Exporting to PNG; phantomjs error

Windows Subsystem Linux Python 3.6.3 Anaconda 64 bit Chrome bokeh = 0.12.14

When saving to a bokeh plot to png I get RuntimeError: PhantomJS is not present in PATH. Try "conda install phantomjs" or "npm install -g phantomjs-prebuilt"

And I saw the issue and this PR.

So I tried running the steps by hand.

>>> from subprocess import Popen
>>> from subprocess import PIPE
>>> phantomjs_path = shutil.which('phantomjs')
>>> phantomjs_path
'/home/solactus/anaconda3/bin/phantomjs'
Popen([phantomjs_path, "--version"], stdout=PIPE, stderr=PIPE)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/solactus/anaconda3/lib/python3.6/subprocess.py", line 709, in __init__
    restore_signals, start_new_session)
  File "/home/solactus/anaconda3/lib/python3.6/subprocess.py", line 1344, in _execute_child
    raise child_exception_type(errno_num, err_msg, err_filename)
OSError: [Errno 8] Exec format error: '/home/solactus/anaconda3/bin/phantomjs'

And when I looked up the error and try adding ‘sh’ like this:

>>>Popen(['sh', phantomjs_path, "--version"], stdout=PIPE, stderr=PIPE)
<subprocess.Popen object at 0x7fc46df93908>

It doesn’t error out; so I was thinking maybe adding the ‘sh’ to line 67 of bokeh/util/dependencies.py?

Here’s a quick sample code that was taken from the previous issue.

from bokeh.io import export_png
from bokeh.plotting import figure, show

# prepare some data
x = [1, 2, 3, 4, 5]
y = [6, 7, 2, 4, 5]

# create a new plot with a title and axis labels
p = figure(title="simple line example", x_axis_label='x', y_axis_label='y')

# add a line renderer with legend and line thickness
p.line(x, y, legend="Temp.", line_width=2)

export_png(p, filename="plot.png")

[zbarry]

I was facing the same issue for my python 3.7 environment which was not the root. I was able to solve it by installing phantomjs in the root.

My scenario (which might have been yours) was running a JupyterLab server from one conda environment and the notebook python kernel from another. conda install -c conda-forge Phantomjs in the JupyterLab conda environment instead of the kernel environment solved the problem.

[PanditPranav]

I was facing the same issue for my python 3.7 environment which was not the root. I was able to solve it by installing phantomjs in the root.

Windows Subsystem Linux Python 3.6.3 Anaconda 64 bit Chrome bokeh = 0.12.14

When saving to a bokeh plot to png I get RuntimeError: PhantomJS is not present in PATH. Try "conda install phantomjs" or "npm install -g phantomjs-prebuilt"

And I saw the issue and this PR.

So I tried running the steps by hand.

>>> from subprocess import Popen
>>> from subprocess import PIPE
>>> phantomjs_path = shutil.which('phantomjs')
>>> phantomjs_path
'/home/solactus/anaconda3/bin/phantomjs'
Popen([phantomjs_path, "--version"], stdout=PIPE, stderr=PIPE)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/solactus/anaconda3/lib/python3.6/subprocess.py", line 709, in __init__
    restore_signals, start_new_session)
  File "/home/solactus/anaconda3/lib/python3.6/subprocess.py", line 1344, in _execute_child
    raise child_exception_type(errno_num, err_msg, err_filename)
OSError: [Errno 8] Exec format error: '/home/solactus/anaconda3/bin/phantomjs'

And when I looked up the error and try adding ‘sh’ like this:

>>>Popen(['sh', phantomjs_path, "--version"], stdout=PIPE, stderr=PIPE)
<subprocess.Popen object at 0x7fc46df93908>

It doesn’t error out; so I was thinking maybe adding the ‘sh’ to line 67 of bokeh/util/dependencies.py?

Here’s a quick sample code that was taken from the previous issue.

from bokeh.io import export_png
from bokeh.plotting import figure, show

# prepare some data
x = [1, 2, 3, 4, 5]
y = [6, 7, 2, 4, 5]

# create a new plot with a title and axis labels
p = figure(title="simple line example", x_axis_label='x', y_axis_label='y')

# add a line renderer with legend and line thickness
p.line(x, y, legend="Temp.", line_width=2)

export_png(p, filename="plot.png")

I was facing the same issue for my python 3.7 environment which was not the root. I was able to solve it by installing phantomjs in the root.