
| 1563 The mex-file interface, as well as using @code{mxGetM}, @code{mxGetN}, | |
| 1564 @code{mxSetM}, @code{mxSetN}, @code{mxGetPr}, @code{mxGetPi}, | |
| 1565 @code{mxSetPr} and @code{mxSetPi}, the mex-file interface supplies the | |
| 1566 functions | |
| 1567 | |
| 1568 @example | |
| 1569 @group | |
| 6686 | 1570 mwIndex *mxGetIr (const mxArray *ptr); |
| 1571 mwIndex *mxGetJc (const mxArray *ptr); | |
| 1572 mwSize mxGetNzmax (const mxArray *ptr); | |
| 6593 | 1573 |
| 6686 | 1574 void mxSetIr (mxArray *ptr, mwIndex *ir); |
| 1575 void mxSetJc (mxArray *ptr, mwIndex *jc); | |
| 1576 void mxSetNzmax (mxArray *ptr, mwSize nzmax); | |
| 6593 | 1577 @end group |
| 1578 @end example | |
| 6580 | 1579 |
| 6593 | 1580 @noindent |
| 1581 @code{mxGetNzmax} gets the maximum number of elements that can be stored | |
| 1582 in the sparse matrix. This is not necessarily the number of non-zero | |
| 1583 elements in the sparse matrix. @code{mxGetJc} returns an array with one | |
| 1584 additional value than the number of columns in the sparse matrix. The | |
| 1585 difference between consecutive values of the array returned by | |
| 1586 @code{mxGetJc} define the number of non-zero elements in each column of | |
| 1587 the sparse matrix. Therefore | |
| 6580 | 1588 |
| 6593 | 1589 @example |
| 1590 @group | |
| 6686 | 1591 mwSize nz, n; |
| 1592 mwIndex *Jc; | |
| 6593 | 1593 mxArray *m; |
| 1594 @dots{} | |
| 1595 n = mxGetN (m); | |
| 1596 Jc = mxGetJc (m); | |
| 1597 nz = Jc[n]; | |
| 1598 @end group | |
| 1599 @end example | |
| 1600 | |
| 1601 @noindent | |
| 1602 returns the actual number of non-zero elements stored in the matrix in | |
| 1603 @code{nz}. As the arrays returned by @code{mxGetPr} and @code{mxGetPi} | |
| 1604 only contain the non-zero values of the matrix, we also need a pointer | |
| 1605 to the rows of the non-zero elements, and this is given by | |
| 1606 @code{mxGetIr}. A complete example of the use of sparse matrices in | |
| 1607 mex-files is given by the file @file{mysparse.c} as seen below | |
| 1608 | |
| 9906 | 1609 @example |
| 1610 @EXAMPLEFILE(mysparse.c) | |
| 1611 @end example | |
| 6569 | 1612 |
| 6580 | 1613 @node Calling Other Functions in Mex-Files |
| 1614 @subsection Calling Other Functions in Mex-Files | |
| 1615 | |
| 1616 It is also possible call other Octave functions from within a mex-file | |
| 6593 | 1617 using @code{mexCallMATLAB}. An example of the use of |
| 6580 | 1618 @code{mexCallMATLAB} can be see in the example below |
| 1619 | |
| 9906 | 1620 @example |
| 1621 @EXAMPLEFILE(myfeval.c) | |
| 1622 @end example | |
| 6580 | 1623 |
| 1624 If this code is in the file @file{myfeval.c}, and is compiled to | |
| 1625 @file{myfeval.mex}, then an example of its use is | |
| 6569 | 1626 |
| 6580 | 1627 @example |
| 1628 @group | |
| 1629 myfeval("sin", 1) | |
| 1630 a = myfeval("sin", 1) | |
| 1631 @result{} Hello, World! | |
| 1632 I have 2 inputs and 1 outputs | |
| 1633 I'm going to call the interpreter function sin | |
| 1634 a = 0.84147 | |
| 1635 @end group | |
| 1636 @end example | |
| 1637 | |
| 1638 Note that it is not possible to use function handles or inline functions | |
| 1639 within a mex-file. | |
| 1640 | |
| 6593 | 1641 @c @node Application Programming Interface for Mex-Files |
| 1642 @c @subsection Application Programming Interface for Mex-Files | |
| 1643 @c | |
| 1644 @c WRITE ME, refer to mex.h and mexproto.h | |
| 6569 | 1645 |
| 1646 @node Standalone Programs | |
| 1647 @section Standalone Programs | |
| 1648 | |
| 1649 The libraries Octave itself uses, can be utilized in standalone | |
| 6571 | 1650 applications. These applications then have access, for example, to the |
| 6569 | 1651 array and matrix classes as well as to all the Octave algorithms. The |
| 1652 following C++ program, uses class Matrix from liboctave.a or | |
| 1653 liboctave.so. | |
| 1654 | |
| 9906 | 1655 @example |
| 1656 @group | |
| 1657 @EXAMPLEFILE(standalone.cc) | |
| 1658 @end group | |
| 1659 @end example | |
| 6569 | 1660 |
| 6580 | 1661 @noindent |
| 6569 | 1662 mkoctfile can then be used to build a standalone application with a |
| 1663 command like | |
| 1664 | |
| 1665 @example | |
| 1666 @group | |
|
8097
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1667 $ mkoctfile --link-stand-alone standalone.cc -o standalone |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1668 $ ./standalone |
| 6569 | 1669 Hello Octave world! |
| 1670 11 12 | |
| 1671 21 22 | |
| 1672 $ | |
| 1673 @end group | |
| 1674 @end example | |
| 1675 | |
| 1676 Note that the application @code{hello} will be dynamically linked | |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1677 against the octave libraries and any octave support libraries. The above |
|
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1678 allows the Octave math libraries to be used by an application. It does |
|
8097
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1679 not however allow the script files, oct-files or builtin functions of |
|
9080
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1680 Octave to be used by the application. To do that the Octave interpreter |
|
ec41eabf4499
Cleanup documentation files dynamic.texi, testfun.texi, tips.texi
Rik <rdrider0-list@yahoo.com>
parents:
9038
diff
changeset
|
1681 needs to be initialized first. An example of how to do this can then be |
|
8097
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1682 seen in the code |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1683 |
| 9906 | 1684 @example |
| 1685 @group | |
| 1686 @EXAMPLEFILE(embedded.cc) | |
| 1687 @end group | |
| 1688 @end example | |
|
8097
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1689 |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1690 @noindent |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1691 which is compiled and run as before as a standalone application with |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1692 |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1693 @example |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1694 @group |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1695 $ mkoctfile --link-stand-alone embedded.cc -o embedded |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1696 $ ./embedded |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1697 GCD of [10, 15] is 5 |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1698 $ |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1699 @end group |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1700 @end example |
|
804c60f92fb1
Add explanationation of initializing the interpreter in a standalone program
David Bateman <dbateman@free.fr>
parents:
7354
diff
changeset
|
1701 |