def test_issue_375():
# minimal example
n_pc = 3
color1, color2 = "b", "r"
ratio = NDDataset([1, 2, 3])
cum = ratio.cumsum()
ax1 = ratio.plot_bar(color=color1, title="Scree plot")
assert len(ax1.lines) == 0, "no lines"
assert len(ax1.patches) == 3, "bar present"
ax2 = cum.plot_scatter(color=color2, pen=True, markersize=7.0, twinx=ax1)
assert len(ax2.lines) == 1, "1 lines"
assert len(ax2.patches) == 0, "no bar present on the second plot"
# TODO: Don't know yet how to get the marker present.
ax1.set_title("Scree plot")
show()
# To load the API, you must import it using one of the following syntax. # # In the first syntax we load the library into a namespace called `scp` (you can choose whatever you want - except # something already in use): # %% import spectrochempy as scp # SYNTAX 1 nd = scp.NDDataset() # %% [markdown] # or in the second syntax, with a wild `*` import. # %% from spectrochempy import * nd = NDDataset() # %% [markdown] # With the second syntax, as often in python, the access to objects/functions can be greatly simplified. For example, # we can use "NDDataset" without a prefix instead of `scp.NDDataset` which is the first syntax) but there is always a # risk of overwriting some variables or functions already present in the namespace. # Therefore, the first syntax is generally highly recommended. # # Alternatively, you can also load only the onbjects and function required by your application: # # %% from spectrochempy import NDDataset nd = NDDataset()
# * `mask`: Data can be partially masked at will
# * `units`: Data can have units, allowing units-aware operations
# * `coordset`: Data can have a set of coordinates, one or several by dimensions
#
# Additional metadata can also be added to the instances of this class through the `meta` properties.
# %% [markdown]
# ## 1D-Dataset (unidimensional dataset)
# %% [markdown]
# In the following example, a minimal 1D dataset is created from a simple list, to which we can add some metadata:
# %%
d1D = NDDataset(
[10.0, 20.0, 30.0],
name="Dataset N1",
author="Blake and Mortimer",
description="A dataset from scratch",
)
d1D
# %% [markdown]
# <div class='alert alert-info'>
# <b>Note</b>
#
# In the above code, run in a notebook, the output of d1D is in html for a nice display.
#
# To get the same effect, from a console script, one can use `print_` (with an underscore) and not the usual python
# function `print`. As you can see below, the `print` function only gives a short summary of the information,
# while the `print_` method gives more detailed output
#
# </div>
def fake_dataset(*args, size=3, **kwargs):
if not args:
ds = NDDataset([range(size)])
else:
ds = NDDataset([[range(4)]], )
return ds
def test_importer(monkeypatch, fs):
fs.create_file("/var/data/xx1.txt")
assert os.path.exists("/var/data/xx1.txt")
# mock filesystem
fs.create_dir(DATADIR)
# try to read unexistent scp file
f = DATADIR / "fakedir/fakescp.scp"
with pytest.raises(FileNotFoundError):
read(f)
# make fake file
fs.create_file(f)
monkeypatch.setattr(NDDataset, "load", fake_dataset)
nd = read(f)
assert nd == fake_dataset(f)
nd = read(f.stem, directory=DATADIR / "fakedir/", protocol="scp")
assert nd == fake_dataset(f)
nd = read(f.stem, directory=DATADIR / "fakedir/")
assert nd == fake_dataset(f)
# Generic read without parameters and dialog cancel
monkeypatch.setattr(spectrochempy.core, "open_dialog", dialog_cancel)
monkeypatch.setenv(
"KEEP_DIALOGS",
"True") # we ask to display dialogs as we will mock them.
nd = read()
assert nd is None
# read as class method
nd1 = NDDataset.read()
assert nd1 is None
# NDDataset instance as first arguments
nd = NDDataset()
nd2 = nd.read()
assert nd2 is None
nd = read(default_filter="matlab")
assert nd is None
# Check if Filetype is not known
f = DATADIR / "fakedir/not_exist_fake.fk"
with pytest.raises(TypeError):
read_fake(f)
# Make fake type acceptable
FILETYPES.append(("fake", "FAKE files (*.fk)"))
ALIAS.append(("fk", "fake"))
monkeypatch.setattr("spectrochempy.core.readers.importer.FILETYPES",
FILETYPES)
monkeypatch.setattr("spectrochempy.core.readers.importer.ALIAS", ALIAS)
# Check not existing filename
f = DATADIR / "fakedir/not_exist_fake.fk"
with pytest.raises(FileNotFoundError):
read_fake(f)
# Generic read with a wrong protocol
with pytest.raises(spectrochempy.utils.exceptions.ProtocolError):
read(f, protocol="wrongfake")
# Generic read with a wrong file extension
with pytest.raises(TypeError):
g = DATADIR / "fakedir/otherfake.farfelu"
read(g)
# Mock file
f = DATADIR / "fakedir/fake.fk"
fs.create_file(f)
# specific read_(protocol) function
nd = read_fk(f)
assert nd == fake_dataset()
# should also be a Class function
nd = NDDataset.read_fk(f)
assert nd == fake_dataset()
# and a NDDataset instance function
nd = NDDataset().read_fk(f)
assert nd == fake_dataset()
# single file without protocol inferred from filename
nd = read(f)
assert nd == fake_dataset()
# single file read with protocol specified
nd = read(f, protocol="fake")
assert nd == fake_dataset()
# attribute a new name
nd = read(f, name="toto")
assert nd.name == "toto"
# mock some fake file and assume they exists
f1 = DATADIR / "fakedir/fake1.fk"
f2 = DATADIR / "fakedir/fake2.fk"
f3 = DATADIR / "fakedir/fake3.fk"
f4 = DATADIR / "fakedir/fake4.fk"
f5 = DATADIR / "fakedir/otherdir/otherfake.fk"
f6 = DATADIR / "fakedir/emptyfake.fk" # return None when reader
fs.create_file(f1)
fs.create_file(f2)
fs.create_file(f3)
fs.create_file(f4)
fs.create_file(f5)
fs.create_file(f6)
# l = list(pathclean("/Users/christian/test_data/fakedir").iterdir())
# multiple compatible 1D files automatically merged
nd = read(f1, f2, f3)
assert nd.shape == (3, 3)
nd = read([f1, f2, f3], name="fake_merged")
assert nd.shape == (3, 3)
assert nd.name == "fake_merged"
# multiple compatible 1D files not merged if the merge keyword is set to False
nd = read([f1, f2, f3], names=["a", "c", "b"], merge=False)
assert isinstance(nd, list)
assert len(nd) == 3 and nd[0] == fake_dataset()
assert nd[1].name == "c"
# do not merge inhomogeneous dataset
nd = read([f1, f2, f5])
assert isinstance(nd, list)
# too short list of names. Not applied
nd = read([f1, f2, f3], names=["a", "c"], merge=False)
assert nd[0].name.startswith("NDDataset")
monkeypatch.setattr(spectrochempy.core, "open_dialog", dialog_open)
nd = (
read()
) # should open a dialog (but to selects individual filename (here only simulated)
assert nd.shape == (2, 3)
# read in a directory
monkeypatch.setattr(pathlib.Path, "glob", directory_glob)
# directory selection
nd = read(protocol="fake", directory=DATADIR / "fakedir")
assert nd.shape == (4, 3)
nd = read(protocol="fake", directory=DATADIR / "fakedir", merge=False)
assert len(nd) == 4
assert isinstance(nd, list)
nd = read(listdir=True, directory=DATADIR / "fakedir")
assert len(nd) == 4
assert not isinstance(nd, list)
# if a directory is passed as a keyword, the behavior is different:
# a dialog for file selection occurs except if listdir is set to True
nd = read(directory=DATADIR / "fakedir", listdir=False)
assert nd.shape == (2, 3) # -> file selection dialog
nd = read(directory=DATADIR / "fakedir", listdir=True)
assert nd.shape == (4, 3) # -> directory selection dialog
# read_dir()
nd = read_dir(DATADIR / "fakedir")
assert nd.shape == (4, 3)
nd1 = read_dir()
assert nd1 == nd
nd = read_dir(
directory=DATADIR / "fakedir"
) # open a dialog to eventually select directory inside the specified
# one
assert nd.shape == (4, 3)
fs.create_file(DATADIR / "fakedir/subdir/fakesub1.fk")
nd = read_dir(directory=DATADIR / "fakedir", recursive=True)
assert nd.shape == (5, 3)
# no merging
nd = read_dir(directory=DATADIR / "fakedir", recursive=True, merge=False)
assert len(nd) == 5
assert isinstance(nd, list)
# Simulate reading a content
nd = read({"somename.fk": "a fake content"})
assert nd == fake_dataset(content=True)
nd = read_fake({"somename.fk": "a fake content"})
assert nd == fake_dataset(content=True)
# read multiple contents and merge them
nd = read({
"somename.fk": "a fake content",
"anothername.fk": "another fake content",
"stillanothername.fk": "still another fake content",
})
assert nd.shape == (3, 3)
# do not merge
nd = read(
{
"somename.fk": "a fake content",
"anothername.fk": "another fake content"
},
merge=False,
)
assert isinstance(nd, list)
assert len(nd) == 2
# handling spectroscopic information, one of the major objectives of the SpectroChemPy package: # # * `mask`: Data can be partially masked at will # * `units`: Data can have units, allowing units-aware operations # * `coordset`: Data can have a set of coordinates, one or sevral by dimensions # # Additional metadata can also be added to the instances of this class through the `meta` properties. # %% [markdown] # ## 1D-Dataset (unidimensional dataset) # %% [markdown] # In the following example, a minimal 1D dataset is created from a simple list, to which we can add some metadata: # %% d1D = NDDataset([10., 20., 30.], name="Dataset N1", author='Blake and Mortimer', description='A dataset from scratch') d1D # %% [markdown] # <div class='alert alert-info'> # <b>Note</b> # # In the above code, run in a notebook, the output of d1D is in html for a nice display. # # To get the same effect, from a console script, one can use `print_` (with an underscore) and not the usual python # function `print`. As you can see below, the `print` function only gives a short summary of the information, # while the `print_` method gives more detailed output # # </div> # %%