import sys, json, time, re
import util
”’ ===============================================================================================
Copyright By PowCoder代写 加微信 powcoder
This class is used to convert a .ipynb file into a .py file for grading using the unittest
python framework
===============================================================================================
def seed_random(source):
”’ This function is used to seed the random library with a common
seed so that the test cases can compare apples-to-apples
source_as_list = source.split(‘\n’)
for i in range(len(source_as_list)):
line = source_as_list[i]
location = re.search(‘import\s+random’, line)
if location is not None:
# Find indentation level, then match and append
prefix = ”
for c in line:
if c == ‘ ‘ or c == ‘\t’:
prefix += c
line += ‘\n{}random.seed(0)’.format(prefix)
source_as_list[i] = line
source = ‘\n’.join(source_as_list)
return source.split(‘\n’)
def list_to_dict_str(the_list):
”’ This function creates a string version of a list to be written
to the output file as a dictionary. Pass in a list, you’ll receive
a dictionary – it’s used to parse the check_val field
ret_val = ‘{‘
for i in range(len(the_list)):
ret_val += ‘”{}”: {}’.format(the_list[i], the_list[i])
if i == len(the_list) – 1:
ret_val += ‘}’
ret_val += ‘, ‘
return ret_val
def parse_cell(cell, outfile, nb_version):
”’ This function will parse a cell and write it to the outfile
if cell[‘cell_type’] == ‘code’:
ret = parse_code_cell(cell, outfile, nb_version)
outfile.write(‘\n\n’)
return ret
return 0, 0
def parse_code_cell(cell, outfile, nb_version):
”’ This function will format a new .py file from .ipynb with structure such that
each named cell (name must be in metadata) is assigned as a variable to
source = cell[‘source’] if nb_version == ‘old’ else cell[‘input’]
source_str = ‘\n’.join(source)
if len(re.findall(‘import\s+random’, source_str)) > 0 or len(re.findall(‘from\s+random’, source_str)) > 0:
source = seed_random(source_str)
if ‘name’ not in cell[‘metadata’].keys():
# Non-named cell, don’t care, just write it
source_to_write = []
for line in source:
for char in line:
if char == ‘ ‘ or char == ‘\t’:
elif char == ‘%’:
source_to_write.append(line)
outfile.write(‘\n’.join(source_to_write))
return 0, 1
name = cell[‘metadata’][‘name’]
if ‘check_val’ in cell[‘metadata’].keys():
# Override case. This is where we tell the script exactly
# which variables to check, and is preferable
outfile.write(‘\n’.join(source))
outfile.write(‘\n\n{} = {}’.format(cell[‘metadata’][‘name’], list_to_dict_str(cell[‘metadata’][‘check_val’])))
elif ‘def’ in ‘\n’.join(source):
# If defined function, get the name of the function and pass that back
# as the return value
func_name = ‘\n’.join(source).split(‘ ‘)[1].split(‘(‘)[0]
outfile.write(‘\n’.join(source))
outfile.write(‘\n\n{} = {}’.format(name, func_name))
elif ‘print’ in ‘\n’.join(source):
# If there are print statements, use those instead
outfile.write(‘\n’.join(source))
text_to_write = cell[‘outputs’][0][‘text’]
for i in range(len(text_to_write)):
text_to_write[i] = text_to_write[i].strip()
outfile.write(‘\n\n{} = {}’.format(name, text_to_write))
# In line case
for i in range(len(source)):
if i == len(source) – 1:
to_write = ”
for c in source[i]:
if c == ‘\t’ or c == ‘ ‘:
to_write += c
stop_location = len(to_write)
to_write += “{} = “.format(name)
to_write += source[i][stop_location:]
outfile.write(to_write)
outfile.write(source[i])
return 1, 0
def convert_notebook(infile, outfile_arg, log_path):
“”” Convert a jupyter notebook to python, write it to the logging location
logger = open(“{}/nbconvert.log”.format(log_path), “a+”)
logger.write(‘\n\n\n\n======================================================================\n’)
logger.write(‘{}\n’.format(util.get_datetime()))
logger.write(‘ > Beginning conversion of {}\n’.format(infile))
f = open(infile)
contents = f.read()
contents = json.loads(contents)
cells = contents[‘cells’] if ‘cells’ in contents.keys() else contents[‘worksheets’][0][‘cells’]
nb_version = ‘old’ if ‘cells’ in contents.keys() else ‘new’
logger.write(‘ > Found {} cells\n’.format(len(cells)))
outfile = open(‘{}’.format(outfile_arg), ‘w+’)
outfile.write(‘#!/bin/python3\n\n\n’)
definitions, declarations = 0, 0
for cell in cells:
x, y = parse_cell(cell, outfile, nb_version)
definitions += x
declarations += y
logger.write(‘ > Wrote {} function definition cells\n’.format(definitions))
logger.write(‘ > Wrote {} declarative cells\n’.format(declarations))
logger.write(‘======================================================================\n’)
logger.write(‘ > Finished writing to {}\n’.format(outfile_arg))
logger.write(‘======================================================================\n’)
def main():
convert_notebook(sys.argv[1], sys.argv[2])
if __name__ == ‘__main__’:
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com