"""
project - Project data onto lines or great circles, or generate tracks.
"""
import pandas as pd
from pygmt.clib import Session
from pygmt.exceptions import GMTInvalidInput
from pygmt.helpers import (
GMTTempFile,
build_arg_string,
data_kind,
dummy_context,
fmt_docstring,
kwargs_to_strings,
use_alias,
)
[docs]@fmt_docstring
@use_alias(
A="azimuth",
E="endpoint",
F="flags",
G="generate",
L="length",
N="flatearth",
Q="maptypeunits",
S="sort",
T="rotationpole",
V="verbose",
W="width",
Z="ellipse",
f="coltypes",
)
@kwargs_to_strings(E="sequence", L="sequence", T="sequence", W="sequence")
def project(points, center, outfile=None, **kwargs):
"""
Project data onto lines or great circles, or generate tracks.
Project reads arbitrary (x, y [, z]) data and returns any combination of
(x, y, z, p, q, r, s), where (p, q) are the coordinates in the projection,
(r, s) is the position in the (x, y) coordinate system of the point on the
profile (q = 0 path) closest to (x, y), and z is all remaining columns in
the input (beyond the required x and y columns).
Alternatively, Project may be used to generate (r, s, p) triples at equal
increments ``dist`` along a profile. In this case (``generate`` option),
the value of ``points`` is ignored (you can use, e.g., ``points=None``).
Projections are defined in any (but only) one of three ways:
(Definition 1) By a ``center`` and an ``azimuth`` in degrees
clockwise from North.
(Definition 2) By a ``center`` and ``endpoint`` of the projection path.
(Definition 3) By a ``center`` and a ``rotationpole`` position.
To spherically project data along a great circle path, an oblique
coordinate system is created which has its equator along that path, and the
zero meridian through the Center. Then the oblique longitude (p)
corresponds to the distance from the Center along the great circle, and the
oblique latitude (q) corresponds to the distance perpendicular to the great
circle path. When moving in the increasing (p) direction, (toward B or in
the azimuth direction), the positive (q) direction is to your left. If a
Pole has been specified, then the positive (q) direction is toward the
pole.
To specify an oblique projection, use the ``rotationpole`` option to set
the pole. Then the equator of the projection is already determined and the
``center`` option is used to locate the p = 0 meridian. The center cx/cy
will be taken as a point through which the p = 0 meridian passes. If you do
not care to choose a particular point, use the South pole (ox = 0, oy =
-90).
Data can be selectively windowed by using the ``length`` and ``width``
options. If ``width`` is used, the projection width is set to use only
points with w_min < q < w_max. If ``length`` is set, then the length is set
to use only those points with l_min < p < l_max. If the ``endpoint`` option
has been used to define the projection, then ``length="w"`` may be used to
window the length of the projection to exactly the span from O to B.
Flat Earth (Cartesian) coordinate transformations can also be made. Set
``flatearth=True`` and remember that azimuth is clockwise from North (the
y axis), NOT the usual cartesian theta, which is counterclockwise from the
x axis. azimuth = 90 - theta.
No assumptions are made regarding the units for x, y, r, s, p, q, dist,
l_min, l_max, w_min, w_max. If -Q is selected, map units are assumed and x,
y, r, s must be in degrees and p, q, dist, l_min, l_max, w_min, w_max will
be in km.
Calculations of specific great-circle and geodesic distances or for
back-azimuths or azimuths are better done using :gmt-docs:`mapproject` as
project is strictly spherical.
Project is CASE SENSITIVE: use lower case for the **xyzpqrs** letters
in ``flags``.
{aliases}
Parameters
----------
points : pandas.DataFrame or str
Either a table with (x, y) or (lon, lat) values in the first two
columns, or a filename (e.g. csv, txt format). More columns may be
present.
center : str or list
*cx*/*cy*.
*cx/cy* sets the origin of the projection, in Definition 1 or 2. If
Definition 3 is used, then *cx/cy* are the coordinates of a
point through which the oblique zero meridian (*p* = 0) should pass.
The *cx/cy* is not required to be 90 degrees from the pole.
azimuth : float or str
defines the azimuth of the projection (Definition 1).
endpoint : str or list
*bx*/*by*.
*bx/by* defines the end point of the projection path (Definition 2).
flags : str
Specify your desired output using any combination of **xyzpqrs**, in
any order [Default is **xyzpqrs**]. Do not space between the letters.
Use lower case. The output will be columns of values corresponding to
your ``flags``. The **z** flag is special and refers to all numerical
columns beyond the leading **x** and **y** in your input record. The
**z** flag also includes any trailing text (which is placed at the end
of the record regardless of the order of **z** in ``flags``). **Note**:
If ``generate`` is True, then the output order is hardwired to be
**rsp** and ``flags`` is not allowed.
generate : str
*dist* [/*colat*][**+c**|**h**].
Generate mode. No input is read and the value of ``points`` is ignored
(you can use, e.g., ``points=None``). Create (*r*, *s*, *p*) output
points every *dist* units of *p*. See `maptypeunits` option.
Alternatively, append */colat* for a small circle instead [Default is a
colatitude of 90, i.e., a great circle]. If setting a pole with
``rotationpole`` and you want the small circle to go through *cx*/*cy*,
append **+c** to compute the required colatitude. Use ``center`` and
``endpoint`` to generate a circle that goes through the center and end
point. Note, in this case the center and end point cannot be farther
apart than 2*| *colat* |. Finally, if you append **+h** then we will
report the position of the pole as part of the segment header [no
header].
length : str or list
[**w**| *l_min*/*l_max*].
Length controls. Project only those points whose *p* coordinate is
within *l_min* < *p* < *l_max*. If ``endpoint`` has been set, then you
may alternatively use **w** to stay within the distance from ``center``
to ``endpoint``.
flatearth : bool
If `True`, Make a Cartesian coordinate transformation in the plane.
[Default uses spherical trigonometry.]
maptypeunits : bool
If `True`, project assumes *x*, *y*, *r*, *s* are in degrees while *p*,
*q*, *dist*, *l_min*, *l_max*, *w_min*, *w_max* are in km. If not set
(or `False`), then all these are assumed to be in the same units.
sort : bool
Sort the output into increasing *p* order. Useful when projecting
random data into a sequential profile.
rotationpole : str or list
*px*/*py*.
*px/py* sets the position of the rotation pole of the projection.
(Definition 3).
{V}
width : str or list
*w_min*/*w_max*
Width controls. Project only those points whose *q* coordinate is
within *w_min* < *q* < *w_max*.
ellipse : str
*major*/*minor*/*azimuth* [**+e** | **n**].
Used in conjunction with ``center`` (sets its center) and ``generate``
(sets the distance increment) to create the coordinates of an ellipse
with *major* and *minor* axes given in km (unless ``flatearth`` is
given for a Cartesian ellipse) and the *azimuth* of the major axis in
degrees. Append **+e** to adjust the increment set via ``generate`` so
that the the ellipse has equal distance increments [Default uses the
given increment and closes the ellipse]. Instead, append **+n** to set
a specific number of unique equidistant points via ``generate``. For
degenerate ellipses you can just supply a single *diameter* instead. A
geographic diameter may be specified in any desired unit other than km
[Default] by appending the unit (e.g., 3d for degrees); if so we assume
the increment is also given in the same unit (see `Units`_). **Note**:
For the Cartesian ellipse (which requires ``flatearth``), we expect
*direction* counter-clockwise from the horizontal instead of an
*azimuth*.
outfile : str
Required if ``points`` is a file. The file name for the output ASCII
file.
{f}
Returns
-------
track: pandas.DataFrame or None
Return type depends on whether the ``outfile`` parameter is set:
- :class:`pandas.DataFrame` table with (x, y, ..., newcolname) if
``outfile`` is not set
- None if ``outfile`` is set (track output will be stored in file set
by ``outfile``)
"""
# center ("C") is a required positional argument, so it cannot be
# processed by decorator `@kwargs_to_strings`
kwargs["C"] = "/".join(f"{item}" for item in center)
with GMTTempFile(suffix=".csv") as tmpfile:
if outfile is None: # Output to tmpfile if outfile is not set
outfile = tmpfile.name
with Session() as lib:
if "G" not in kwargs:
# Store the pandas.DataFrame points table in virtualfile
if data_kind(points) == "matrix":
table_context = lib.virtualfile_from_matrix(points.values)
elif data_kind(points) == "file":
if outfile is None:
raise GMTInvalidInput("Please pass in a str to 'outfile'")
table_context = dummy_context(points)
else:
raise GMTInvalidInput(f"Unrecognized data type {type(points)}")
# Run project on the temporary (csv) points table
with table_context as csvfile:
arg_str = " ".join(
[csvfile, build_arg_string(kwargs), "->" + outfile]
)
lib.call_module(module="project", args=arg_str)
else:
arg_str = " ".join([build_arg_string(kwargs), "->" + outfile])
lib.call_module(module="project", args=arg_str)
# if user did not set outfile, return pd.DataFrame
if outfile == tmpfile.name:
if "G" in kwargs:
column_names = list("rsp")
else:
# Set output column names according to the "F" flag or use
# default value
if "F" in kwargs:
column_names = list(kwargs["F"])
else:
column_names = list("xyzpqrs")
# Find the indexes of "x", "y" and "z" column and
# replace with input column names
i_x = column_names.index("x")
i_y = column_names.index("y")
i_z = column_names.index("z")
input_column_names = points.columns.to_list()
column_names[i_x] = input_column_names[0]
column_names[i_y] = input_column_names[1]
# "z" can be actually more than one column
column_names.pop(i_z)
for col in reversed(input_column_names[2:]):
column_names.insert(i_z, col)
result = pd.read_csv(tmpfile.name, sep="\t", names=column_names)
# return None if outfile set, output in outfile
elif outfile != tmpfile.name:
result = None
return result