How to export a matplotlib plot from a Python data function to Spotfire as an image

[Artur Veloso]

Dear Community

I am running an analysis in a Python data function that leads to a complicated graph which I would not be able to plot using Spotfire's built-in visualizations.

I would like to create the plot within Python (using matplotlib) and display it Spotfire. In the past I have done this with R data functions following this tutorial:

https://community.spotfire.com/wiki/spotfire-tips-tricks-create-r-graphics-...

The Spotfire Python Data Functions guide says that matplotlib figure objects can be return to Spotfire from Python:

https://docs.tibco.com/pub/sfire-analyst/latest/doc/pdf/TIB_sfire-analys...

 

I have tried out the code:

import numpy as np

import matplotlib

matplotlib.use('Agg')

 

import matplotlib.pyplot as plt

 

 

ax = plt.subplot(111)

x = np.linspace(0, 10)

y = np.exp(x)

w = plt.plot(x, y)

where w is my output variable (Document Property, Binary), but that fails with the following error message:

 

Could not execute function call 'Untitled'

 

Error executing Python script:

 

spotfire.sbdf.SBDFError: unknown type 'matplotlib.lines.Line2D' in list

 

Traceback (most recent call last):

File "data_function.py", line 358, in _write_outputs

output.write(self.globals, self.debug)

File "data_function.py", line 141, in write

sbdf.export_data(globals_dict[self.name], self.file, default_column_name=self.name)

File "sbdf.py", line 160, in export_data

default_column_name)

File "sbdf.py", line 243, in _export_columnize_data

column_types = {default_column_name: _ValueTypeId.infer_from_type(list(obj), "list")}

File "sbdf.py", line 1007, in infer_from_type

raise SBDFError("unknown type '%s' in %s" % (_utils.type_name(vals_type), value_description))

 

 

at Spotfire.Dxp.Data.DataFunctions.Executors.LocalPythonFunctionClient.d__8.MoveNext()

at Spotfire.Dxp.Data.DataFunctions.Executors.PythonScriptExecutor.d__11.MoveNext()

at Spotfire.Dxp.Data.DataFunctions.DataFunctionExecutorService.d__8.MoveNext()

 

 

TheSpotfire build I am using is 10.10.0.64.

Does anyone have any suggestions

 

Thank you!

[Floris Remmen]

Hi

 

The function needs to return a figure object. If you instantiate a figure like so:

```

import numpy as np

import matplotlib

matplotlib.use('Agg')

import matplotlib.pyplot as plt

 

f, ax = plt.subplots()

x = np.linspace(0, 10)

y = np.exp(x)

w = ax.plot(x, y)

```

and you return the 'f' object into a document property it works.

[Osama Alsalmi]

use these two functions:

import os

import matplotlib.pyplot as plt

import tempfile

import bitstring

from PIL import Image

import numpy as np

 

 

def _image_file_to_binary(file):

bits = bitstring.BitArray(filename=file.name)

img = bits.bytes

os.unlink(file.name)

return img

 

def _pyplot_to_binary(fig):

fig.set_canvas(plt.gcf().canvas)

file = tempfile.NamedTemporaryFile(delete=False)

fig.savefig(file)

file.close()

return _image_file_to_binary(file)#Example

 

output_binary_document_prop = _pyplot_to_binary()Finally, to view this make sure you to input this as a lable in the text area.