| 1 |
% \iffalse |
| 2 |
% |
| 3 |
% functan.dtx |
| 4 |
% Copyright 2004 Antoine Lejay |
| 5 |
% |
| 6 |
% This program is provided under the terms of the |
| 7 |
% LaTeX Project Public License distributed from CTAN |
| 8 |
% archives in directory macros/latex/base/lppl.txt. |
| 9 |
% |
| 10 |
% Author: Antoine Lejay |
| 11 |
% \fi |
| 12 |
% \CheckSum{360} |
| 13 |
%% \CharacterTable |
| 14 |
%% {Upper-case \A\B\C\D\E\F\G\H\I\J\K\L\M\N\O\P\Q\R\S\T\U\V\W\X\Y\Z |
| 15 |
%% Lower-case \a\b\c\d\e\f\g\h\i\j\k\l\m\n\o\p\q\r\s\t\u\v\w\x\y\z |
| 16 |
%% Digits \0\1\2\3\4\5\6\7\8\9 |
| 17 |
%% Exclamation \! Double quote \" Hash (number) \# |
| 18 |
%% Dollar \$ Percent \% Ampersand \& |
| 19 |
%% Acute accent \' Left paren \( Right paren \) |
| 20 |
%% Asterisk \* Plus \+ Comma \, |
| 21 |
%% Minus \- Point \. Solidus \/ |
| 22 |
%% Colon \: Semicolon \; Less than \< |
| 23 |
%% Equals \= Greater than \> Question mark \? |
| 24 |
%% Commercial at \@ Left bracket \[ Backslash \\ |
| 25 |
%% Right bracket \] Circumflex \^ Underscore \_ |
| 26 |
%% Grave accent \` Left brace \{ Vertical bar \| |
| 27 |
%% Right brace \} Tilde \~} |
| 28 |
%% |
| 29 |
% \iffalse |
| 30 |
%<*dtx> |
| 31 |
\ProvidesFile{functan.dtx} |
| 32 |
%</dtx> |
| 33 |
%<package>\ProvidesPackage{functan} |
| 34 |
%<driver>\ProvidesFile{functan.drv} |
| 35 |
% \fi |
| 36 |
% \ProvidesFile{functan.dtx}% |
| 37 |
[2004/07/03 v1.0 Macros for functional analysis] |
| 38 |
% \iffalse |
| 39 |
%<*driver> |
| 40 |
\documentclass[11pt]{ltxdoc} |
| 41 |
\usepackage{functan} |
| 42 |
\EnableCrossrefs |
| 43 |
\CodelineIndex |
| 44 |
\RecordChanges |
| 45 |
%%\OnlyDescription |
| 46 |
\begin{document} |
| 47 |
\DocInput{functan.dtx} |
| 48 |
\end{document} |
| 49 |
%</driver> |
| 50 |
% \fi |
| 51 |
% \def\docdate{July, 13 2004} |
| 52 |
% \DoNotIndex{\newcommand,\fi,\def,\else,\edef,\ifx,\let} |
| 53 |
% \DoNotIndex{\global,\NeedsTeXFormat,\;,\ProcessOptions,\ProvidesPackage} |
| 54 |
% \DoNotIndex{\PassOptionsToPackage} |
| 55 |
% \GetFileInfo{functan.dtx} |
| 56 |
% \MakeShortVerb{\"} |
| 57 |
% \DeleteShortVerb{\|} |
| 58 |
% \Macro{L2}{\mathrm{L}^2} |
| 59 |
% \Macro{H10}{\mathrm{H}^1_0} |
| 60 |
% \Macro{L2Om}{\m{L2}(\Omega)} |
| 61 |
% \Macro{H10Om}{\m{H10}(\Omega)} |
| 62 |
% \newnorm{aver}{\delonearg[#1]{#2}{\langle}{\rangle}} |
| 63 |
% \newconv{*L2Om}{\conv*{\star\text{-}\m{L2Om}}{#1}} |
| 64 |
% \newcommand{\itm}{\medskip\noindent The command \ignorespaces} |
| 65 |
% \newcommand{\itms}{\medskip\noindent The commands \ignorespaces} |
| 66 |
% \newenvironment{examplenotab}{\medskip\hrule\noindent\strut\ignorespaces}{\par\smallskip\hrule} |
| 67 |
% \newenvironment{example}{\medskip\hrule\noindent\begin{tabular}{@{\strut}l l}}{\end{tabular}\hrule} |
| 68 |
% \newcommand{\event}[2][auto]{\delonearg[#1]{#2}{\{}{\}}} |
| 69 |
% \let\package\textsf |
| 70 |
% \newcommand{\Nota}{\medskip\noindent\textit{Nota: }\ignorespaces} |
| 71 |
% |
| 72 |
% |
| 73 |
% \title{\package{functan}\thanks{Package version \fileversion{} of |
| 74 |
% \filedate{}.}\\ Macros for functional analysis and related domains} |
| 75 |
% \author{Antoine Lejay\\ |
| 76 |
% $<$\texttt{[email protected]}$>$} |
| 77 |
% \date{\docdate\\ Printed on \today} |
| 78 |
% \maketitle |
| 79 |
% \section{Introduction} |
| 80 |
% |
| 81 |
% This package is designed especially for the people |
| 82 |
% working in domains of mathematics such |
| 83 |
% as functional analysis and PDE theory. Of course, |
| 84 |
% this package may be used in other domains. |
| 85 |
% |
| 86 |
% In functional analysis and PDE theory, one may face |
| 87 |
% a lot of names of spaces, sometimes depending |
| 88 |
% on one or more parameters, such as $\m{L2Om}$, $\m{H10Om}$, ... |
| 89 |
% Besides, one has to deal |
| 90 |
% with norms, convergence and scalar product |
| 91 |
% for each of these spaces. |
| 92 |
% As a control sequence, i.e., a \TeX{} command, consists only |
| 93 |
% of letters, using \TeX's standard macros is not convenient |
| 94 |
% for dealing with such spaces. Yet |
| 95 |
% composing each of these space names at each their |
| 96 |
% occurrence increases the risk of errors, |
| 97 |
% and tends to be tedious. |
| 98 |
% |
| 99 |
% Fortunately, \TeX{} provides some way to |
| 100 |
% create commands whose name may contain |
| 101 |
% arbitrary characters. We exploit this |
| 102 |
% possibility to deal with names of |
| 103 |
% functional spaces and related objects |
| 104 |
% (norms, convergence, ...). |
| 105 |
% |
| 106 |
% This package allows for example to typeset: |
| 107 |
% \begin{quote} |
| 108 |
% We denote by $\m{L2Om}$ the space of measurable |
| 109 |
% functions on $\Omega$ which are square integrable. |
| 110 |
% The associated norm $\norm{L2Om}{\cdot}$ is defined |
| 111 |
% by $\norm{L2Om}{f}=\left(\int_\Omega f(x)^2 dx\right)^{1/2}$ |
| 112 |
% and the corresponding scalar product is |
| 113 |
% $\scalprod{L2Om}{f}{g}=\int_\Omega f(x)g(x)dx$. |
| 114 |
% We write $f_n\conv{L2Om}{n\to\infty}f$ |
| 115 |
% if the sequence $(f_n)_n$ converges to $f$ in $\m{L2Om}$. |
| 116 |
% If for all $g\in\m{L2Om}$, $\scalprod{L2Om}{f_n}{g}$ |
| 117 |
% converges to $\scalprod{L2Om}{f}{g}$, then |
| 118 |
% we say that $(f_n)_n$ converges weakly to $f$ |
| 119 |
% in $\m{L2Om}$, and we write $f_n\wkconv{L2Om}{n\to\infty}f$. |
| 120 |
% \end{quote} |
| 121 |
% The code for this example is the following: |
| 122 |
% \begin{verbatim} |
| 123 |
% \Macro{L2Om}{\mathrm{L}^2(\Omega)} |
| 124 |
% |
| 125 |
% We denote by $\m{L2Om}$ the space of measurable |
| 126 |
% functions on $\Omega$ which are square integrable. |
| 127 |
% The associated norm $\norm{L2Om}{\cdot}$ is defined |
| 128 |
% by $\norm{L2Om}{f}=\left(\int_\Omega f(x)^2 dx\right)^{1/2}$ |
| 129 |
% and the corresponding scalar product is |
| 130 |
% $\scalprod{L2Om}{f}{g}=\int_\Omega f(x)g(x)dx$. |
| 131 |
% We write $f_n\conv{L2Om}{n\to\infty}f$ |
| 132 |
% if the sequence $(f_n)_n$ converges to $f$ in $\m{L2Om}$. |
| 133 |
% If for all $g\in\m{L2Om}$, $\scalprod{L2Om}{f_n}{g}$ |
| 134 |
% converges to $\scalprod{L2Om}{f}{g}$, then |
| 135 |
% we say that $(f_n)_n$ converges weakly to $f$ |
| 136 |
% in $\m{L2Om}$, and we write $f_n\wkconv{L2Om}{n\to\infty}f$. |
| 137 |
% \end{verbatim} |
| 138 |
% |
| 139 |
% \begin{center} |
| 140 |
% \fbox{\begin{minipage}{0.75\textwidth} This package |
| 141 |
% uses of some commands of the package |
| 142 |
% \package{amsmath} (some of these commands are available |
| 143 |
% only for version \textbf{2.0} or higher), which is then automatically |
| 144 |
% loaded. All the options of \package{amsmath} |
| 145 |
% may be defined as options of \package{functan}. |
| 146 |
% |
| 147 |
% Before using \package{functan}, |
| 148 |
% be sure you have a recent version of \package{amsmath} |
| 149 |
% installed in your computer. |
| 150 |
% \end{minipage}} |
| 151 |
% \end{center} |
| 152 |
% |
| 153 |
% \section{Loading the package} |
| 154 |
% |
| 155 |
% This package is loaded as any package by the command |
| 156 |
% "\usepackage". This package has no options, except |
| 157 |
% those of \package{amsmath} in order to avoir |
| 158 |
% an option clash. Thus, the possible options |
| 159 |
% are: |
| 160 |
% "leqno", |
| 161 |
% "intlimits", |
| 162 |
% "nointlimits", |
| 163 |
% "sumlimits", |
| 164 |
% "nosumlimits", |
| 165 |
% "namelimits", |
| 166 |
% "nonamelinits", |
| 167 |
% "reqno", |
| 168 |
% "centertags", |
| 169 |
% "tbtags", |
| 170 |
% "cmex10" and |
| 171 |
% "fleqn". |
| 172 |
% |
| 173 |
% \medskip |
| 174 |
% |
| 175 |
% \noindent\textbf{Incompatibilities : } This package |
| 176 |
% has no known incompatibilities with other packages. |
| 177 |
% |
| 178 |
% \section{The commands} |
| 179 |
% |
| 180 |
% \subsection{An alternative system of macros} |
| 181 |
% |
| 182 |
% This system consists in associating a sequence |
| 183 |
% of tokens to a name by the command "\Macro", |
| 184 |
% and call it later by the command "\m". |
| 185 |
% |
| 186 |
% |
| 187 |
% \itm "\Macro"\marg{name}\marg{expansion} |
| 188 |
% \DescribeMacro{\Macro} |
| 189 |
% associates the list of tokens \meta{expansion} |
| 190 |
% to the list of tokens |
| 191 |
% \meta{name}, which may contain any arbitrary |
| 192 |
% sequence of characters (except~"{", "}" and |
| 193 |
% active characters). This macro produces |
| 194 |
% a warning message ``The macro `\meta{name}' has |
| 195 |
% already been defined'' if \meta{name} already exists. |
| 196 |
% |
| 197 |
% \Nota The sequence \meta{name} may correspond to the control |
| 198 |
% sequence of a \TeX{} macro without creating interferences. |
| 199 |
% Internally, the system stores the expansion |
| 200 |
% of \meta{name} as the \TeX{} control sequence |
| 201 |
% "functan@macro@"\meta{name}. |
| 202 |
% |
| 203 |
% \itm "\m"\marg{name} gives the expansion corresponding |
| 204 |
% \DescribeMacro{\m} |
| 205 |
% to \meta{name} if a couple \meta{name} and \meta{expansion} |
| 206 |
% has already been defined by |
| 207 |
% "\Macro". Otherwise, an error message |
| 208 |
% ``The macro `\meta{name}' is not defined'' is produced. |
| 209 |
% The error message ``No macro's name given'' is |
| 210 |
% also produced if the argument is given, that is |
| 211 |
% if "\m{}" is encountered. |
| 212 |
% |
| 213 |
% |
| 214 |
% \begin{examplenotab} |
| 215 |
% "\Macro{L2Om}{\mathrm{L}^2(\Omega)}"\\ |
| 216 |
% "Denote by $\m{L2Om}$ the space of functions such that..."\\ |
| 217 |
% \hbox{}\quad |
| 218 |
% Denote by $\m{L2Om}$ the space of functions such that... |
| 219 |
% \end{examplenotab} |
| 220 |
% |
| 221 |
% \Nota A command "\mempty" |
| 222 |
% is also defined whose action is similar to |
| 223 |
% "\m", except that it does not produce |
| 224 |
% an error message if its argument is empty. |
| 225 |
% |
| 226 |
% \subsection{Norms} |
| 227 |
% |
| 228 |
% The command "\norm" allows to typeset |
| 229 |
% norms easily. |
| 230 |
% \begin{table} |
| 231 |
% \begin{center} |
| 232 |
% \begin{tabular}{l l l l l l} |
| 233 |
% \hline |
| 234 |
% (empty string) & \texttt{auto} & |
| 235 |
% \texttt{big} & \texttt{Big} & \texttt{bigg} & \texttt{Bigg}\\ |
| 236 |
% \hline |
| 237 |
% \end{tabular} |
| 238 |
% \caption{\label{table-1} Possible arguments for the size} |
| 239 |
% \end{center} |
| 240 |
% \end{table} |
| 241 |
% |
| 242 |
% \itm "\norm"\oarg{size}\marg{name}\marg{arg} |
| 243 |
% \DescribeMacro{\norm} |
| 244 |
% is roughly an equivalent to "\|"\meta{arg}"\|_{\m"\marg{name}"}" |
| 245 |
% but with a control on the size of the vertical bars. |
| 246 |
% If \meta{name} is empty, |
| 247 |
% then nothing is put in subscript. |
| 248 |
% The optional argument \meta{size} is one of the |
| 249 |
% element in Table~\ref{table-1}. |
| 250 |
% By default, \meta{size} is set to \texttt{auto}, |
| 251 |
% which is equivalent in using "\left\|"..."\right\|" |
| 252 |
% around \meta{arg}. |
| 253 |
% If \meta{size} is empty, that is when one call |
| 254 |
% "\norm[]{...}{...}", then the delimitor $\|$ |
| 255 |
% is used with its normal size. If \meta{size} is equal to |
| 256 |
% \texttt{big}, \texttt{Big}, \texttt{bigg} or \texttt{Bigg}, |
| 257 |
% then \meta{size} is inserted before the delimitors $\|$. |
| 258 |
% |
| 259 |
% \Nota If \meta{size} is not one of the elements of Table~\ref{table-1}, |
| 260 |
% the the normal size is used. |
| 261 |
% |
| 262 |
% \begin{example} |
| 263 |
% "\norm{}{f}" & $\norm{}{f}$\\ |
| 264 |
% "\norm{L2Om}{f}" & $\norm{L2Om}{f}$\\ |
| 265 |
% "\norm[Big]{L2Om}{f}" & $\norm[Big]{L2Om}{f}$\\ |
| 266 |
% "\norm{L2Om}{\frac{1}{f}}" & $\norm{L2Om}{\frac{1}{f}}$\\ |
| 267 |
% "\norm[]{L2Om}{\frac{1}{f}}" & $\norm[]{L2Om}{\frac{1}{f}}$\\ |
| 268 |
% \end{example} |
| 269 |
% |
| 270 |
% \itm "\norm*"\oarg{size}\marg{subscr}\marg{arg} |
| 271 |
% \DescribeMacro{\norm*} |
| 272 |
% works as "\norm", except that \meta{subscr} is not |
| 273 |
% one of the macro defined by "\Macro", |
| 274 |
% but any sequence of tokens which is put in subscript. |
| 275 |
% For example "\norm*{\infty}{f}" is equivalent |
| 276 |
% to "\norm{}{f}_{\infty}". |
| 277 |
% |
| 278 |
% \begin{example} |
| 279 |
% "\norm*{\infty}{f}" & $\norm*{\infty}{f}$\\ |
| 280 |
% "\norm*[Big]{\infty}{f}" & $\norm*[Big]{\infty}{f}$\\ |
| 281 |
% \end{example} |
| 282 |
% |
| 283 |
% \itm "\newnorm"\marg{name}\marg{expansion} |
| 284 |
% \DescribeMacro{\newnorm} |
| 285 |
% allows to override the behavior of the "\norm" macro. |
| 286 |
% Instead of using a "\|"..."\|" |
| 287 |
% structure, the user may for example use other delimitors. |
| 288 |
% Here \meta{name} is any sequence of tokens (except |
| 289 |
% "{", "}" and active characters), already |
| 290 |
% defined by "\Macro" or not. |
| 291 |
% The arguments "#1" (for the size) et "#2" (for the argument) may be |
| 292 |
% used in \meta{expansion}. The macros |
| 293 |
% "\casesize" and "\delonearg" (see Section~\ref{sub-macro} below) |
| 294 |
% may be used to create some new kind of norms. |
| 295 |
% |
| 296 |
% \newnorm{L2Om}{\delonearg[#1]{#2}{|}{|}} |
| 297 |
% \begin{example} |
| 298 |
% "\newnorm{L2Om}{\delonearg[#1]{#2}{|}{|}}" \\ |
| 299 |
% "\norm{L2Om}{f}" & $\norm{L2Om}{f}$\\ |
| 300 |
% "\norm[Big]{L2Om}{f}" & $\norm[Big]{L2Om}{f}$\\ |
| 301 |
% \end{example} |
| 302 |
% |
| 303 |
% \Nota It seems that some side effects |
| 304 |
% appear when "\newnorm" is used inside a group. |
| 305 |
% Hence, it is recommended to use this command |
| 306 |
% in the preamble. |
| 307 |
% |
| 308 |
% |
| 309 |
% |
| 310 |
% \subsection{Scalar products} |
| 311 |
% |
| 312 |
% Macros for scalar products are similar to |
| 313 |
% macros for the norms, except that two |
| 314 |
% argument are needed. |
| 315 |
% |
| 316 |
% \itm "\scalprod"\oarg{size}\marg{name}\marg{arg 1}\marg{arg 2} |
| 317 |
% \DescribeMacro{\scalprod} |
| 318 |
% creates roughly |
| 319 |
% "\langle"\meta{arg 1}","\meta{arg 2}"\rangle_{\m"\marg{name}"}". |
| 320 |
% The optional |
| 321 |
% \meta{size} argument shall be taken in Table~\ref{table-1}, |
| 322 |
% and acts as a modifier for the size of the brackets. |
| 323 |
% |
| 324 |
% \begin{example} |
| 325 |
% "\scalprod{}{f}{g}" & $\scalprod{}{f}{g}$\\ |
| 326 |
% "\scalprod{L2Om}{f}{g}" & $\scalprod{L2Om}{f}{g}$\\ |
| 327 |
% "\scalprod{L2Om}{f}{\frac{1}{g}}" & $\scalprod{L2Om}{f}{\frac{1}{g}}$\\ |
| 328 |
% "\scalprod[big]{L2Om}{f}{\frac{1}{g}}" & $\scalprod[big]{L2Om}{f}{\frac{1}{g}}$ |
| 329 |
% \end{example} |
| 330 |
% |
| 331 |
% \itm "\scalprod*"\oarg{size}\marg{subscr}\marg{arg 1}\marg{arg 2} |
| 332 |
% \DescribeMacro{\scalprod*} |
| 333 |
% acts like "\scalprod", except that \meta{subscr} is |
| 334 |
% not the name already defined by "\Macro", but |
| 335 |
% a sequence of tokens put in subscript after the closing |
| 336 |
% bracket. The rules are the same as for "\norm". |
| 337 |
% |
| 338 |
% \begin{example} |
| 339 |
% "\scalprod*{X,Y}{Ax}{y}" & $\scalprod*{X,Y}{Ax}{y}$\\ |
| 340 |
% "\scalprod*{X,Y}{B^{-1}x}{y}" & $\scalprod*{X,Y}{B^{-1}x}{y}$\\ |
| 341 |
% "\scalprod*[]{X,Y}{B^{-1}x}{y}" & $\scalprod*[]{X,Y}{B^{-1}x}{y}$ |
| 342 |
% \end{example} |
| 343 |
% |
| 344 |
% \itm "\newscalprod"\marg{name}\marg{expansion} |
| 345 |
% \DescribeMacro{\newscalprod} |
| 346 |
% allows to override the behavior of "\scalprod", |
| 347 |
% as for "\newnorm". Here, \meta{expansion} may |
| 348 |
% used the three arguments "#1" (for the optional size), |
| 349 |
% "#2" (\meta{arg 1}) and "#3" (\meta{arg 2}). |
| 350 |
% The command "\towarg" (see Section~\ref{sub-macro} below) |
| 351 |
% may be used to help the user to deal with new types |
| 352 |
% of scalar products. |
| 353 |
% |
| 354 |
% \newscalprod{L2}{\deltwoarg[#1]{#2}{#3}{(}{)}{|}} |
| 355 |
% \begin{examplenotab} |
| 356 |
% \noindent\strut"\newscalprod{L2Om}{\deltwoarg[#1]{#2}{#3}{(}{)}{|}}" |
| 357 |
% |
| 358 |
% \noindent\begin{tabular}{@{\strut} l l} |
| 359 |
% "\scalprod{L2}{f}{g}" & $\scalprod{L2}{f}{g}$ \\ |
| 360 |
% "\scalprod[Big]{L2}{f}{g}" & $\scalprod[Big]{L2}{f}{g}$\\ |
| 361 |
% "\scalprod[]{L2}{f^2}{g^2}" & $\scalprod[]{L2}{f^2}{g^2}$ |
| 362 |
% \end{tabular} |
| 363 |
% \end{examplenotab} |
| 364 |
% |
| 365 |
% \Nota It seems that some side effects |
| 366 |
% appear when "\newnorm" is used inside a group. |
| 367 |
% Hence, it is recommended to used this command |
| 368 |
% in the preamble. |
| 369 |
% |
| 370 |
% \subsection{Convergence} |
| 371 |
% The package \package{amsmath} provides a way |
| 372 |
% to create \emph{extensible arrows}, whose |
| 373 |
% size depends on the material put above and |
| 374 |
% below the arrow. |
| 375 |
% |
| 376 |
% \itm "\conv"\marg{name}\marg{below} |
| 377 |
% \DescribeMacro{\conv} |
| 378 |
% creates an extensible arrow by putting the |
| 379 |
% expansion of \meta{name} above the arrow, |
| 380 |
% and \meta{below} under the arrow. |
| 381 |
% This command uses the commands "\xrightarrow" of |
| 382 |
% \package{amsmath}. However, the "\xrightarrow" command |
| 383 |
% takes as an optional argument the material to be put |
| 384 |
% below. Here, it is a mandatory second argument. |
| 385 |
% |
| 386 |
% \begin{example} |
| 387 |
% "x_n\conv{}{} x" & $x_n\conv{}{} x$\\ |
| 388 |
% "f_n\conv{}{n\to\infty} f" & $f_n\conv{}{n\to\infty} f$\\ |
| 389 |
% "f_n\conv{L2Om}{n\to\infty} f" & $f_n\conv{L2Om}{n\to\infty} f$\\ |
| 390 |
% \end{example} |
| 391 |
% |
| 392 |
% \itm "\conv*"\marg{above}\marg{below} |
| 393 |
% \DescribeMacro{\conv*} |
| 394 |
% is similar to "\conv", except that both |
| 395 |
% \meta{above} and \meta{below} are lists of tokens |
| 396 |
% that are put respectively above and below the arrow. |
| 397 |
% |
| 398 |
% \begin{example} |
| 399 |
% "f_n\conv*{\text{uniformly}}{n\to\infty} f" & $f_n\conv*{\text{uniformly}}{n\to\infty} f$ |
| 400 |
% \end{example} |
| 401 |
% |
| 402 |
% \itm "\newconv"\marg{name}\marg{expansion} allows |
| 403 |
% \DescribeMacro{\newconv} |
| 404 |
% to override the behavior of "\conv"\marg{name}, |
| 405 |
% whether \meta{name} corresponds to a name |
| 406 |
% defined by "\Macro" or not. The token list |
| 407 |
% \meta{expansion} may use one argument "#1". |
| 408 |
% |
| 409 |
% \begin{examplenotab} |
| 410 |
% "\newconv{*L2Om}{\conv*{\star\text{-}\m{L2Om}}{#1}}"\\ |
| 411 |
% "f_n\conv{*L2Om}{n\to\infty} f" \quad $f_n\conv{*L2Om}{n\to\infty}f$\\ |
| 412 |
% \end{examplenotab} |
| 413 |
% |
| 414 |
% \itms "\leftconv"\marg{name}\marg{below} |
| 415 |
% \DescribeMacro{\leftconv} |
| 416 |
% \DescribeMacro{\leftconv*} |
| 417 |
%\\ and "\leftconv*"\marg{above}\marg{below} |
| 418 |
% act like "\conv" and "\conv*", except that |
| 419 |
% an arrow pointing to the left ($\leftconv{}{}$) is used instead |
| 420 |
% of an arrow pointing to the right ($\conv{}{}$). |
| 421 |
% |
| 422 |
% \itms "\leftrightconv"\marg{name}\marg{below} |
| 423 |
% \DescribeMacro{\leftrightconv} |
| 424 |
% \DescribeMacro{\leftrightconv*} |
| 425 |
% \\and "\leftrightconv*"\marg{above}\marg{below} |
| 426 |
% act like "\conv" and "\conv*", except that |
| 427 |
% an arrow pointing both to the left and to |
| 428 |
% the right ($\leftrightconv{}{}$) is used instead |
| 429 |
% of an arrow pointing to the right ($\conv{}{}$). |
| 430 |
% |
| 431 |
% |
| 432 |
% There are other commands similar to "\conv", |
| 433 |
% "\conv*", "\leftconv", "\leftconv*", |
| 434 |
% "\leftrightconv" and "\leftrightconv*" |
| 435 |
% that is with the same arguments, but with |
| 436 |
% different types of arrows. |
| 437 |
% |
| 438 |
% \itms "\wkconv", "\wkconv*", "\leftwkconv", |
| 439 |
% \DescribeMacro{\wkconv} |
| 440 |
% \DescribeMacro{\wkconv*} |
| 441 |
% \DescribeMacro{\leftwkconv} |
| 442 |
% \DescribeMacro{\leftwkconv*} |
| 443 |
% \DescribeMacro{\leftrightwkconv} |
| 444 |
% \DescribeMacro{\leftrightwkconv*} |
| 445 |
% "\leftwkconv*",\\ |
| 446 |
% "\leftrightwkconv" and "\leftrightwkconv*" |
| 447 |
% use $\wkconv{}{}$ instead of~$\conv{}{}$. Such an arrow |
| 448 |
% is generally used to denote \emph{weak convergence} |
| 449 |
% (hence the "wk"). |
| 450 |
% |
| 451 |
% \begin{example} |
| 452 |
% "f_n\wkconv{}{n\to\infty} f" & |
| 453 |
% $f_n\wkconv{}{n\to\infty} f$\\ |
| 454 |
% "f_n\wkconv{H10}{n\to\infty} f" & |
| 455 |
% $f_n\wkconv{H10}{n\to\infty} f$ |
| 456 |
% \end{example} |
| 457 |
% |
| 458 |
% \itms "\Conv", "\Conv*", "\Leftconv", "\Leftconv*", |
| 459 |
% \DescribeMacro{\Conv} |
| 460 |
% \DescribeMacro{\Conv*} |
| 461 |
% \DescribeMacro{\Leftconv} |
| 462 |
% \DescribeMacro{\Leftconv*} |
| 463 |
% \DescribeMacro{\Leftrightconv} |
| 464 |
% \DescribeMacro{\Leftrightconv*} |
| 465 |
% "\Leftrightconv" and "\Leftrightconv*" |
| 466 |
% use $\Conv{}{}$ instead of~$\conv{}{}$. |
| 467 |
% This may be used to denote convergence in |
| 468 |
% distribution in probability theory. |
| 469 |
% This could also be used for the implication |
| 470 |
% and equivalence symbols in logic. |
| 471 |
% \begin{examplenotab} |
| 472 |
% "\mu_n\Conv{}{n\to\infty} \mu" \quad |
| 473 |
% $\mu_n\Conv{}{n\to\infty} \mu$\\ |
| 474 |
% "x<0\Leftrightconv*{\text{Lemma 1.1}}{} f(x)\geq 0"\\ |
| 475 |
% \hbox{}\quad $x<0\Leftrightconv*{\text{Lemma 1.1}}{} f(x)\geq 0$ |
| 476 |
% \end{examplenotab} |
| 477 |
% |
| 478 |
% \subsection{Defining macros} |
| 479 |
% |
| 480 |
% \label{sub-macro} |
| 481 |
% |
| 482 |
% The package \package{functan} provides a few |
| 483 |
% macros to help the user to define its own |
| 484 |
% norms, scalar products, ... |
| 485 |
% |
| 486 |
% \itm "\delonearg"\oarg{size}\marg{arg}\marg{left del}\marg{right del} |
| 487 |
% \DescribeMacro{\delonearg} |
| 488 |
%\\ surrounds \meta{arg} the left delimitor |
| 489 |
% \meta{left del} and the right delimitor \meta{right del}. |
| 490 |
% The possible values of the optional argument \meta{size} (which is by |
| 491 |
% default set to "auto") are given in Table~\ref{table-1}. |
| 492 |
% \newcommand{\ket}[2][auto]{\delonearg[#1]{#2}{|}{\rangle}} |
| 493 |
% \newcommand{\bra}[2][auto]{\delonearg[#1]{#2}{\langle}{|}} |
| 494 |
% \begin{examplenotab} |
| 495 |
% "\newcommand{\event}[2][auto]{\delonearg[#1]{#2}{\{}{\}}}"\\ |
| 496 |
% "\event{1,\dotsc,\frac{1}{n}}"\\ |
| 497 |
% "\cup\event[]{\frac{1}{n^2},\dotsc,\frac{2}{n^2}}"\\ |
| 498 |
% \hbox{}\quad$\event{1,\dotsc,\frac{1}{n}}\cup\event[]{\frac{1}{n},\dotsc,\frac{2}{n}}$\\ |
| 499 |
% "\newcommand{\ket}[2][auto]{\delonearg[#1]{#2}{|}{\rangle}}"\\ |
| 500 |
% "\ket{Ax}" \quad $\ket{Ax}$\\ |
| 501 |
% "\newcommand{\bra}[2][auto]{\delonearg[#1]{#2}{\langle}{|}}"\\ |
| 502 |
% "\bra{Ax}" \quad $\bra{Ax}$ |
| 503 |
% \end{examplenotab} |
| 504 |
% |
| 505 |
% \itm "\deltwoarg"\oarg{size}\marg{arg 1}\marg{arg 2}... |
| 506 |
% \DescribeMacro{\deltwoarg} |
| 507 |
% \\...\marg{left del}\marg{right del}\marg{sep} |
| 508 |
% is similar to "\delonearg", except that it |
| 509 |
% separates \meta{arg 1} and \meta{arg 2} |
| 510 |
% by a separator \meta{sep.}. If |
| 511 |
% \meta{sep} is set to "|", then a vertical bar |
| 512 |
% is used, with an automatic adjustment to the size |
| 513 |
% of the delimitors. |
| 514 |
% |
| 515 |
% \newcommand{\set}[3][auto]{\deltwoarg[#1]{#2}{#3}{\{}{\}}{\,;\,}} |
| 516 |
% \newcommand{\braket}[3][auto]{\deltwoarg[#1]{#2}{#3}{\langle}{\rangle}{|}} |
| 517 |
% \begin{examplenotab} |
| 518 |
% "\newcommand{\set}[3][auto]{\deltwoarg[#1]{#2}{#3}{\{}{\}}{\,;\,}}"\\ |
| 519 |
% "\set{x\geq 0}{P(x)=0}" \quad $\set{x\geq 0}{P(x)=0}$\\ |
| 520 |
% "\newcommand{\braket}[3][auto]{%"\\ |
| 521 |
% "\deltwoarg[#1]{#2}{#3}{\langle}{\rangle}{|}}"\\ |
| 522 |
% "\braket{Ax}{y}" \quad $\braket{Ax}{y}$ |
| 523 |
% \end{examplenotab} |
| 524 |
% |
| 525 |
% \itm "\casesize"\marg{size}\marg{auto}\marg{empty}... |
| 526 |
% \DescribeMacro{\casesize} |
| 527 |
%\\...\marg{big}\marg{Big}\marg{bigg}\marg{Bigg} |
| 528 |
% takes \meta{size} in Table~\ref{table-1} |
| 529 |
% and executes \marg{empty} if \meta{size} is an empty |
| 530 |
% list of tokens, \meta{auto} if \meta{size} is "auto", |
| 531 |
% \meta{big} if \meta{size} is "big", ... |
| 532 |
% The error message ``Size argument `\meta{size}' of `casesize' not valid'' |
| 533 |
% is produced if \meta{size} is not in Table~\ref{table-1} |
| 534 |
% |
| 535 |
% \itm "\delcasesize"\marg{size}\marg{auto}\marg{empty}\marg{other} |
| 536 |
% \DescribeMacro{\delcasesize} |
| 537 |
% is similar to "\casesize", except that \meta{other} is |
| 538 |
% executed if \meta{size} is different from "auto" |
| 539 |
% or from an empty list of tokens. |
| 540 |
% |
| 541 |
% |
| 542 |
% \section{Examples} |
| 543 |
% |
| 544 |
% \subsection{Macros with arguments} |
| 545 |
% |
| 546 |
% It is possible to define a macro \meta{name} |
| 547 |
% such that "\m"\marg{name} accepts some arguments. |
| 548 |
% For that, it is sufficient |
| 549 |
% that it expands into a \TeX{} command |
| 550 |
% that accepts arguments. |
| 551 |
% |
| 552 |
% \newcommand{\evolution}[2]{\m{#1}(0,T;\m{#2})} |
| 553 |
% \Macro{L20T}{\evolution{L2}} |
| 554 |
% \begin{examplenotab} |
| 555 |
% "\newcommand{\evolution}[2]{\m{#1}(0,T;\m{#2})}"\\ |
| 556 |
% "\Macro{L20T}{\evolution{L2}}"\\ |
| 557 |
% "\m{L20T}{H10Om}\cap\m{L20T}{L2Om}"\quad |
| 558 |
% $\m{L20T}{H10Om}\cap\m{L20T}{L2Om}$ |
| 559 |
% \end{examplenotab} |
| 560 |
% |
| 561 |
% |
| 562 |
% \subsection{Matrix norms} |
| 563 |
%\newnorm{matrix}{\delcasesize{#1}% |
| 564 |
%{\left|\left|\left| #2\right|\right|\right|}% |
| 565 |
%{|||#2|||}% |
| 566 |
%{#1|#1|#1|#2 #1|#1|#1|}} |
| 567 |
% |
| 568 |
% A matrix norm with three bars may be |
| 569 |
% defined by |
| 570 |
%\begin{verbatim} |
| 571 |
%\newnorm{matrix}{\delcasesize{#1}% |
| 572 |
%{\left|\left|\left| #2\right|\right|\right|}% |
| 573 |
%{|||#2|||}% |
| 574 |
%{#1|#1|#1|#2 #1|#1|#1|}} |
| 575 |
%\end{verbatim} |
| 576 |
% so that |
| 577 |
% \begin{examplenotab} |
| 578 |
% "\norm{matrix}{A^2}\leq \norm{matrix}{A}^2"\quad |
| 579 |
% $\norm{matrix}{A^2}\leq \norm{matrix}{A}^2$\\ |
| 580 |
% "\norm[]{matrix}{A^2}\leq \norm{matrix}{A}^2"\quad |
| 581 |
% $\norm[]{matrix}{A^2}\leq \norm{matrix}{A}^2$ |
| 582 |
% \end{examplenotab} |
| 583 |
% |
| 584 |
% \subsection{Sets} |
| 585 |
% |
| 586 |
% An example of code for defining sets was given |
| 587 |
% with the documentation of "\deltwoarg". Here is |
| 588 |
% another possibility |
| 589 |
% \newcommand{\settwo}[3][auto]{\deltwoarg[#1]{#2\,}{\,#3}{\{}{\}}{|}} |
| 590 |
% \begin{examplenotab} |
| 591 |
% "\newcommand{\set}[3][auto]{%"\\ |
| 592 |
% "\deltwoarg[#1]{#2\,}{\,#3}{\{}{\}}{|}}"\\ |
| 593 |
% "\set{x>-1}{\frac{x}{1+x}\geq \frac{1}{2}}"\quad |
| 594 |
% $\displaystyle \settwo{x>-1}{\frac{x}{1+x}\geq \frac{1}{2}}$ |
| 595 |
% \end{examplenotab} |
| 596 |
% |
| 597 |
% \subsection{Duality products} |
| 598 |
% |
| 599 |
% The command "\casesize" may be used for defining |
| 600 |
% a duality product, which is a bit tricky since |
| 601 |
% we want the position of name of the dual space |
| 602 |
% to be adjusted to the size of the delimitors. |
| 603 |
% The command "\dual" defined by |
| 604 |
% |
| 605 |
%\newcommand{\dual}[5][auto]{ |
| 606 |
%\casesize{#1}{ |
| 607 |
%\left.\vphantom{\left\langle #4;#5\right\rangle}% |
| 608 |
%\right._{\m{#2}}\left.% |
| 609 |
%\left\langle #4;#5\right\rangle\right._{\m{#3}}}% |
| 610 |
%{{}_{\m{#2}}\langle #4;#5\rangle_{\m{#3}}}% |
| 611 |
%{\bigl._{\m{#2}}\bigl\langle #4;#5\bigr\rangle_{\m{#3}}}% |
| 612 |
%{\Bigl._{\m{#2}}\Bigl\langle #4;#5\Bigr\rangle_{\m{#3}}}% |
| 613 |
%{\biggl._{\m{#2}}\biggl\langle #4;#5\biggr\rangle_{\m{#3}}}% |
| 614 |
%{\Biggl._{\m{#2}}\Biggl\langle #4;#5\Biggr\rangle_{\m{#3}}}} |
| 615 |
% \begin{verbatim} |
| 616 |
%\newcommand{\dual}[5][auto]{ |
| 617 |
%\casesize{#1}{ |
| 618 |
%\left.\vphantom{\left\langle #4;#5\right\rangle}% |
| 619 |
%\right._{\m{#2}}\left.% |
| 620 |
%\left\langle #4;#5\right\rangle\right._{\m{#3}}}% |
| 621 |
%{{}_{\m{#2}}\langle #4;#5\rangle_{\m{#3}}}% |
| 622 |
%{\bigl._{\m{#2}}\bigl\langle #4;#5\bigr\rangle_{\m{#3}}}% |
| 623 |
%{\Bigl._{\m{#2}}\Bigl\langle #4;#5\Bigr\rangle_{\m{#3}}}% |
| 624 |
%{\biggl._{\m{#2}}\biggl\langle #4;#5\biggr\rangle_{\m{#3}}}% |
| 625 |
%{\Biggl._{\m{#2}}\Biggl\langle #4;#5\Biggr\rangle_{\m{#3}}}} |
| 626 |
% \end{verbatim} |
| 627 |
% may be used to produce |
| 628 |
% \Macro{H1}{\mathrm{H}^1} |
| 629 |
% \Macro{H-1}{\mathrm{H}^{-1}} |
| 630 |
% \begin{example} |
| 631 |
% "\Macro{H1}{\mathrm{H}^1}"\\ |
| 632 |
% "\Macro{H-1}{\mathrm{H}^{-1}}"\\ |
| 633 |
% "\dual{H-1}{H1}{f}{\frac{1}{g}}" & |
| 634 |
% $\dual{H-1}{H1}{f}{\frac{1}{g}}$\\ |
| 635 |
% "\dual[]{H-1}{H1}{f}{\frac{1}{g}}" & |
| 636 |
% $\dual[]{H-1}{H1}{f}{\frac{1}{g}}$\\ |
| 637 |
% "\dual[bigg]{H-1}{H1}{f}{\frac{1}{g}}" & |
| 638 |
% $\dual[bigg]{H-1}{H1}{f}{\frac{1}{g}}$ |
| 639 |
% \end{example} |
| 640 |
% |
| 641 |
% |
| 642 |
% |
| 643 |
% |
| 644 |
% \section{The code} |
| 645 |
% |
| 646 |
% \subsection{Package heading} |
| 647 |
% This package uses some code of the \package{amsmath} package. |
| 648 |
% All the options of \package{amsmath} may |
| 649 |
% be set as options of \package{functan} in |
| 650 |
% order to avoir an option clash. |
| 651 |
% \begin{macrocode} |
| 652 |
%<*package> |
| 653 |
\NeedsTeXFormat{LaTeX2e} |
| 654 |
\ProvidesPackage{functan}[2004/07/03 v1.0, Macros for functional analysis] |
| 655 |
\DeclareOption{leqno}{% |
| 656 |
\PassOptionsToPackage{leqno}{amsmath}} |
| 657 |
\DeclareOption{intlimits}{% |
| 658 |
\PassOptionsToPackage{intlimits}{amsmath}} |
| 659 |
\DeclareOption{nointlimits}{% |
| 660 |
\PassOptionsToPackage{leqno}{amsmath}} |
| 661 |
\DeclareOption{sumlimits}{% |
| 662 |
\PassOptionsToPackage{sumlimits}{amsmath}} |
| 663 |
\DeclareOption{nosumlimits}{% |
| 664 |
\PassOptionsToPackage{leqno}{amsmath}} |
| 665 |
\DeclareOption{namelimits}{% |
| 666 |
\PassOptionsToPackage{namelimits}{amsmath}} |
| 667 |
\DeclareOption{nonamelinits}{% |
| 668 |
\PassOptionsToPackage{nonamelinits}{amsmath}} |
| 669 |
\DeclareOption{reqno}{% |
| 670 |
\PassOptionsToPackage{reqno}{amsmath}} |
| 671 |
\DeclareOption{centertags}{% |
| 672 |
\PassOptionsToPackage{centertags}{amsmath}} |
| 673 |
\DeclareOption{tbtags}{% |
| 674 |
\PassOptionsToPackage{tbtags}{amsmath}} |
| 675 |
\DeclareOption{cmex10}{% |
| 676 |
\PassOptionsToPackage{cmex10}{amsmath}} |
| 677 |
\DeclareOption{fleqn}{% |
| 678 |
\PassOptionsToPackage{fleqn}{amsmath}} |
| 679 |
\ProcessOptions |
| 680 |
\RequirePackage{amsmath}[2000/01/01 v2.0] |
| 681 |
% \end{macrocode} |
| 682 |
% \subsection{Defining macros} |
| 683 |
% |
| 684 |
% \begin{macro}{\Macro} |
| 685 |
% This commands defines a new macro with |
| 686 |
% name \meta{name}, by creating |
| 687 |
% a \TeX{} command \verb+functan@macro@+\meta{name}. |
| 688 |
% A warning message is sent if \meta{name} has already |
| 689 |
% been defined. |
| 690 |
% \begin{macrocode} |
| 691 |
\newcommand{\Macro}[2]{\@ifundefined{functan@macro@#1}{}{% |
| 692 |
\PackageWarning{functan}{The macro '#1' has already been defined}} |
| 693 |
\@namedef{functan@macro@#1}{#2}} |
| 694 |
% \end{macrocode} |
| 695 |
% \end{macro} |
| 696 |
% \begin{macro}{\m} |
| 697 |
% \begin{macro}{\mempty} |
| 698 |
% This command allows to call the macro defined by \meta{name}. |
| 699 |
% An error message is sent if \meta{name} has not been |
| 700 |
% defined, or if no argument is given. |
| 701 |
% The macro \verb+\mempty+ provides no error |
| 702 |
% message is the argument is empty. It is used |
| 703 |
% to deal with \verb+\conv+ macros. |
| 704 |
% \begin{macrocode} |
| 705 |
\newcommand{\m}[1]{\@ifempty{#1}{% |
| 706 |
\PackageError{functan}{No macro's name given}{} |
| 707 |
}{% |
| 708 |
\@ifundefined{functan@macro@#1}{% |
| 709 |
\PackageError{functan}{The macro `#1' is not defined}{}}{% |
| 710 |
\@nameuse{functan@macro@#1}}}} |
| 711 |
\newcommand{\mempty}[1]{\@ifempty{#1}{}{% |
| 712 |
\@ifundefined{functan@macro@#1}{% |
| 713 |
\PackageError{functan}{The macro `#1' is not defined}{}}{% |
| 714 |
\@nameuse{functan@macro@#1}}}} |
| 715 |
% \end{macrocode} |
| 716 |
% \end{macro} |
| 717 |
% \end{macro} |
| 718 |
% |
| 719 |
% |
| 720 |
% \subsection{Dealing with size} |
| 721 |
% \begin{macro}{\functan@casesize} |
| 722 |
% \begin{macro}{\delcasesize} |
| 723 |
% The macro "\delcasesize" is an alias for "\functan@casesize". |
| 724 |
% These macros require four arguments. This first one |
| 725 |
% is the size among $\emptyset$ (empty argument for normal size) |
| 726 |
% \texttt{auto} and one of the delimitors possible |
| 727 |
% size (\texttt{big}, \texttt{Big}, \texttt{bigg}, \texttt{Bigg}). |
| 728 |
% It \texttt{auto} is detected, then the second argument |
| 729 |
% is executed. It no argument is given, then the third |
| 730 |
% argument is executed. Otherwise, the fourth argument |
| 731 |
% is executed. |
| 732 |
% Another macro, "\casesize" (see below), is intended |
| 733 |
% to execute a different code the the six distinct possibilities |
| 734 |
% ($\emptyset$, \texttt{auto}, |
| 735 |
% \texttt{big}, \texttt{Big}, \texttt{bigg}, \texttt{Bigg}). |
| 736 |
% \begin{macrocode} |
| 737 |
\global\def\functan@size@auto{auto} |
| 738 |
\newcommand{\functan@casesize}[4]{\def\functan@size{#1}% |
| 739 |
\ifx\functan@size\functan@size@auto #2\else% |
| 740 |
\@ifempty{#1}{#3}{#4}\fi} |
| 741 |
\let\delcasesize\functan@casesize |
| 742 |
% \end{macrocode} |
| 743 |
% \end{macro} |
| 744 |
% \end{macro} |
| 745 |
% \begin{macro}{\casesize} |
| 746 |
% The first argument is one of the delimitor's size |
| 747 |
% specification (\texttt{auto}, $\emptyset$, |
| 748 |
% \texttt{big}, \texttt{Big}, \texttt{bigg}, \texttt{Bigg}), |
| 749 |
% and the six others are the different codes |
| 750 |
% for each of the possible size in the order given just above. |
| 751 |
% \begin{macrocode} |
| 752 |
\global\def\functan@size@big{big} |
| 753 |
\global\def\functan@size@Big{Big} |
| 754 |
\global\def\functan@size@bigg{bigg} |
| 755 |
\global\def\functan@size@Bigg{Bigg} |
| 756 |
\newcommand{\casesize}[7]{% |
| 757 |
\def\functan@size{#1} |
| 758 |
\@ifempty{#1}{#3}{% |
| 759 |
\ifx\functan@size\functan@size@auto{#2}\else% |
| 760 |
\ifx\functan@size\functan@size@big{#4}\else% |
| 761 |
\ifx\functan@size\functan@size@Big{#5}\else% |
| 762 |
\ifx\functan@size\functan@size@bigg{#6}\else% |
| 763 |
\ifx\functan@size\functan@size@Bigg{#7}\else% |
| 764 |
\PackageError{functan}{Size argument `#1' of `casesize' not valid}% |
| 765 |
\fi\fi\fi\fi\fi}} |
| 766 |
% \end{macrocode} |
| 767 |
% \end{macro} |
| 768 |
% \begin{macro}{\delonearg} |
| 769 |
% \begin{macro}{\deltwoarg} |
| 770 |
% These two macros may be used for defining a pair |
| 771 |
% of delimitors with automatic or manual size adjustment. |
| 772 |
% |
| 773 |
% \Nota A "big-g-g" argument is transformed |
| 774 |
% into \TeX{} control sequence with the command |
| 775 |
% "\@nameuse". If the argument of \@nameuse |
| 776 |
% is not a control sequence for an already defined |
| 777 |
% control sequence, then "\@nameuse"\marg{...} |
| 778 |
% is transformed into an empty list. |
| 779 |
% \begin{macrocode} |
| 780 |
\newcommand{\delonearg}[4][auto]{% |
| 781 |
\functan@casesize{#1}{\left#3 #2 \right#4}{% |
| 782 |
#3 #2 #4}{\@nameuse{#1l}#3 #2 \@nameuse{#1r}#4}} |
| 783 |
\newcommand{\deltwoarg}[6][auto]{% |
| 784 |
\def\functan@vert@bar{|}% |
| 785 |
\edef\functan@arg@bar{#6}% |
| 786 |
\ifx\functan@arg@bar\functan@vert@bar% |
| 787 |
\functan@casesize{#1}{\left#4 #2\;\vrule\; #3\right#5}{% |
| 788 |
#4 #2\;\vert\;#3 #5}{% |
| 789 |
\@nameuse{#1l}#4 #2\;\@nameuse{#1}\vert\;#3 \@nameuse{#1r}#5} |
| 790 |
\else |
| 791 |
\functan@casesize{#1}{\left#4 #2 #6 #3 \right#5}{% |
| 792 |
#4 #2 #6 #3 #5}{% |
| 793 |
\@nameuse{#1l}#4 #2 #6 #3 \@nameuse{#1r}#5} |
| 794 |
\fi} |
| 795 |
% \end{macrocode} |
| 796 |
% \end{macro} |
| 797 |
% \end{macro} |
| 798 |
% \subsection{Norms} |
| 799 |
% \begin{macro}{\functan@norm} |
| 800 |
% This is the generic macro for the norms, which |
| 801 |
% uses double bars $\Vert$ as delimitors. |
| 802 |
% \begin{macrocode} |
| 803 |
\newcommand{\functan@norm}[3]{% |
| 804 |
\functan@casesize{#1}{\left\lVert #3\right\rVert% |
| 805 |
\@ifnotempty{#2}{_{#2}}}{% |
| 806 |
\lVert #3\rVert\@ifnotempty{#2}{_{#2}}}{% |
| 807 |
\@nameuse{#1l}\lVert #3\@nameuse{#1r}\rVert% |
| 808 |
\@ifnotempty{#2}{_{#2}}}} |
| 809 |
% \end{macrocode} |
| 810 |
% \end{macro} |
| 811 |
% \begin{macro}{\functan@starred@norm} |
| 812 |
% \begin{macro}{\functan@nonstarred@norm} |
| 813 |
% This two macros call the command |
| 814 |
% "\functan@norm" either with (non-starred form) |
| 815 |
% or without (starred form) expending the second |
| 816 |
% argument \meta{name} as the name of a macro defined by "\Macro". |
| 817 |
% Note that if the second argument is empty |
| 818 |
% in "\functan@nonstarred@norm", then |
| 819 |
% no error message is produced. This allows |
| 820 |
% to have a subscript just after the norm |
| 821 |
% such as in "\norm{}{f}_1" |
| 822 |
% without producing an error message |
| 823 |
% of type ``double subscript error''. Note |
| 824 |
% that the previous example may also be |
| 825 |
% written "\norm*{1}{f}". |
| 826 |
% For the non-starred version, |
| 827 |
% if a \TeX{} macro |
| 828 |
% "functan@named@norm@"\meta{name} |
| 829 |
% exists (which is defined by "\newnorm"), |
| 830 |
% then this macro is called instead |
| 831 |
% of \verb+\functan@norm+. |
| 832 |
% \begin{macrocode} |
| 833 |
\newcommand{\functan@starred@norm}[3][auto]{% |
| 834 |
\functan@norm{#1}{#2}{#3}} |
| 835 |
\newcommand{\functan@nonstarred@norm}[3][auto]{% |
| 836 |
\@ifundefined{functan@named@norm@#2}% |
| 837 |
{\@ifempty{#2}{\functan@norm{#1}{}{#3}}{% |
| 838 |
\functan@norm{#1}{\m{#2}}{#3}}}% |
| 839 |
{\@nameuse{functan@named@norm@#2}{#1}{#3}}} |
| 840 |
% \end{macrocode} |
| 841 |
% \end{macro} |
| 842 |
% \end{macro} |
| 843 |
% \begin{macro}{\newnorm} |
| 844 |
% This macro allows to override the |
| 845 |
% behavior of the "\norm" macro |
| 846 |
% by defining a \TeX{} command |
| 847 |
% "functan@named@norm@"\meta{name}. |
| 848 |
% \begin{macrocode} |
| 849 |
\newcommand{\newnorm}[1]{% |
| 850 |
\@namedef{functan@named@norm@#1}##1##2} |
| 851 |
% \end{macrocode} |
| 852 |
% \end{macro} |
| 853 |
% \begin{macro}{\norm} This macro is the |
| 854 |
%one which is finally used. The presence |
| 855 |
%of a star "*" is checked |
| 856 |
%and either "\functan@starred@norm" or |
| 857 |
%"\functan@nonstarred@norm" is called in |
| 858 |
%consequence. |
| 859 |
% \begin{macrocode} |
| 860 |
\newcommand{\norm}{\@ifstar{\functan@starred@norm}% |
| 861 |
{\functan@nonstarred@norm}} |
| 862 |
% \end{macrocode} |
| 863 |
% \end{macro} |
| 864 |
% |
| 865 |
% |
| 866 |
% \subsection{Convergence} |
| 867 |
% |
| 868 |
% \Nota It is for this set of macros that |
| 869 |
% the compatibility with version 2.0 of "amsmath" |
| 870 |
% or higher is important. |
| 871 |
% |
| 872 |
% \begin{macro}{\newconv} This macro |
| 873 |
% allows to override the behavior of the |
| 874 |
% \verb+\conv+ macro, by defining a |
| 875 |
% \TeX{} control sequence "functan@named@conv@"\meta{name}. |
| 876 |
% \begin{macrocode} |
| 877 |
\newcommand{\newconv}[1]{% |
| 878 |
\@namedef{functan@named@conv@#1}##1} |
| 879 |
% \end{macrocode} |
| 880 |
% \end{macro} |
| 881 |
% \begin{macro}{\conv} |
| 882 |
% This macros stands for the usual convergence |
| 883 |
% symbol ($\rightarrow$). The presence of a star |
| 884 |
% \verb+*+ is checked, and either |
| 885 |
% \verb+\functan@nonstarred@conv+ |
| 886 |
% or \verb+\functan@starred@conv+ is called in consequence. |
| 887 |
% If a \TeX{} command called |
| 888 |
% \verb+functan@named@conv@+\meta{name} exists, |
| 889 |
% where \meta{name} is the first argument of \verb+\conv+, |
| 890 |
% then this command is called instead of the others. |
| 891 |
% |
| 892 |
% The command \verb+\xrightarrow+ is defined |
| 893 |
% in the \package{amsmath} package. |
| 894 |
% \begin{macrocode} |
| 895 |
\newcommand{\conv}{% |
| 896 |
\@ifstar{\functan@starred@conv}{\functan@nonstarred@conv}} |
| 897 |
\newcommand{\functan@nonstarred@conv}[2]{% |
| 898 |
\@ifundefined{functan@named@conv@#1}% |
| 899 |
{\xrightarrow[#2]{\mempty{#1}}}{\@nameuse{functan@named@conv@#1}{#2}}} |
| 900 |
\newcommand{\functan@starred@conv}[2]{\xrightarrow[#2]{#1}} |
| 901 |
% \end{macrocode} |
| 902 |
% \end{macro} |
| 903 |
% \begin{macro}{\xleftrightglobalarrow} |
| 904 |
% \begin{macro}{\Xleftrightglobalarrow} |
| 905 |
% \begin{macro}{\Xrightglobalarrow} |
| 906 |
% \begin{macro}{\Xleftglobalarrow} |
| 907 |
% These commands are defined in a way similar to the |
| 908 |
% one in \verb+\xrightarrow+ in the package \package{amsmath}. |
| 909 |
% The first four arguments are lengths added on each |
| 910 |
% side of the argument. |
| 911 |
% \begin{macrocode} |
| 912 |
\newcommand{\xleftrightarrow}[2][]{% |
| 913 |
\ext@arrow 9999\leftrightarrowfill@{#1}{#2}} |
| 914 |
\newcommand{\Xrightarrow}[2][]{% |
| 915 |
\ext@arrow 0359\Rightarrowfill@{#1}{#2}} |
| 916 |
\newcommand{\Xleftarrow}[2][]{% |
| 917 |
\ext@arrow 3095\Leftarrowfill@{#1}{#2}} |
| 918 |
\newcommand{\Xleftrightarrow}[2][]{% |
| 919 |
\ext@arrow 9999\Leftrightarrowfill@{#1}{#2}} |
| 920 |
% \end{macrocode} |
| 921 |
% \end{macro} |
| 922 |
% \end{macro} |
| 923 |
% \end{macro} |
| 924 |
% \end{macro} |
| 925 |
% \begin{macro}{\leftconv} |
| 926 |
% \begin{macro}{\leftrightconv} |
| 927 |
% The commands \verb+\leftconv+ and \verb+\leftrightconv+ are defined |
| 928 |
% as the command \verb+\conv+, except that it |
| 929 |
% is not possible to override them (however, |
| 930 |
% any kind of arrow may be used with the command |
| 931 |
% \verb+\newconv+). |
| 932 |
% \begin{macrocode} |
| 933 |
\newcommand{\leftconv}{% |
| 934 |
\@ifstar{\functan@starred@leftconv}% |
| 935 |
{\functan@nonstarred@leftconv}} |
| 936 |
\newcommand{\functan@nonstarred@leftconv}[2]{% |
| 937 |
\xleftarrow[#2]{\mempty{#1}}} |
| 938 |
\newcommand{\functan@starred@leftconv}[2]{% |
| 939 |
\xleftarrow[#2]{#1}} |
| 940 |
\newcommand{\leftrightconv}{% |
| 941 |
\@ifstar{\functan@starred@leftrightconv}% |
| 942 |
{\functan@nonstarred@leftrightconv}} |
| 943 |
\newcommand{\functan@nonstarred@leftrightconv}[2]{% |
| 944 |
\xleftrightarrow[#2]{\mempty{#1}}} |
| 945 |
\newcommand{\functan@starred@leftrightconv}[2]{% |
| 946 |
\xleftrightarrow[#2]{#1}} |
| 947 |
% \end{macrocode} |
| 948 |
% \end{macro} |
| 949 |
% \end{macro} |
| 950 |
% \begin{macro}{\Conv} |
| 951 |
% \begin{macro}{\Leftconv} |
| 952 |
% \begin{macro}{\Leftrightconv} |
| 953 |
% These macros are defined like \verb+\conv+. |
| 954 |
% \begin{macrocode} |
| 955 |
\newcommand{\Conv}{% |
| 956 |
\@ifstar{\functan@starred@Conv}{\functan@nonstarred@Conv}} |
| 957 |
\newcommand{\functan@nonstarred@Conv}[2]{% |
| 958 |
\Xrightarrow[#2]{\mempty{#1}}} |
| 959 |
\newcommand{\functan@starred@Conv}[2]{% |
| 960 |
\Xrightarrow[#2]{#1}} |
| 961 |
\newcommand{\Leftconv}{% |
| 962 |
\@ifstar{\functan@starred@Leftconv}% |
| 963 |
{\functan@nonstarred@Leftconv}} |
| 964 |
\newcommand{\functan@nonstarred@Leftconv}[2]{% |
| 965 |
\Xleftarrow[#2]{\mempty{#1}}} |
| 966 |
\newcommand{\functan@starred@Leftconv}[2]{% |
| 967 |
\Xleftarrow[#2]{#1}} |
| 968 |
\newcommand{\Leftrightconv}{% |
| 969 |
\@ifstar{\functan@starred@Leftrightconv}% |
| 970 |
{\functan@nonstarred@Leftrightconv}} |
| 971 |
\newcommand{\functan@nonstarred@Leftrightconv}[2]{% |
| 972 |
\Xleftrightarrow[#2]{\mempty{#1}}} |
| 973 |
\newcommand{\functan@starred@Leftrightconv}[2]{% |
| 974 |
\Xleftrightarrow[#2]{#1}} |
| 975 |
% \end{macrocode} |
| 976 |
% \end{macro} |
| 977 |
% \end{macro} |
| 978 |
% \end{macro} |
| 979 |
% \begin{macro}{\xrightharpoon} |
| 980 |
% \begin{macro}{\xleftharpoon} |
| 981 |
% \begin{macro}{\xleftrightharpoon} |
| 982 |
% This type of arrow behaves as the other ones, except |
| 983 |
% that an $\rightharpoonup$ is used at the end |
| 984 |
% of the arrow. |
| 985 |
% \begin{macrocode} |
| 986 |
\def\rightharpoonfill@{\arrowfill@\relbar\relbar\rightharpoonup} |
| 987 |
\newcommand{\xrightharpoon}[2][]% |
| 988 |
{\ext@arrow 0359\rightharpoonfill@{#1}{#2}} |
| 989 |
\def\leftharpoonfill@{\arrowfill@\leftharpoonup\relbar\relbar} |
| 990 |
\newcommand{\xleftharpoon}[2][]% |
| 991 |
{\ext@arrow 3095\leftharpoonfill@{#1}{#2}} |
| 992 |
\def\leftrightharpoonfill@% |
| 993 |
{\arrowfill@\leftharpoonup\relbar\rightharpoonup} |
| 994 |
\newcommand{\xleftrightharpoon}[2][]% |
| 995 |
{\ext@arrow 3095\leftrightharpoonfill@{#1}{#2}} |
| 996 |
% \end{macrocode} |
| 997 |
% \end{macro} |
| 998 |
% \end{macro} |
| 999 |
% \end{macro} |
| 1000 |
% \begin{macro}{\wkconv} |
| 1001 |
% \begin{macro}{\leftwkconv} |
| 1002 |
% \begin{macro}{\leftrightwkconv} |
| 1003 |
% These macro may be used for the weak convergence. |
| 1004 |
% Its definition is similar to the one of \verb+\conv+. |
| 1005 |
% \begin{macrocode} |
| 1006 |
\newcommand{\wkconv}{% |
| 1007 |
\@ifstar{\functan@starred@wkconv}% |
| 1008 |
{\functan@nonstarred@wkconv}} |
| 1009 |
\newcommand{\functan@nonstarred@wkconv}[2]{% |
| 1010 |
\xrightharpoon[#2]{\mempty{#1}}} |
| 1011 |
\newcommand{\functan@starred@wkconv}[2]{% |
| 1012 |
\xrightharpoon[#2]{#1}} |
| 1013 |
\newcommand{\leftwkconv}{% |
| 1014 |
\@ifstar{\functan@starred@leftwkconv}% |
| 1015 |
{\functan@nonstarred@leftwkconv}} |
| 1016 |
\newcommand{\functan@nonstarred@leftwkconv}[2]{% |
| 1017 |
\xleftharpoon[#2]{\mempty{#1}}} |
| 1018 |
\newcommand{\functan@starred@leftwkconv}[2]{% |
| 1019 |
\xleftharpoon[#2]{#1}} |
| 1020 |
\newcommand{\leftrightwkconv}{% |
| 1021 |
\@ifstar{\functan@starred@leftrightwkconv}% |
| 1022 |
{\functan@nonstarred@leftrightwkconv}} |
| 1023 |
\newcommand{\functan@nonstarred@leftrightwkconv}[2]{% |
| 1024 |
\xleftrightharpoon[#2]{\mempty{#1}}} |
| 1025 |
\newcommand{\functan@starred@leftrightwkconv}[2]{% |
| 1026 |
\xleftrightharpoon[#2]{#1}} |
| 1027 |
% \end{macrocode} |
| 1028 |
% \end{macro} |
| 1029 |
% \end{macro} |
| 1030 |
% \end{macro} |
| 1031 |
% |
| 1032 |
% \subsection{Scalar products} |
| 1033 |
% |
| 1034 |
% \begin{macro}{\scalprod} |
| 1035 |
% \begin{macro}{\newscalprod} |
| 1036 |
% The definition of the scalar product is |
| 1037 |
% similar to the definition of norms. |
| 1038 |
% \begin{macrocode} |
| 1039 |
\newcommand{\scalprod}{\@ifstar{\functan@starred@scalprod}% |
| 1040 |
{\functan@nonstarred@scalprod}} |
| 1041 |
\newcommand{\functan@starred@scalprod}[4][auto]{ |
| 1042 |
\deltwoarg[#1]{#3}{#4}{\langle}% |
| 1043 |
{\rangle}{,}\@ifempty{#2}{}{_{#2}}} |
| 1044 |
\newcommand{\functan@nonstarred@scalprod}[4][auto]{ |
| 1045 |
\@ifundefined{functan@scalprod@named@#2}{ |
| 1046 |
\deltwoarg[#1]{#3}{#4}{\langle}% |
| 1047 |
{\rangle}{,}\@ifempty{#2}{}{_{\m{#2}}}}{% |
| 1048 |
\@nameuse{functan@scalprod@named@#2}{#1}{#3}{#4}}} |
| 1049 |
\newcommand{\newscalprod}[1]{% |
| 1050 |
\@namedef{functan@scalprod@named@#1}##1##2##3} |
| 1051 |
%</package> |
| 1052 |
% \end{macrocode} |
| 1053 |
% \end{macro} |
| 1054 |
% \end{macro} |
| 1055 |
% \Finale |
| 1056 |
% \setcounter{IndexColumns}{2} |
| 1057 |
% \PrintIndex |