Drawing rectangles

[Nasuyo]

Hello,

When trying to plot (fig.plot()) rectangles, basically nothing happens..^^
Following code -- with squres -- works:

fig = pygmt.Figure()
fig.coast(region=[0, 26, 44, 62], projection='S13/90/6i', frame="ag", shorelines="0.3p,black", borders="1/0.5p,black")
fig.plot(10, 50, style='s1')
fig.show()

This should do the exact same thing, shouldn't it?:

fig = pygmt.Figure()
fig.coast(region=[0, 26, 44, 62], projection='S13/90/6i', frame="ag", shorelines="0.3p,black", borders="1/0.5p,black")
fig.plot(10, 50, style='r1/1')
fig.show()

Problem: No rectangle is drawn. Alterations fomo the style command like 'r1', 'r1,1', 'r1c', 'r1c/1c' etc. neither work.

Problem 2: Substituting r by R for a rounded rectangle produces an error:

Module 'psconvert' failed with status code 78:
psconvert [ERROR]: System call [gs -q -dSAFER -dNOPAUSE -dBATCH -sDEVICE=bbox -DPSL_no_pagefill -dMaxBitmap=2147483647 -dUseFastColor=true '/home/schroeder/.gmt/sessions/gmt6.20169/gmt_124.ps-' 2> '/home/schroeder/.gmt/sessions/gmt6.20169/psconvert_20169c.bb'] returned error 256.

I'm not sure whether this is actually a bug or just my own lack of knowledge.

Thanks in advance,
Stefan

[weiji14]

Hi Nasuyo,

Sorry for the really late reply, only got back into the flow quite recently 😳. I've had a look at the code and you're right, it only works with squares because you can pass a size argument directly into the style (e.g. style='s1' for a square of size 1). For rectangles it's a bit harder as GMT requires the length and width to be in the data column (rather than as you tried, 'r1/1'). See also https://docs.generic-mapping-tools.org/latest/plot.html#s.

One solution I can think of at the moment would be to put your x, y, length, width numbers into a textfile, and call something like pygmt.plot(data="file.txt", style="r"). Ideally though, we'll need to fix the relevant code here

pygmt/pygmt/base_plotting.py

Lines 321 to 436 in 320a728

	@fmt_docstring
	@use_alias(
	R="region",
	J="projection",
	B="frame",
	S="style",
	G="color",
	W="pen",
	i="columns",
	l="label",
	C="cmap",
	)
	@kwargs_to_strings(R="sequence", i="sequence_comma")
	def plot(self, x=None, y=None, data=None, sizes=None, direction=None, **kwargs):
	"""
	Plot lines, polygons, and symbols on maps.
	Used to be psxy.
	Takes a matrix, (x,y) pairs, or a file name as input and plots lines,
	polygons, or symbols at those locations on a map.
	Must provide either *data* or *x* and *y*.
	If providing data through *x* and *y*, *color* (G) can be a 1d array
	that will be mapped to a colormap.
	If a symbol is selected and no symbol size given, then psxy will
	interpret the third column of the input data as symbol size. Symbols
	whose size is <= 0 are skipped. If no symbols are specified then the
	symbol code (see *S* below) must be present as last column in the
	input. If *S* is not used, a line connecting the data points will be
	drawn instead. To explicitly close polygons, use *L*. Select a fill
	with *G*. If *G* is set, *W* will control whether the polygon outline
	is drawn or not. If a symbol is selected, *G* and *W* determines the
	fill and outline/no outline, respectively.
	Full option list at :gmt-docs:`plot.html`
	{aliases}
	Parameters
	----------
	x, y : float or 1d arrays
	The x and y coordinates, or arrays of x and y coordinates of the
	data points
	data : str or 2d array
	Either a data file name or a 2d numpy array with the tabular data.
	Use option *columns* (i) to choose which columns are x, y, color,
	and size, respectively.
	sizes : 1d array
	The sizes of the data points in units specified in *style* (S).
	Only valid if using *x* and *y*.
	direction : list of two 1d arrays
	If plotting vectors (using ``style='V'`` or ``style='v'``), then
	should be a list of two 1d arrays with the vector directions. These
	can be angle and length, azimuth and length, or x and y components,
	depending on the style options chosen.
	{J}
	{R}
	A : bool or str
	``'[m|p|x|y]'``
	By default, geographic line segments are drawn as great circle
	arcs. To draw them as straight lines, use *A*.
	{B}
	{CPT}
	D : str
	``'dx/dy'``: Offset the plot symbol or line locations by the given
	amounts dx/dy.
	E : bool or str
	``'[x|y|X|Y][+a][+cl|f][+n][+wcap][+ppen]'``.
	Draw symmetrical error bars.
	{G}
	S : str
	Plot symbols (including vectors, pie slices, fronts, decorated or
	quoted lines).
	{W}
	{U}
	l : str
	Add a legend entry for the symbol or line being plotted.
	"""
	kwargs = self._preprocess(**kwargs)
	kind = data_kind(data, x, y)
	extra_arrays = []
	if "S" in kwargs and kwargs["S"][0] in "vV" and direction is not None:
	extra_arrays.extend(direction)
	if "G" in kwargs and not isinstance(kwargs["G"], str):
	if kind != "vectors":
	raise GMTInvalidInput(
	"Can't use arrays for color if data is matrix or file."
	)
	extra_arrays.append(kwargs["G"])
	del kwargs["G"]
	if sizes is not None:
	if kind != "vectors":
	raise GMTInvalidInput(
	"Can't use arrays for sizes if data is matrix or file."
	)
	extra_arrays.append(sizes)
	with Session() as lib:
	# Choose how data will be passed in to the module
	if kind == "file":
	file_context = dummy_context(data)
	elif kind == "matrix":
	file_context = lib.virtualfile_from_matrix(data)
	elif kind == "vectors":
	file_context = lib.virtualfile_from_vectors(
	np.atleast_1d(x), np.atleast_1d(y), *extra_arrays
	)
	with file_context as fname:
	arg_str = " ".join([fname, build_arg_string(kwargs)])
	lib.call_module("plot", arg_str)

Alternatively, if you know the West/South/East/North coordinates of your area of interest, you can supply them directly and use style="r+s" like so:

import numpy as np
import pygmt

fig = pygmt.Figure()
fig.coast(region=[0, 26, 44, 62], projection='S13/90/6i', frame="ag", shorelines="0.3p,black", borders="1/0.5p,black")
fig.plot(data=np.array([[5,47,15,52]]), style='r+s', pen="3p,yellow")
fig.show()

gives:

[seisman]

FYI, GMT recently allows specifying rectangle's width and height from command line in GenericMappingTools/gmt#4164. It will be available in GMT 6.2.0.

Before that, the rectangle width and height must be given from the input data.

[maxrjones]

I just tested this with the development version of gmt (6.2.0_d8a7766_2021.02.11).

This now produces a rectangle:

fig = pygmt.Figure()
fig.coast(region=[0, 26, 44, 62], projection='S13/90/6i', frame="ag", shorelines="0.3p,black", borders="1/0.5p,black")
fig.plot(10, 50, style='r1/2')
fig.show()

And the rounded rectangle style produced an error because the corner radius is required. This works for a rounded rectangle:

fig = pygmt.Figure()
fig.coast(region=[0, 26, 44, 62], projection='S13/90/6i', frame="ag", shorelines="0.3p,black", borders="1/0.5p,black")
fig.plot(10, 50, style='R1/2/0.5')
fig.show()

Does the issue stay open until GMT 6.2 is released or does it get closed as soon as the problem is fixed in the main gmt branch?

[seisman]

We will keep the issue open until we bump the minimum required GMT version to 6.2.0 in PyGMT v0.4.0.

[maxrjones]

Would it be possible to use a label for these types of issues (i.e., upstream problems that are solved but require an update to the minimum required GMT version)? In reading through PyGMT issues, it appears there are at least a couple of this type of issue currently open. I think a tag would help for keeping track and quickly identifying which issues do not require any action.

[weiji14]

Would it be possible to use a label for these types of issues (i.e., upstream problems that are solved but require an update to the minimum required GMT version)? In reading through PyGMT issues, it appears there are at least a couple of this type of issue currently open. I think a tag would help for keeping track and quickly identifying which issues do not require any action.

Yeah there are quite a few of these. I think labeling it 'upstream' and adding it to the v0.4.0 milestone should suffice, but feel free to suggest a new label.