| 1 |
% Copyright 2018 by Till Tantau |
| 2 |
% |
| 3 |
% This file may be distributed and/or modified |
| 4 |
% |
| 5 |
% 1. under the LaTeX Project Public License and/or |
| 6 |
% 2. under the GNU Free Documentation License. |
| 7 |
% |
| 8 |
% See the file doc/generic/pgf/licenses/LICENSE for more details. |
| 9 |
|
| 10 |
|
| 11 |
\section{Transformations} |
| 12 |
|
| 13 |
\pgfname\ has a powerful transformation mechanism that is similar to the |
| 14 |
transformation capabilities of \textsc{metafont}. The present section explains |
| 15 |
how you can access it in \tikzname. |
| 16 |
|
| 17 |
|
| 18 |
\subsection{The Different Coordinate Systems} |
| 19 |
|
| 20 |
It is a long process from a coordinate like, say, $(1,2)$ or |
| 21 |
$(1\mathrm{cm},5\mathrm{pt})$, to the position a point is finally placed on the |
| 22 |
display or paper. In order to find out where the point should go, it is |
| 23 |
constantly ``transformed'', which means that it is mostly shifted around and |
| 24 |
possibly rotated, slanted, scaled, and otherwise mutilated. |
| 25 |
|
| 26 |
In detail, (at least) the following transformations are applied to a coordinate |
| 27 |
like $(1,2)$ before a point on the screen is chosen: |
| 28 |
% |
| 29 |
\begin{enumerate} |
| 30 |
\item \pgfname\ interprets a coordinate like $(1,2)$ in its |
| 31 |
$xy$-coordinate system as ``add the current $x$-vector once and the |
| 32 |
current $y$-vector twice to obtain the new point''. |
| 33 |
\item \pgfname\ applies its coordinate transformation matrix to the |
| 34 |
resulting coordinate. This yields the final position of the point |
| 35 |
inside the picture. |
| 36 |
\item The backend driver (like |dvips| or |pdftex|) adds transformation |
| 37 |
commands such that the coordinate is shifted to the correct position in |
| 38 |
\TeX's page coordinate system. |
| 39 |
\item \textsc{pdf} (or PostScript) apply the canvas transformation matrix |
| 40 |
to the point, which can once more change the position on the page. |
| 41 |
\item The viewer application or the printer applies the device |
| 42 |
transformation matrix to transform the coordinate to its final pixel |
| 43 |
coordinate on the screen or paper. |
| 44 |
\end{enumerate} |
| 45 |
|
| 46 |
In reality, the process is even more involved, but the above should give the |
| 47 |
idea: A point is constantly transformed by changes of the coordinate system. |
| 48 |
|
| 49 |
In \tikzname, you only have access to the first two coordinate systems: The |
| 50 |
$xy$-coordinate system and the coordinate transformation matrix (these will be |
| 51 |
explained later). \pgfname\ also allows you to change the canvas transformation |
| 52 |
matrix, but you have to use commands of the core layer directly to do so and |
| 53 |
you ``better know what you are doing'' when you do this. The moment you start |
| 54 |
modifying the canvas matrix, \pgfname\ immediately loses track of all |
| 55 |
coordinates and shapes, anchors, and bounding box computations will no longer |
| 56 |
work. |
| 57 |
|
| 58 |
|
| 59 |
\subsection{The XY- and XYZ-Coordinate Systems} |
| 60 |
\label{section-xyz} |
| 61 |
|
| 62 |
The first and easiest coordinate systems are \pgfname's $xy$- and |
| 63 |
$xyz$-coordinate systems. The idea is very simple: Whenever you specify a |
| 64 |
coordinate like |(2,3)| this means $2v_x + 3v_y$, where $v_x$ is the current |
| 65 |
\emph{$x$-vector} and $v_y$ is the current \emph{$y$-vector}. Similarly, the |
| 66 |
coordinate |(1,2,3)| means $v_x + 2v_y + 3v_z$. |
| 67 |
|
| 68 |
Unlike other packages, \pgfname\ does not insist that $v_x$ actually has a |
| 69 |
$y$-component of $0$, that is, that it is a horizontal vector. Instead, the |
| 70 |
$x$-vector can point anywhere you want. Naturally, \emph{normally} you will |
| 71 |
want the $x$-vector to point horizontally. |
| 72 |
|
| 73 |
One undesirable effect of this flexibility is that it is not possible to |
| 74 |
provide mixed coordinates as in $(1,2\mathrm{pt})$. Life is hard. |
| 75 |
|
| 76 |
To change the $x$-, $y$-, and $z$-vectors, you can use the following options: |
| 77 |
|
| 78 |
\begin{key}{/tikz/x=\meta{value} (initially 1cm)} |
| 79 |
If \meta{value} is a dimension, the $x$-vector of \pgfname's |
| 80 |
$xyz$-coordinate system is set up to point \meta{value} to the right, that |
| 81 |
is, to $(\meta{value},0pt)$. |
| 82 |
% |
| 83 |
\begin{codeexample}[] |
| 84 |
\begin{tikzpicture} |
| 85 |
\draw (0,0) -- +(1,0); |
| 86 |
\draw[x=2cm,color=red] (0,0.1) -- +(1,0); |
| 87 |
\end{tikzpicture} |
| 88 |
\end{codeexample} |
| 89 |
|
| 90 |
\begin{codeexample}[] |
| 91 |
\tikz \draw[x=1.5cm] (0,0) grid (2,2); |
| 92 |
\end{codeexample} |
| 93 |
|
| 94 |
The last example shows that the size of steppings in grids, just like all |
| 95 |
other dimensions, are not affected by the $x$-vector. After all, the |
| 96 |
$x$-vector is only used to determine the coordinate of the upper right |
| 97 |
corner of the grid. |
| 98 |
|
| 99 |
If \meta{value} is a coordinate, the $x$-vector of \pgfname's |
| 100 |
$xyz$-coordinate system is set to the specified coordinate. If \meta{value} |
| 101 |
contains a comma, it must be put in braces. |
| 102 |
% |
| 103 |
\begin{codeexample}[] |
| 104 |
\begin{tikzpicture} |
| 105 |
\draw (0,0) -- (1,0); |
| 106 |
\draw[x={(2cm,0.5cm)},color=red] (0,0) -- (1,0); |
| 107 |
\end{tikzpicture} |
| 108 |
\end{codeexample} |
| 109 |
|
| 110 |
You can use this, for example, to exchange the meaning of the $x$- and |
| 111 |
$y$-coordinate. |
| 112 |
% |
| 113 |
\begin{codeexample}[] |
| 114 |
\begin{tikzpicture}[smooth] |
| 115 |
\draw plot coordinates{(1,0) (2,0.5) (3,0) (3,1)}; |
| 116 |
\draw[x={(0cm,1cm)},y={(1cm,0cm)},color=red] |
| 117 |
plot coordinates{(1,0) (2,0.5) (3,0) (3,1)}; |
| 118 |
\end{tikzpicture} |
| 119 |
\end{codeexample} |
| 120 |
% |
| 121 |
\end{key} |
| 122 |
|
| 123 |
\begin{key}{/tikz/y=\meta{value} (initially 1cm)} |
| 124 |
Works like the |x=| option, only if \meta{value} is a dimension, the |
| 125 |
resulting vector points to $(0,\meta{value})$. |
| 126 |
\end{key} |
| 127 |
|
| 128 |
\begin{key}{/tikz/z=\meta{value} (initially \normalfont$-3.85$mm)} |
| 129 |
Works like the |y=| option, but now a dimension is the point |
| 130 |
$(\meta{value},\meta{value})$. |
| 131 |
% |
| 132 |
\begin{codeexample}[] |
| 133 |
\begin{tikzpicture}[z=-1cm,->,thick] |
| 134 |
\draw[color=red] (0,0,0) -- (1,0,0); |
| 135 |
\draw[color=blue] (0,0,0) -- (0,1,0); |
| 136 |
\draw[color=orange] (0,0,0) -- (0,0,1); |
| 137 |
\end{tikzpicture} |
| 138 |
\end{codeexample} |
| 139 |
% |
| 140 |
\end{key} |
| 141 |
|
| 142 |
|
| 143 |
\subsection{Coordinate Transformations} |
| 144 |
|
| 145 |
\pgfname\ and \tikzname\ allow you to specify \emph{coordinate |
| 146 |
transformations}. Whenever you specify a coordinate as in |(1,0)| or |
| 147 |
|(1cm,1pt)| or |(30:2cm)|, this coordinate is first ``reduced'' to a position |
| 148 |
of the form ``$x$ points to the right and $y$ points upwards''. For example, |
| 149 |
|(1in,5pt)| is reduced to ``$72\frac{27}{100}$ points to the right and 5 points |
| 150 |
upwards'' and |(90:100pt)| means ``0pt to the right and 100 points upwards''. |
| 151 |
|
| 152 |
The next step is to apply the current \emph{coordinate transformation matrix} |
| 153 |
to the coordinate. For example, the coordinate transformation matrix might |
| 154 |
currently be set such that it adds a certain constant to the $x$ value. Also, |
| 155 |
it might be set up such that it, say, exchanges the $x$ and $y$ value. In |
| 156 |
general, any ``standard'' transformation like translation, rotation, slanting, |
| 157 |
or scaling or any combination thereof is possible. (Internally, \pgfname\ keeps |
| 158 |
track of a coordinate transformation matrix very much like the concatenation |
| 159 |
matrix used by \textsc{pdf} or PostScript.) |
| 160 |
% |
| 161 |
\begin{codeexample}[] |
| 162 |
\begin{tikzpicture} |
| 163 |
\draw[help lines] (0,0) grid (3,2); |
| 164 |
\draw (0,0) rectangle (1,0.5); |
| 165 |
\begin{scope}[xshift=1cm] |
| 166 |
\draw [red] (0,0) rectangle (1,0.5); |
| 167 |
\draw[yshift=1cm] [blue] (0,0) rectangle (1,0.5); |
| 168 |
\draw[rotate=30] [orange] (0,0) rectangle (1,0.5); |
| 169 |
\end{scope} |
| 170 |
\end{tikzpicture} |
| 171 |
\end{codeexample} |
| 172 |
|
| 173 |
The most important aspect of the coordinate transformation matrix is \emph{that |
| 174 |
it applies to coordinates only!} In particular, the coordinate transformation |
| 175 |
has no effect on things like the line width or the dash pattern or the shading |
| 176 |
angle. In certain cases, it is not immediately clear whether the coordinate |
| 177 |
transformation matrix \emph{should} apply to a certain dimension. For example, |
| 178 |
should the coordinate transformation matrix apply to grids? (It does.) And what |
| 179 |
about the size of arced corners? (It does not.) The general rule is: ``If there |
| 180 |
is no `coordinate' involved, even `indirectly', the matrix is not applied.''. |
| 181 |
However, sometimes, you simply have to try or look it up in the documentation |
| 182 |
whether the matrix will be applied. |
| 183 |
|
| 184 |
Setting the matrix cannot be done directly. Rather, all you can do is to |
| 185 |
``add'' another transformation to the current matrix. However, all |
| 186 |
transformations are local to the current \TeX-group. All transformations are |
| 187 |
added using graphic options, which are described below. |
| 188 |
|
| 189 |
Transformations apply immediately when they are encountered ``in the middle of |
| 190 |
a path'' and they apply only to the coordinates on the path following the |
| 191 |
transformation option. |
| 192 |
% |
| 193 |
\begin{codeexample}[] |
| 194 |
\tikz \draw (0,0) rectangle (1,0.5) [xshift=2cm] (0,0) rectangle (1,0.5); |
| 195 |
\end{codeexample} |
| 196 |
|
| 197 |
A final word of warning: You should refrain from using ``aggressive'' |
| 198 |
transformations like a scaling of a factor of 10\,000. The reason is that all |
| 199 |
transformations are done using \TeX, which has a fairly low accuracy. |
| 200 |
Furthermore, in certain situations it is necessary that \tikzname\ |
| 201 |
\emph{inverts} the current transformation matrix and this will fail if the |
| 202 |
transformation matrix is badly conditioned or even singular (if you do not know |
| 203 |
what singular matrices are, you are blessed). |
| 204 |
|
| 205 |
\begin{key}{/tikz/shift={\ttfamily\char`\{}\meta{coordinate}{\ttfamily\char`\}}} |
| 206 |
Adds the \meta{coordinate} to all coordinates. |
| 207 |
% |
| 208 |
\begin{codeexample}[] |
| 209 |
\begin{tikzpicture} |
| 210 |
\draw[help lines] (0,0) grid (3,2); |
| 211 |
\draw (0,0) -- (1,1) -- (1,0); |
| 212 |
\draw[shift={(1,1)},blue] (0,0) -- (1,1) -- (1,0); |
| 213 |
\draw[shift={(30:1cm)},red] (0,0) -- (1,1) -- (1,0); |
| 214 |
\end{tikzpicture} |
| 215 |
\end{codeexample} |
| 216 |
% |
| 217 |
\end{key} |
| 218 |
|
| 219 |
\begin{key}{/tikz/shift only} |
| 220 |
This option does not take any parameter. Its effect is to cancel all |
| 221 |
current transformations except for the shifting. This means that the origin |
| 222 |
will remain where it is, but any rotation around the origin or scaling |
| 223 |
relative to the origin or skewing will no longer have an effect. |
| 224 |
|
| 225 |
This option is useful in situations where a complicated transformation is |
| 226 |
used to ``get to a position'', but you then wish to draw something |
| 227 |
``normal'' at this position. |
| 228 |
% |
| 229 |
\begin{codeexample}[] |
| 230 |
\begin{tikzpicture} |
| 231 |
\draw[help lines] (0,0) grid (3,2); |
| 232 |
\draw (0,0) -- (1,1) -- (1,0); |
| 233 |
\draw[rotate=30,xshift=2cm,blue] (0,0) -- (1,1) -- (1,0); |
| 234 |
\draw[rotate=30,xshift=2cm,shift only,red] (0,0) -- (1,1) -- (1,0); |
| 235 |
\end{tikzpicture} |
| 236 |
\end{codeexample} |
| 237 |
% |
| 238 |
\end{key} |
| 239 |
|
| 240 |
\begin{key}{/tikz/xshift=\meta{dimension}} |
| 241 |
Adds \meta{dimension} to the $x$ value of all coordinates. |
| 242 |
% |
| 243 |
\begin{codeexample}[] |
| 244 |
\begin{tikzpicture} |
| 245 |
\draw[help lines] (0,0) grid (3,2); |
| 246 |
\draw (0,0) -- (1,1) -- (1,0); |
| 247 |
\draw[xshift=2cm,blue] (0,0) -- (1,1) -- (1,0); |
| 248 |
\draw[xshift=-10pt,red] (0,0) -- (1,1) -- (1,0); |
| 249 |
\end{tikzpicture} |
| 250 |
\end{codeexample} |
| 251 |
% |
| 252 |
\end{key} |
| 253 |
|
| 254 |
\begin{key}{/tikz/yshift=\meta{dimension}} |
| 255 |
Adds \meta{dimension} to the $y$ value of all coordinates. |
| 256 |
\end{key} |
| 257 |
|
| 258 |
\begin{key}{/tikz/scale=\meta{factor}} |
| 259 |
Multiplies all coordinates by the given \meta{factor}. The \meta{factor} |
| 260 |
should not be excessively large in absolute terms or very close to zero. |
| 261 |
% |
| 262 |
\begin{codeexample}[] |
| 263 |
\begin{tikzpicture} |
| 264 |
\draw[help lines] (0,0) grid (3,2); |
| 265 |
\draw (0,0) -- (1,1) -- (1,0); |
| 266 |
\draw[scale=2,blue] (0,0) -- (1,1) -- (1,0); |
| 267 |
\draw[scale=-1,red] (0,0) -- (1,1) -- (1,0); |
| 268 |
\end{tikzpicture} |
| 269 |
\end{codeexample} |
| 270 |
% |
| 271 |
\end{key} |
| 272 |
|
| 273 |
\begin{key}{/tikz/scale around={\ttfamily\char`\{}\meta{factor}|:|\meta{coordinate}{\ttfamily\char`\}}} |
| 274 |
Scales the coordinate system by \meta{factor}, with the ``origin of |
| 275 |
scaling'' centered on \meta{coordinate} rather than the origin. |
| 276 |
% |
| 277 |
\begin{codeexample}[] |
| 278 |
\begin{tikzpicture} |
| 279 |
\draw[help lines] (0,0) grid (3,2); |
| 280 |
\draw (0,0) -- (1,1) -- (1,0); |
| 281 |
\draw[scale=2,blue] (0,0) -- (1,1) -- (1,0); |
| 282 |
\draw[scale around={2:(1,1)},red] (0,0) -- (1,1) -- (1,0); |
| 283 |
\end{tikzpicture} |
| 284 |
\end{codeexample} |
| 285 |
% |
| 286 |
\end{key} |
| 287 |
|
| 288 |
\begin{key}{/tikz/xscale=\meta{factor}} |
| 289 |
Multiplies only the $x$-value of all coordinates by the given |
| 290 |
\meta{factor}. |
| 291 |
\begin{codeexample}[] |
| 292 |
\begin{tikzpicture} |
| 293 |
\draw[help lines] (0,0) grid (3,2); |
| 294 |
\draw (0,0) -- (1,1) -- (1,0); |
| 295 |
\draw[xscale=2,blue] (0,0) -- (1,1) -- (1,0); |
| 296 |
\draw[xscale=-1,red] (0,0) -- (1,1) -- (1,0); |
| 297 |
\end{tikzpicture} |
| 298 |
\end{codeexample} |
| 299 |
% |
| 300 |
\end{key} |
| 301 |
|
| 302 |
\begin{key}{/tikz/yscale=\meta{factor}} |
| 303 |
Multiplies only the $y$-value of all coordinates by \meta{factor}. |
| 304 |
\end{key} |
| 305 |
|
| 306 |
\begin{key}{/tikz/xslant=\meta{factor}} |
| 307 |
Slants the coordinate horizontally by the given \meta{factor}: |
| 308 |
% |
| 309 |
\begin{codeexample}[] |
| 310 |
\begin{tikzpicture} |
| 311 |
\draw[help lines] (0,0) grid (3,2); |
| 312 |
\draw (0,0) -- (1,1) -- (1,0); |
| 313 |
\draw[xslant=2,blue] (0,0) -- (1,1) -- (1,0); |
| 314 |
\draw[xslant=-1,red] (0,0) -- (1,1) -- (1,0); |
| 315 |
\end{tikzpicture} |
| 316 |
\end{codeexample} |
| 317 |
% |
| 318 |
\end{key} |
| 319 |
|
| 320 |
\begin{key}{/tikz/yslant=\meta{factor}} |
| 321 |
Slants the coordinate vertically by the given \meta{factor}: |
| 322 |
% |
| 323 |
\begin{codeexample}[] |
| 324 |
\begin{tikzpicture} |
| 325 |
\draw[help lines] (0,0) grid (3,2); |
| 326 |
\draw (0,0) -- (1,1) -- (1,0); |
| 327 |
\draw[yslant=2,blue] (0,0) -- (1,1) -- (1,0); |
| 328 |
\draw[yslant=-1,red] (0,0) -- (1,1) -- (1,0); |
| 329 |
\end{tikzpicture} |
| 330 |
\end{codeexample} |
| 331 |
% |
| 332 |
\end{key} |
| 333 |
|
| 334 |
\begin{key}{/tikz/rotate=\meta{degree}} |
| 335 |
Rotates the coordinate system by \meta{degree}: |
| 336 |
% |
| 337 |
\begin{codeexample}[] |
| 338 |
\begin{tikzpicture} |
| 339 |
\draw[help lines] (0,0) grid (3,2); |
| 340 |
\draw (0,0) -- (1,1) -- (1,0); |
| 341 |
\draw[rotate=40,blue] (0,0) -- (1,1) -- (1,0); |
| 342 |
\draw[rotate=-20,red] (0,0) -- (1,1) -- (1,0); |
| 343 |
\end{tikzpicture} |
| 344 |
\end{codeexample} |
| 345 |
% |
| 346 |
\end{key} |
| 347 |
|
| 348 |
\begin{key}{/tikz/rotate around={\ttfamily\char`\{}\meta{degree}|:|\meta{coordinate}{\ttfamily\char`\}}} |
| 349 |
Rotates the coordinate system by \meta{degree} around the point |
| 350 |
\meta{coordinate}. |
| 351 |
% |
| 352 |
\begin{codeexample}[] |
| 353 |
\begin{tikzpicture} |
| 354 |
\draw[help lines] (0,0) grid (3,2); |
| 355 |
\draw (0,0) -- (1,1) -- (1,0); |
| 356 |
\draw[rotate around={40:(1,1)},blue] (0,0) -- (1,1) -- (1,0); |
| 357 |
\draw[rotate around={-20:(1,1)},red] (0,0) -- (1,1) -- (1,0); |
| 358 |
\end{tikzpicture} |
| 359 |
\end{codeexample} |
| 360 |
% |
| 361 |
\end{key} |
| 362 |
|
| 363 |
\begin{key}{/tikz/rotate around x=\meta{angle}} |
| 364 |
This key sets the $x$, $y$ and $z$ vectors of the \pgfname\ |
| 365 |
$xyz$-coordinate system so that they are rotated by \meta{angle} around the |
| 366 |
axis corresponding to the $x$-vector. The rotation is applied so that when |
| 367 |
looking towards the origin along this axis, positive angles result in an |
| 368 |
anticlockwise rotation. |
| 369 |
% |
| 370 |
\begin{codeexample}[] |
| 371 |
\begin{tikzpicture}[>=stealth] |
| 372 |
\draw [->] (0,0,0) -- (2,0,0) node [at end, right] {$x$}; |
| 373 |
\draw [->] (0,0,0) -- (0,2,0) node [at end, left] {$y$}; |
| 374 |
\draw [->] (0,0,0) -- (0,0,2) node [at end, left] {$z$}; |
| 375 |
|
| 376 |
\draw [red, rotate around x=0] (0,0,0) -- (1,1,0) -- (1,0,0); |
| 377 |
\draw [green, rotate around x=45] (0,0,0) -- (1,1,0) -- (1,0,0); |
| 378 |
\draw [blue, rotate around x=90] (0,0,0) -- (1,1,0) -- (1,0,0); |
| 379 |
\end{tikzpicture} |
| 380 |
\end{codeexample} |
| 381 |
% |
| 382 |
\end{key} |
| 383 |
|
| 384 |
\begin{key}{/tikz/rotate around y=\meta{angle}} |
| 385 |
This key sets the $x$, $y$ and $z$ vectors of the \pgfname\ |
| 386 |
$xyz$-coordinate system so that they are rotated by \meta{angle} around |
| 387 |
the axis corresponding to the $y$-vector. The rotation is applied so that |
| 388 |
when looking towards the origin along this axis, positive angles result in |
| 389 |
an anticlockwise rotation. |
| 390 |
% |
| 391 |
\begin{codeexample}[] |
| 392 |
\begin{tikzpicture}[>=stealth] |
| 393 |
\draw [->] (0,0,0) -- (2,0,0) node [at end, right] {$x$}; |
| 394 |
\draw [->] (0,0,0) -- (0,2,0) node [at end, left] {$y$}; |
| 395 |
\draw [->] (0,0,0) -- (0,0,2) node [at end, left] {$z$}; |
| 396 |
|
| 397 |
\draw [red, rotate around y=0] (0,0,0) -- (1,1,0) -- (1,0,0); |
| 398 |
\draw [green, rotate around y=-45] (0,0,0) -- (1,1,0) -- (1,0,0); |
| 399 |
\draw [blue, rotate around y=-90] (0,0,0) -- (1,1,0) -- (1,0,0); |
| 400 |
\end{tikzpicture} |
| 401 |
\end{codeexample} |
| 402 |
% |
| 403 |
\end{key} |
| 404 |
|
| 405 |
\begin{key}{/tikz/rotate around z=\meta{angle}} |
| 406 |
This key sets the $x$, $y$ and $z$ vectors of the \pgfname\ |
| 407 |
$xyz$-coordinate system so that they are rotated by \meta{angle} around the |
| 408 |
axis corresponding to the $z$-vector. The rotation is applied so that when |
| 409 |
looking towards the origin along this axis, positive angles result in an |
| 410 |
anticlockwise rotation. |
| 411 |
% |
| 412 |
\begin{codeexample}[] |
| 413 |
\begin{tikzpicture}[>=stealth] |
| 414 |
\draw [->] (0,0,0) -- (2,0,0) node [at end, right] {$x$}; |
| 415 |
\draw [->] (0,0,0) -- (0,2,0) node [at end, left] {$y$}; |
| 416 |
\draw [->] (0,0,0) -- (0,0,2) node [at end, left] {$z$}; |
| 417 |
|
| 418 |
\draw [red, rotate around z=0] (0,0) -- (1,1) -- (1,0); |
| 419 |
\draw [green, rotate around z=45] (0,0) -- (1,1) -- (1,0); |
| 420 |
\draw [blue, rotate around z=90] (0,0) -- (1,1) -- (1,0); |
| 421 |
\end{tikzpicture} |
| 422 |
\end{codeexample} |
| 423 |
% |
| 424 |
\end{key} |
| 425 |
|
| 426 |
\begin{key}{/tikz/cm={\ttfamily\char`\{}\meta{$a$}|,|\meta{$b$}|,|\meta{$c$}|,|\meta{$d$}|,|\meta{coordinate}{\ttfamily\char`\}}} |
| 427 |
applies the following transformation to all coordinates: Let $(x,y)$ be the |
| 428 |
coordinate to be transformed and let \meta{coordinate} specify the point |
| 429 |
$(t_x,t_y)$. Then the new coordinate is given by |
| 430 |
$\left(\begin{smallmatrix} a & c \\ b & d\end{smallmatrix}\right) |
| 431 |
\left(\begin{smallmatrix} x \\ y \end{smallmatrix}\right) + |
| 432 |
\left(\begin{smallmatrix} t_x \\ t_y \end{smallmatrix}\right)$. |
| 433 |
Usually, you do not use this option directly. |
| 434 |
% |
| 435 |
\begin{codeexample}[] |
| 436 |
\begin{tikzpicture} |
| 437 |
\draw[help lines] (0,0) grid (3,2); |
| 438 |
\draw (0,0) -- (1,1) -- (1,0); |
| 439 |
\draw[cm={1,1,0,1,(0,0)},blue] (0,0) -- (1,1) -- (1,0); |
| 440 |
\draw[cm={0,1,1,0,(1cm,1cm)},red] (0,0) -- (1,1) -- (1,0); |
| 441 |
\end{tikzpicture} |
| 442 |
\end{codeexample} |
| 443 |
% |
| 444 |
\end{key} |
| 445 |
|
| 446 |
\begin{key}{/tikz/reset cm} |
| 447 |
Completely resets the coordinate transformation matrix to the identity |
| 448 |
matrix. This will destroy not only the transformations applied in the |
| 449 |
current scope, but also all transformations inherited from surrounding |
| 450 |
scopes. Do not use this option, unless you really, really know what you are |
| 451 |
doing. |
| 452 |
\end{key} |
| 453 |
|
| 454 |
|
| 455 |
\subsection{Canvas Transformations} |
| 456 |
|
| 457 |
A \emph{canvas transformation}, see |
| 458 |
Section~\ref{section-design-transformations} for details, is best thought of as |
| 459 |
a transformation in which the drawing canvas is stretched or rotated. Imaging |
| 460 |
writing something on a balloon (the canvas) and then blowing air into the |
| 461 |
balloon: Not only does the text become larger, the thin lines also become |
| 462 |
larger. In particular, if you scale the canvas by a factor of two, all lines |
| 463 |
are twice as thick. |
| 464 |
|
| 465 |
Canvas transformations should be used with great care. In most circumstances |
| 466 |
you do \emph{not} want line widths to change in a picture as this creates |
| 467 |
visual inconsistency. |
| 468 |
|
| 469 |
Just as important, when you use canvas transformations \emph{\pgfname\ loses |
| 470 |
track of positions of nodes and of picture sizes} since it does not take the |
| 471 |
effect of canvas transformations into account when it computes coordinates of |
| 472 |
nodes (do not, however, rely on this; it may change in the future). |
| 473 |
|
| 474 |
Finally, note that a canvas transformation always applies to a path as a whole, |
| 475 |
it is not possible (as for coordinate transformations) to use different |
| 476 |
transformations in different parts of a path. |
| 477 |
|
| 478 |
In short, you should not use canvas transformations unless you really know what |
| 479 |
you are doing. |
| 480 |
|
| 481 |
\begin{key}{/tikz/transform canvas=\meta{options}} |
| 482 |
The \meta{options} should contain coordinate transformations options like |
| 483 |
|scale| or |xshift|. Multiple options can be given, their effects |
| 484 |
accumulate in the usual manner. The effect of these \meta{options} |
| 485 |
(immediately) changes the current canvas transformation matrix. The |
| 486 |
coordinate transformation matrix is not changed. Tracking of the picture |
| 487 |
size is (locally) switched off and the node coordinate will no longer be |
| 488 |
correct. |
| 489 |
% |
| 490 |
\begin{codeexample}[] |
| 491 |
\begin{tikzpicture} |
| 492 |
\draw[help lines] (0,0) grid (3,2); |
| 493 |
\draw (0,0) -- (1,1) -- (1,0); |
| 494 |
\draw[transform canvas={scale=2},blue] (0,0) -- (1,1) -- (1,0); |
| 495 |
\draw[transform canvas={rotate=180},red] (0,0) -- (1,1) -- (1,0); |
| 496 |
\end{tikzpicture} |
| 497 |
\end{codeexample} |
| 498 |
% |
| 499 |
\end{key} |