Placement & layout¶

add_svg can drop a drawing at any data coordinate, scale it to a target size, rotate it, and stamp it repeatedly. This is what makes it useful as a source of custom markers and inline icons.

In [1]:

%matplotlib inline

%matplotlib inline

In [ ]:

import matplotlib.pyplot as plt
import numpy as np

import svg2mpl

import matplotlib.pyplot as plt import numpy as np import svg2mpl

In [3]:

STAR_SVG = """
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100">
  <polygon id="star"
    points="50,5 61,38 96,38 68,59 79,93 50,72 21,93 32,59 4,38 39,38"
    fill="gold" stroke="darkorange" stroke-width="3"/>
</svg>
"""

STAR_SVG = """ <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100"> <polygon id="star" points="50,5 61,38 96,38 68,59 79,93 50,72 21,93 32,59 4,38 39,38" fill="gold" stroke="darkorange" stroke-width="3"/> </svg> """

Placing a single drawing¶

The placement arguments work in data coordinates:

• xy -- where the anchor lands.
• width / height -- the final bounding-box size (give one to preserve aspect).
• origin -- the anchor inside the drawing, normalized ((0.5, 0.5) = center).
• rotation -- degrees.

Here we place the star centered at (0.5, 0.5) on a unit axes, scaled to a width of 0.4, and rotated by 20 degrees.

In [4]:

fig, ax = plt.subplots(figsize=(4, 4))
ax.set_xlim(0, 1)
ax.set_ylim(0, 1)
ax.set_aspect("equal")

svg2mpl.add_svg(
    STAR_SVG, xy=(0.5, 0.5), width=0.4, origin=(0.5, 0.5), rotation=20, ax=ax
)
ax.plot(0.5, 0.5, "k+", markersize=12)  # mark the anchor
ax.set_title("one star, centered on (0.5, 0.5)")
plt.show()

fig, ax = plt.subplots(figsize=(4, 4)) ax.set_xlim(0, 1) ax.set_ylim(0, 1) ax.set_aspect("equal") svg2mpl.add_svg( STAR_SVG, xy=(0.5, 0.5), width=0.4, origin=(0.5, 0.5), rotation=20, ax=ax ) ax.plot(0.5, 0.5, "k+", markersize=12) # mark the anchor ax.set_title("one star, centered on (0.5, 0.5)") plt.show()

Stamping a drawing as a marker¶

Because each SVG source is parsed once and cached, calling add_svg in a loop to place the same drawing many times is cheap. Here we scatter stars whose size and rotation vary with the data.

In [ ]:

rng = np.random.default_rng(0)
n = 12
xs = rng.uniform(0, 10, n)
ys = rng.uniform(0, 10, n)
sizes = rng.uniform(0.4, 1.2, n)

fig, ax = plt.subplots(figsize=(6, 6))
for x, y, s in zip(xs, ys, sizes, strict=True):
    svg2mpl.add_svg(
        STAR_SVG,
        xy=(x, y),
        width=s,
        rotation=rng.uniform(0, 90),
        ax=ax,
    )
ax.set_xlim(-1, 11)
ax.set_ylim(-1, 11)
ax.set_aspect("equal")
ax.set_title("stars as data-driven markers")
plt.show()

rng = np.random.default_rng(0) n = 12 xs = rng.uniform(0, 10, n) ys = rng.uniform(0, 10, n) sizes = rng.uniform(0.4, 1.2, n) fig, ax = plt.subplots(figsize=(6, 6)) for x, y, s in zip(xs, ys, sizes, strict=True): svg2mpl.add_svg( STAR_SVG, xy=(x, y), width=s, rotation=rng.uniform(0, 90), ax=ax, ) ax.set_xlim(-1, 11) ax.set_ylim(-1, 11) ax.set_aspect("equal") ax.set_title("stars as data-driven markers") plt.show()

place=False: faithful absolute layout¶

With place=False, the drawing is rendered in its own coordinate system, exactly reproducing the SVG's absolute layout (only flip_y still applies). Compare it with the placed version, which has been re-centered and scaled.

In [6]:

fig, axes = plt.subplots(1, 2, figsize=(8, 4))

svg2mpl.add_svg(STAR_SVG, place=False, ax=axes[0])
axes[0].set_title("place=False (SVG coordinates)")

svg2mpl.add_svg(STAR_SVG, xy=(0, 0), width=2, ax=axes[1])
axes[1].set_title("placed at (0, 0), width=2")

for ax in axes:
    ax.set_aspect("equal")
plt.show()

fig, axes = plt.subplots(1, 2, figsize=(8, 4)) svg2mpl.add_svg(STAR_SVG, place=False, ax=axes[0]) axes[0].set_title("place=False (SVG coordinates)") svg2mpl.add_svg(STAR_SVG, xy=(0, 0), width=2, ax=axes[1]) axes[1].set_title("placed at (0, 0), width=2") for ax in axes: ax.set_aspect("equal") plt.show()