Prevent empty Line2D from creating blank tables (#134)

Empty Line2D objects will create empty PGFPlot addplot tables. These
tables are interpreted as being external data tables from file '' or
'.txt' which can cause Latex to crash.

See #134

Author: dougnd

matplotlib2tikz/line2d.py

@@ -10,6 +10,12 @@ def draw_line2d(data, obj):
     content = []
     addplot_options = []
 
+    # If line is of length 0, do nothing.  Otherwise, an empty \addplot table
+    # will be created, which will be interpreted as an external data source
+    # in either the file '' or '.tex'.  Instead, render nothing.
+    if len(obj.get_xdata()) == 0:
+        return data, []
+
     # get the linewidth (in pt)
     line_width = _mpl_linewidth2pgfp_linewidth(data, obj.get_linewidth())
 

test/testfunctions/__init__.py

@@ -3,6 +3,7 @@
 __all__ = [
   'annotate',
   'basic_sin',
+  'boxplot',
   'dual_axis',
   'errorband',
   'errorbar',

test/testfunctions/boxplot.py

@@ -0,0 +1,26 @@
+# -*- coding: utf-8 -*-
+""" Box Plot test
+
+This test plots a box plot with three data series. The causes an empty Line2D
+to be plotted.  Without care, this can turn into an empty table in PGFPlot
+which crashes latex (due to it treating an empty table as a table with 
+external data in the file '' or '.tex')
+See: https://github.com/nschloe/matplotlib2tikz/pull/134
+"""
+desc = 'BoxPlot'
+phash = '6bc6f6b95e0000fe'
+
+def plot():
+    import matplotlib.pyplot as plt
+
+    # plot data
+    fig = plt.figure()
+    ax = fig.add_subplot(111)
+
+    data = [[0.8792419963142024, 0.8842648555256405, 0.8830545971510088, 0.8831310510125482, 0.8839926059865629, 0.8795815040451961, 0.8780455489941472, 0.8785436398314896, 0.8830947020953477, 0.8853267660041949, 0.8888678711018956, 0.8852975957910832, 0.8806832729996307, 0.8757157004574541, 0.8767001155960863, 0.8840806038864472, 0.8817619814119265, 0.8888364252374024, 0.8812448127688732, 0.8831027782255365], [0.8977874209274417, 0.8941751386130553, 0.8896779411432865, 0.8971274869048325, 0.8974081692527065, 0.8942767272739647, 0.8875248054826029, 0.8777267389916926, 0.8950411839136605, 0.8927553406630346, 0.8950822278376636, 0.8987940094730611, 0.8921713177345106, 0.8875512496817447, 0.8897284821652239, 0.8910385725900226, 0.8879321741542129, 0.889056167587369, 0.884905350828982, 0.89214934207348], [0.8841888415170959, 0.8922931655807687, 0.8896153674950393, 0.8875992162118492, 0.890776178375901, 0.8889109386518265, 0.8879119743598638, 0.8912870099488378, 0.8981046527087161, 0.8920725720963792, 0.8841683225315845, 0.8857539590587772, 0.8945156112818913, 0.8894879283167035, 0.8912651966639861, 0.8929190818922158, 0.8943297597492411, 0.8888594626359189, 0.8912494597675972, 0.8917524004164856]]
+
+
+    ax.boxplot(data)
+
+    return fig
+