Possible text bug with fig.meca when plotting from a file

[hazardgoat]

Description of the problem
I think I may have run across a bug with fig.meca when attempting to plot from a file. Column names are being plotted in the bottom left corner, and the issue disappears when column names are removed from the file.

Full code that generated the error

import pygmt
import os

# Main File Path ----------------------------------
main_dir = r'C:\Users\USER\Desktop\TESTING'

# Save Path ----------------------------------
save_name = os.path.join(main_dir, 'Results', 'map_TEST.png')

# Focal Mechanism Path ----------------------------------    
focal_mecha = os.path.join(main_dir, 'Data', 'selected_fm.csv')


fig = pygmt.Figure()

fig.basemap(
    region = [-115.85, -115.35, 32.80, 33.38],
    projection = 'M6i',
    frame = True
)
fig.coast(
    land = 'grey',
    water = 'lightskyblue2',
)
fig.meca(
    spec = focal_mecha,
    convention = 'aki',
    scale = '1.0c',
    offset = True,
    )

fig.savefig(save_name)
print('done')

System information
PyGMT information:
version: v0.4.0
System information:
python: 3.9.5 (tags/v3.9.5:0a7dcbd, May 3 2021, 17:27:52) [MSC v.1928 64 bit (AMD64)]
executable: C:\Users\USER\AppData\Local\Programs\Python\Python39\python.exe
machine: Windows-10-10.0.19043-SP0
Dependency information:
numpy: 1.21.0
pandas: 1.2.5
xarray: 0.18.2
netCDF4: 1.5.7
packaging: 20.9
ghostscript: 9.54.0
gmt: 6.2.0
GMT library information:
binary dir: C:/Users/USER/AppData/Local/Programs/Python/Python39
cores: 4
grid layout: rows
library path: c:/programs/gmt6/bin/gmt_w64.dll
padding: 2
plugin dir: c:/programs/gmt6/bin/gmt_plugins
share dir: c:/programs/gmt6/share
version: 6.2.0

My data:
selected_fm.csv

My map output:

[seisman]

It's not a bug. For text files with a header like

longitude	latitude	depth	strike	dip	rake	magnitude	offset_longitude	offset_latitude	label_mag

you have to tell GMT or PyGMT to skip the header.

[seisman]

@PaulWessel I'm wondering if meca can be more clever about these headers.

You can run the following command to reproduce the issue. You will see the "header line" appear at the lower left corner of the map:

gmt meca -Sa1c -R-116/-115/32.5/33.5 -Baf -pdf map << EOF
longitude	latitude	depth	strike	dip	rake	magnitude	offset_longitude	offset_latitude	label_mag
-115.52666 33.01917 10.11 314 85 -171 5.32 -115.52666 33.17
-115.52783 33.017 8.69 143 87 158 5.44 -115.52783 32.117
-115.6275 33.08567 9.29 158 86 -151 5.75 -115.6275 33.18567
-115.78167 33.08183 8.2 280 86 171 6.2 -115.78167 33.18183
-115.83434 33.01367 5.45 127 72 -159 6.6 -115.83434 33.11367
-115.634 33.14283 5.64 190,62 -106 5.11 -115.634 33.24283
EOF

But, it seems plot can skip the header line quietly.

gmt plot -R-116/-115/32.5/33.5 -Baf -Sc0.2c -pdf points << EOF
longitude	latitude	depth	strike	dip	rake	magnitude	offset_longitude	offset_latitude	label_mag
-115.52666 33.01917 10.11 314 85 -171 5.32 -115.52666 33.17
-115.52783 33.017 8.69 143 87 158 5.44 -115.52783 32.117
-115.6275 33.08567 9.29 158 86 -151 5.75 -115.6275 33.18567
-115.78167 33.08183 8.2 280 86 171 6.2 -115.78167 33.18183
-115.83434 33.01367 5.45 127 72 -159 6.6 -115.83434 33.11367
-115.634 33.14283 5.64 190,62 -106 5.11 -115.634 33.24283
EOF

[seisman]

This is caused by an upstream issue and has been fixed by GenericMappingTools/gmt#5427.

[weiji14]

So running conda install -c conda-forge/label/dev gmt=6.3.0 or building GMT from source (see https://github.com/GenericMappingTools/gmt/blob/master/BUILDING.md) should allow users to get the patched version of GMT.

Not sure if we need to add a test for this (probably not), but I think we can close this in PyGMT v0.6.0 once we bump the minimum GMT version to 6.3.0, agree?

[seisman]

I think we can close the issue without adding a test.