def visualize_as_box_plot():
font = {'size': 20}
plt.rc('font', **font)
project_list = project_config.get('projects', 'project_list').split(",")
# if len(project_list) > 1:
# print("reduce number of projects to 1")
# exit(0)
project_list = project_list[0]
statement_coverage_r = []
checked_coverage_r = []
file_name = "/point_biserial_correlation__full" + ".csv"
path = str(get_project_root()) + results_folder + file_name
try:
with open(path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
statement_coverage_r.append(
float(row['adequacy_based_tests_statement']))
checked_coverage_r.append(
float(row['adequacy_based_tests_checked']))
except FileNotFoundError:
pass
data_to_plot = [statement_coverage_r, checked_coverage_r]
fig = plt.figure(1, figsize=(9, 6))
ax = fig.add_subplot()
ax.set_ylabel('Probabilistic coupling')
# ax.set_xlabel('Indicates whether or not, a bug detecting test is included in generated test suite')
# ax.set_title(project_list)
ax.set_xticks([1, 2])
ax.set_xticklabels(tuple(['Statement coverage', 'Checked coverage']))
bp = ax.violinplot(data_to_plot, showmeans=True, showmedians=True)
# loop over the paths of the mean lines
xy = [[l.vertices[:, 0].mean(), l.vertices[0, 1]]
for l in bp['cmeans'].get_paths()]
xy = np.array(xy)
ax.scatter(xy[:, 0], xy[:, 1], s=121, c="crimson", marker="x", zorder=3)
# make lines invisible
bp['cmeans'].set_visible(False)
bp['bodies'][0].set(color='#e69138', linewidth=2)
bp['bodies'][1].set(color='#3d85c6', linewidth=2)
ax.legend().remove()
plt.show()
save_path = str(get_project_root()) + results_folder + '/' + str(
project_list) + '_violin-plot'
fig.savefig(save_path, dpi=100)
def visualize_as_violin_plot():
font = {'size': 20}
plt.rc('font', **font)
project_list = project_config.get('projects', 'project_list').split(",")
if len(project_list) > 1:
print("reduce number of projects to 1")
exit(0)
project_list = project_list[0]
statement_increase = []
checked_increase = []
mutation_increase = []
file_name = "/" + project_list + ".csv"
path = str(get_project_root()) + results_folder + file_name
with open(path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
statement_increase.append(float(row['statement_coverage_increase']))
checked_increase.append(float(row['checked_coverage_increase'])
if float(row['checked_coverage_increase']) > 0 else 0)
mutation_increase.append(float(row['mutation_coverage_increase'])
if float(row['mutation_coverage_increase']) > 0 else 0)
data_to_plot = [statement_increase, checked_increase, mutation_increase]
fig = plt.figure(1, figsize=(9, 6))
ax = fig.add_subplot()
ax.set_ylabel('% coverage score increase')
# ax.set_xlabel('Indicates whether or not, a bug detecting test is included in generated test suite')
ax.set_title(project_list)
ax.set_xticks([1,2,3])
ax.set_xticklabels(tuple(['Statement coverage', 'Checked coverage', 'Mutation score']))
bp = ax.violinplot(data_to_plot, showmeans=True, showmedians=True)
# loop over the paths of the mean lines
xy = [[l.vertices[:, 0].mean(), l.vertices[0, 1]] for l in bp['cmeans'].get_paths()]
xy = np.array(xy)
ax.scatter(xy[:, 0], xy[:, 1], s=121, c="crimson", marker="x", zorder=3)
# make lines invisible
bp['cmeans'].set_visible(False)
bp['bodies'][0].set(color='#e69138', linewidth=2)
bp['bodies'][1].set(color='#3d85c6', linewidth=2)
bp['bodies'][2].set(color='#6aa84f', linewidth=2)
ax.legend().remove()
plt.show()
save_path = str(get_project_root()) + results_folder + '/' + str(project_list) + '_violin-plot'
fig.savefig(save_path, dpi=100)
def visualize_as_violin_plot():
font = {'size': 20}
plt.rc('font', **font)
statement_increase = []
checked_increase = []
mutation_increase = []
file_name = "/max_pc" + ".csv"
path = str(get_project_root()) + results_folder + file_name
with open(path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
statement_increase.append(float(row['statement_pc_max']))
checked_increase.append(float(row['checked_pc_max']))
# mutation_increase.append(float(row['mutation_coverage_increase'])
# if float(row['mutation_coverage_increase']) > 0 else 0)
# statement_increase = list(filter(lambda num: num != 0, statement_increase))
# checked_increase = list(filter(lambda num: num != 0, checked_increase))
data_to_plot = [statement_increase, checked_increase]
fig = plt.figure(1, figsize=(9, 6))
ax = fig.add_subplot()
ax.set_ylabel('Max PC')
# ax.set_xlabel('Indicates whether or not, a bug detecting test is included in generated test suite')
ax.set_title("Probabilistic Coupling")
ax.set_xticks([1, 2, 3])
if len(mutation_increase) > 0:
ax.set_xticklabels(
tuple(['Statement cov.', 'Checked cov.', 'Mutation score']))
else:
ax.set_xticklabels(tuple(['Statement coverage', 'Checked coverage']))
bp = ax.violinplot(data_to_plot, showmeans=True, showmedians=True)
# loop over the paths of the mean lines
xy = [[l.vertices[:, 0].mean(), l.vertices[0, 1]]
for l in bp['cmeans'].get_paths()]
xy = np.array(xy)
ax.scatter(xy[:, 0], xy[:, 1], s=121, c="crimson", marker="x", zorder=3)
# make lines invisible
bp['cmeans'].set_visible(False)
bp['bodies'][0].set(color='#e69138', linewidth=2)
bp['bodies'][1].set(color='#3d85c6', linewidth=2)
ax.legend().remove()
plt.show()
save_path = str(
get_project_root()) + results_folder + '/max_pc_violin-plot'
fig.savefig(save_path, dpi=100)
def read_file_and_visualise_pc():
project_list = project_config.get('projects', 'project_list').split(",")
for project in project_list:
project_range = project_config.get('projects', project).split(",")
project_count = 0
statement_increase = []
checked_increase = []
mutation_coverage = []
coverage_scores = []
for project_id in range(int(project_range[0]),
int(project_range[1]) + 1):
file_name = "/prob_coupling_" + project + "_" + str(
project_id) + ".csv"
file_path = str(get_project_root()) + results_folder + file_name
mutation_file_name = "/mutant_pc_" + project + "_" + str(
project_id) + ".csv"
mutation_file_path = str(
get_project_root()) + results_folder + mutation_file_name
if path.isfile(file_path):
project_count = project_count + 1
with open(file_path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
statement_increase.append(float(row['statement_pc']))
checked_increase.append(float(row['checked_pc']))
coverage_scores.append(
(float(row['checked_pc']),
float(row['statement_pc']), 0,
row['checked_nature'], row['statement_nature'],
row['checked_modified'],
row['statement_modified']))
if path.isfile(mutation_file_path):
with open(mutation_file_path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
#mutation_coverage.append(float(row['pc_score']))
if float(row['pc_score']) > 0:
mutation_coverage.append(float(row['pc_score']))
save_path = str(
get_project_root()) + results_folder + '/' + project
# visualize_as_scatterplot_skew(coverage_scores, save_path,
# "Probabilistic Coupling: " + project + "-" + str(project_id) +
# "\n Bug count: " + str(project_count), "Count of statements")
visualize_as_violin_plot_jitter(checked_increase, statement_increase,
mutation_coverage, save_path,
"Probabilistic Coupling: " + project,
"PC score")
def visualize_correlation_as_bar():
names = []
checked_cor_score = []
statement_cor_score = []
with open(get_correlation_file()) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
checked_cor_score.append(float(row['checked']))
statement_cor_score.append(float(row['statement']))
names.append(row['name'])
df = pandas.DataFrame(
dict(graph=names[::-1],
n=checked_cor_score[::-1],
m=statement_cor_score[::-1]))
ind = np.arange(len(df))
width = 0.2
fig, ax = plt.subplots()
ax.barh(ind, df.n, width, label='Checked Coverage')
ax.barh(ind + width, df.m, width, label='Statement Coverage')
ax.set(yticks=ind + width, yticklabels=df.graph, ylim=[width - 4, len(df)])
ax.legend()
fig.set_size_inches(7, 15, forward=True)
plt.show()
file_name_to_save = "Cli"
save_path = str(
get_project_root()) + results_folder + '/' + str(file_name_to_save)
fig.savefig(save_path, dpi=100)
def visualize_as_box_plot():
font = {'size': 20}
plt.rc('font', **font)
project_list = project_config.get('projects', 'project_list').split(",")
if len(project_list) > 1:
print("reduce number of projects to 1")
exit(0)
project_list = project_list[0]
statement_increase = []
checked_increase = []
mutation_increase = []
file_name = "/" + project_list + ".csv"
path = str(get_project_root()) + results_folder + file_name
with open(path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
statement_increase.append(float(row['statement_coverage_increase']))
checked_increase.append(float(row['checked_coverage_increase'])
if float(row['checked_coverage_increase']) > 0 else 0)
mutation_increase.append(float(row['mutation_coverage_increase'])
if float(row['mutation_coverage_increase']) > 0 else 0)
data_to_plot = [statement_increase, checked_increase, mutation_increase]
fig = plt.figure(1, figsize=(9, 6))
ax = fig.add_subplot(111)
ax.set_ylabel('% coverage score increase')
# ax.set_xlabel('Indicates whether or not, a bug detecting test is included in generated test suite')
ax.set_title(project_list)
# ax.set_xticks(ind + width / 2)
ax.set_xticklabels(tuple(['Statement coverage', 'Checked coverage', 'Mutation score']))
bp = ax.boxplot(data_to_plot)
bp['medians'][0].set(color='#3d85c6', linewidth=2)
bp['medians'][1].set(color='#e69138', linewidth=2)
bp['medians'][2].set(color='#6aa84f', linewidth=2)
ax.legend().remove()
plt.show()
save_path = str(get_project_root()) + results_folder + '/' + str(project_list) + '_box-plot'
def get_configured_csv_path():
project_list = project_config.get('projects', 'project_list').split(",")
if len(project_list) > 1:
print("reduce number of projects to 1")
exit(0)
project_list = project_list[0]
return str(get_project_root()) + results_folder + '/' + project_list + '.csv'
def get_latest_csv_path():
path = str(get_project_root()) + results_folder + '/*.csv'
list_of_files = glob.glob(path)
if len(list_of_files) <= 0:
print("compute the solution first")
exit(0)
latest_file = max(list_of_files, key=os.path.getctime)
return latest_file
def clean_source_dir():
for root, dirs, files in os.walk(defects4j_project_path):
folder_path = os.path.join(str(get_project_root()), root)
if is_deletable(folder_path):
shutil.rmtree(folder_path, ignore_errors=False, onerror=None)
print("deleted " + folder_path)
for file in files:
if is_deletable(str(file)):
file_path = folder_path + '/' + str(file)
os.remove(file_path)
print("deleted " + file_path)
def visualize_by_project_id(start, end):
with open(get_latest_csv_path()) as csv_file:
n = [int(start), int(end)]
file_name_to_save = os.path.basename(csv_file.name).split(".")[0]
reader = csv.DictReader(csv_file, delimiter=',')
sorted_list = sorted(reader, key=lambda c_row: int(c_row['project_id']), reverse=False)
ratio_statement_coverage = []
ratio_checked_coverage = []
project_ids = []
checked_coverage_label = []
statement_coverage_label = []
project_name = ','.join(set([s['project'] for s in sorted_list]))
for row in islice(sorted_list, n[0], n[1]):
ratio_checked_coverage.append(float(row['ratio_checked_coverage']))
ratio_statement_coverage.append(float(row['ratio_statement_coverage']))
project_ids.append(project_name + '-' + row['project_id'])
statement_coverage_label.append(row['len_of_bug_detecting_tests'] + '/' +
row['len_of_statement_covering_tests'])
checked_coverage_label.append(row['len_of_bug_detecting_tests'] + '/' +
row['len_of_checked_covering_tests'])
ind = np.arange(len(range(n[0], n[1])))
width = 0.40
fig, ax = plt.subplots()
# rects1 = ax.bar(ind, ratio_checked_coverage, width, color='r')
# rects2 = ax.bar(ind + width, ratio_statement_coverage, width, color='y')
checked_coverage_bars = ax.barh(ind, ratio_checked_coverage, width)
statement_coverage_bars = ax.barh(ind + width, ratio_statement_coverage, width)
# add some text for labels, title and axes ticks
ax.set_ylabel('Project bug Id')
ax.set_xlabel('Ratio')
ax.set_yticks(ind + width/2)
ax.set_yticklabels(tuple(project_ids))
formula_legend = Rectangle((0, 0), 1, 1, fc="w", fill=False, edgecolor='none', linewidth=0)
ax.legend((formula_legend, checked_coverage_bars[0], statement_coverage_bars[0]),
('Ratio = No. of bug detecting tests / No. of Covering tests',
'Checked Coverage',
'Statement Coverage'))
autolabel(checked_coverage_bars, checked_coverage_label)
autolabel(statement_coverage_bars, statement_coverage_label)
fig.set_size_inches(7, 15)
# save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save) + '_' +
# str(max(project_ids))
save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save)
fig.savefig(save_path, dpi=200)
plt.show()
def visualize_by_project_id(start, end):
with open(get_latest_csv_path()) as csv_file:
n = [int(start), int(end)]
file_name_to_save = os.path.basename(csv_file.name).split(".")[0]
reader = csv.DictReader(csv_file, delimiter=',')
sorted_list = sorted(reader, key=lambda c_row: int(c_row['project_id']), reverse=False)
statement_coverage_increase = []
checked_coverage_increase = []
mutation_coverage_increase = []
project_ids = []
project_name = ','.join(set([s['project'] for s in sorted_list]))
for row in islice(sorted_list, n[0], n[1]):
checked_coverage_increase.append(float(row['checked_coverage_increase']))
statement_coverage_increase.append(float(row['statement_coverage_increase']))
mutation_coverage_increase.append(float(row['mutation_coverage_increase']))
project_ids.append(int(row['project_id']))
ind = np.arange(len(range(n[0], n[1])))
width = 0.35
fig, ax = plt.subplots()
# rects1 = ax.bar(ind, ratio_checked_coverage, width, color='r')
# rects2 = ax.bar(ind + width, ratio_statement_coverage, width, color='y')
checked_coverage_bars = ax.bar(ind - width / 2, checked_coverage_increase, width)
statement_coverage_bars = ax.bar(ind + width / 2, statement_coverage_increase, width)
mutation_coverage_bars = ax.bar(ind + width * 3 / 2, mutation_coverage_increase, width)
# add some text for labels, title and axes ticks
ax.set_ylabel('% increase in coverage score')
ax.set_xlabel('Project bug Id')
ax.set_title('% of coverage increase for statement and checked coverage for project : ' + project_name)
ax.set_xticks(ind + width / 2)
ax.set_xticklabels(tuple(project_ids))
# formula_legend = Rectangle((0, 0), 1, 1, fc="w", fill=False, edgecolor='none', linewidth=0)
ax.legend((checked_coverage_bars[0], statement_coverage_bars[0], mutation_coverage_bars[0]),
('Checked Coverage', 'Statement Coverage', 'Mutation Coverage'))
# autolabel(ax, checked_coverage_bars)
# autolabel(ax, statement_coverage_bars)
fig.set_size_inches(15, 5)
# save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save) + '_' +
# str(max(project_ids))
save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save)
fig.savefig(save_path, dpi=100)
plt.show()
def read_big_file_and_visualise():
file_name = "/max_pc" + ".csv"
path = str(get_project_root()) + results_folder + file_name
statement_increase = []
checked_increase = []
mutation_coverage = []
with open(path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
statement_increase.append(float(row['statement_pc_max']))
checked_increase.append(float(row['checked_pc_max']))
mutation_coverage.append(float(row['mutation_pc_max']))
# mutation_increase.append(float(row['mutation_coverage_increase'])
# if float(row['mutation_coverage_increase']) > 0 else 0)
save_path = str(
get_project_root()) + results_folder + '/max_pc_violin-plot'
visualize_as_boxplot_plot_jitter(checked_increase, statement_increase,
mutation_coverage, save_path,
"Probabilistic Coupling: All Projects",
"Max PC")
def visualize_whole_project():
with open(get_configured_csv_path()) as csv_file:
file_name_to_save = os.path.basename(csv_file.name).split(".")[0]
reader = csv.DictReader(csv_file, delimiter=',')
sum_len_of_bug_detecting_tests = 0
sum_len_of_statement_covering_tests = 0
sum_len_of_checked_covering_tests = 0
project_name = set()
for row in reader:
sum_len_of_bug_detecting_tests = sum_len_of_bug_detecting_tests + \
int(row['len_of_bug_detecting_tests'])
sum_len_of_statement_covering_tests = sum_len_of_statement_covering_tests + \
int(row['len_of_statement_covering_tests'])
sum_len_of_checked_covering_tests = sum_len_of_checked_covering_tests + \
int(row['len_of_checked_covering_tests'])
project_name.add(row['project'])
sum_checked_coverage = sum_len_of_bug_detecting_tests / sum_len_of_checked_covering_tests
checked_coverage_label = [str(sum_len_of_bug_detecting_tests) + '/' + str(sum_len_of_checked_covering_tests)]
sum_statement_coverage = sum_len_of_bug_detecting_tests / sum_len_of_statement_covering_tests
statement_coverage_label = [str(sum_len_of_bug_detecting_tests) + '/' + str(sum_len_of_statement_covering_tests)]
project_name = ','.join([s for s in project_name])
ind = np.arange(1)
width = 0.35
fig, ax = plt.subplots()
checked_coverage_bars = ax.bar(ind - width / 2, sum_checked_coverage, width, color='#3d85c6')
statement_coverage_bars = ax.bar(ind + width / 2, sum_statement_coverage, width, color='#e69138')
# add some text for labels, title and axes ticks
ax.set_ylabel('Ratio')
ax.set_xlabel(project_name)
# ax.set_title('Overall result: \n Ratio between bug detecting and covering tests for project: ' + project_name)
ax.set_xticks(tuple())
ax.set_xticklabels(tuple(''))
ax.legend().remove()
autolabel1(ax, checked_coverage_bars, checked_coverage_label)
autolabel1(ax, statement_coverage_bars, statement_coverage_label)
# save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save) + '_' +
# str(max(project_ids))
save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save) + '_whole_result'
fig.savefig(save_path, dpi=100, width=4)
plt.show()
def visualize_whole_project():
with open(get_latest_csv_path()) as csv_file:
file_name_to_save = os.path.basename(csv_file.name).split(".")[0]
reader = csv.DictReader(csv_file, delimiter=',')
statement_coverage = []
checked_coverage = []
percent_coverage = []
project_name = set()
for row in reader:
statement_coverage.append(float(row['statement_coverage']))
checked_coverage.append(float(row['checked_coverage']))
percent_coverage.append(int(row['percent_coverage']))
project_name.add(row['project'])
percent_coverage = np.array(percent_coverage)
checked_coverage = np.array(checked_coverage)
percent_coverage_smooth = np.linspace(min(percent_coverage),
max(percent_coverage), 500)
f = interp1d(percent_coverage, checked_coverage, kind='quadratic')
checked_coverage_smooth = f(percent_coverage_smooth)
f1 = interp1d(percent_coverage, statement_coverage, kind='quadratic')
statement_coverage_smooth = f1(percent_coverage_smooth)
project_name = ','.join([s for s in project_name])
save_path = str(get_project_root()) + results_folder + '/' + str(
file_name_to_save) + '_whole_result'
fig = plt.gcf()
fig.set_size_inches(15, 5)
plt.plot(percent_coverage_smooth,
checked_coverage_smooth,
label="checked coverage")
plt.plot(percent_coverage_smooth,
statement_coverage_smooth,
label="statement coverage")
plt.xlabel('coverage score in percentage for project: ' + project_name)
plt.ylabel('correlation (rpb) value')
plt.title('Point Biserial correlation for statement and checked coverage')
plt.legend()
plt.show()
fig.savefig(save_path, dpi=100)
def read_file_and_visualise():
project_list = project_config.get('projects', 'project_list').split(",")
results_folder = str(get_project_root()) + results_folder_max_pc_per_proj
for project in project_list:
file_name = "/" + project + ".csv"
csv_file_path = results_folder + file_name
statement_increase = []
checked_increase = []
mutation_coverage = []
with open(csv_file_path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
statement_increase.append(float(row['statement_pc_max']))
checked_increase.append(float(row['checked_pc_max']))
# mutation_increase.append(float(row['mutation_coverage_increase'])
# if float(row['mutation_coverage_increase']) > 0 else 0)
save_path = results_folder + '/' + project
visualize_as_stackedbar(checked_increase, statement_increase,
mutation_coverage, save_path,
"Probabilistic Coupling: " + project, "Max PC")
def visualize_as_bar_plot():
with open(get_latest_csv_path()) as csv_file:
file_name_to_save = os.path.basename(csv_file.name).split(".")[0]
reader = csv.DictReader(csv_file, delimiter=',')
statement_coverage = []
checked_coverage = []
percent_coverage = []
counts_statement = {}
counts_checked = {}
counts_statement_f = {}
counts_checked_f = {}
project_name = set()
for row in reader:
statement_coverage.append(
int(1 if row['statement_coverage'] == "True" else 0))
checked_coverage.append(
int(1 if row['checked_coverage'] == "True" else 0))
percent_coverage.append(int(row['percent_coverage_s']))
project_name.add(row['project'])
try:
counts_statement[int(row['percent_coverage_s'])] = counts_statement[int(row['percent_coverage_s'])] + \
int(1 if row['statement_coverage'] == "True" else 0)
except KeyError:
counts_statement[int(row['percent_coverage_s'])] = int(
1 if row['statement_coverage'] == "True" else 0)
try:
counts_checked[int(row['percent_coverage_c'])] = counts_checked[int(row['percent_coverage_c'])] + \
int(1 if row['checked_coverage'] == "True" else 0)
except KeyError:
counts_checked[int(row['percent_coverage_c'])] = int(
1 if row['checked_coverage'] == "True" else 0)
except ValueError:
pass
try:
counts_statement_f[int(row['percent_coverage_s'])] = counts_statement_f[int(row['percent_coverage_s'])]\
+ int(1)
except KeyError:
counts_statement_f[int(row['percent_coverage_s'])] = int(1)
try:
counts_checked_f[int(row['percent_coverage_c'])] = counts_checked_f[int(row['percent_coverage_c'])] \
+ int(1)
except KeyError:
counts_checked_f[int(row['percent_coverage_c'])] = int(1)
except ValueError:
pass
label = [item for item, _ in sorted(counts_statement.items())]
statement_bar_heights = [
item / counts_statement_f[_]
for _, item in sorted(counts_statement.items())
]
checked_bar_heights = [
item / counts_checked_f[_]
for _, item in sorted(counts_checked.items())
]
statement_bar_heights_label = [
str(item) + '/' + str(counts_statement_f[_])
for _, item in sorted(counts_statement.items())
]
checked_bar_heights_label = [
str(item) + '/' + str(counts_checked_f[_])
for _, item in sorted(counts_checked.items())
]
project_name = ','.join([s for s in project_name])
ind = np.arange(len(label))
width = 0.35
fig, ax = plt.subplots()
checked_coverage_bars = ax.bar(ind - width / 2, checked_bar_heights, width)
statement_coverage_bars = ax.bar(ind + width / 2, statement_bar_heights,
width)
ax.set_ylabel('Ratio')
ax.set_xlabel('% coverage score')
ax.set_title(
'Ratio of Total No. of test suites to test suite includes bug detecting tests: Lang'
)
ax.set_xticks(ind - width / 2)
ax.set_xticklabels(tuple(label))
ax.legend((checked_coverage_bars[0], statement_coverage_bars[0]),
('Checked Coverage', 'Statement Coverage'))
# ax.legend((checked_coverage_bars[0]), ('Checked Coverage'))
autolabel(ax, checked_coverage_bars, checked_bar_heights_label)
autolabel(ax, statement_coverage_bars, statement_bar_heights_label)
fig.set_size_inches(15, 5)
# save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save) + '_' +
# str(max(project_ids))
save_path = str(
get_project_root()) + results_folder + '/' + str(file_name_to_save)
fig.savefig(save_path, dpi=100)
plt.show()
def visualize_as_box_plot():
font = {'size': 20}
plt.rc('font', **font)
project_list = project_config.get('projects', 'project_list').split(",")
if len(project_list) > 1:
print("reduce number of projects to 1")
exit(0)
project_list = project_list[0]
statement_coverage_t = []
statement_coverage_f = []
checked_coverage_t = []
checked_coverage_f = []
project_range = project_config.get('projects', project_list).split(",")
for project_id in range(int(project_range[0]), int(project_range[1]) + 1):
file_name = "/point_biserial_correlation__" + project_list + "_" + str(
project_id) + ".csv"
path = str(get_project_root()) + results_folder + file_name
try:
with open(path) as csv_file:
reader = csv.DictReader(csv_file, delimiter=',')
for row in reader:
if row['statement_coverage'] == "True":
statement_coverage_t.append(
int(row['percent_coverage_s']))
if row['checked_coverage'] == "True":
checked_coverage_t.append(
int(row['percent_coverage_c']))
if row['statement_coverage'] == "False":
statement_coverage_f.append(
int(row['percent_coverage_s']))
if row['checked_coverage'] == "False":
checked_coverage_f.append(
int(row['percent_coverage_c']))
except FileNotFoundError:
pass
data_to_plot = [
checked_coverage_f, checked_coverage_t, statement_coverage_f,
statement_coverage_t
]
print(len(checked_coverage_f), len(checked_coverage_t),
len(statement_coverage_f), len(statement_coverage_t))
fig = plt.figure(1, figsize=(9, 6))
# Create an axes instance
ax = fig.add_subplot(111)
ax.set_ylabel('No. of tests suites with % cov. score')
ax.set_xlabel(
'Indicates whether or not, a bug detecting test is included in \n generated test suite'
)
ax.set_title(project_list)
# ax.set_xticks(ind + width / 2)
ax.set_xticklabels(tuple(['False', 'True', 'False', 'True']))
# Create the boxplot
bp = ax.boxplot(data_to_plot)
bp['medians'][0].set(color='#3d85c6', linewidth=4)
bp['medians'][1].set(color='#3d85c6', linewidth=4)
bp['medians'][2].set(color='#e69138', linewidth=4)
bp['medians'][3].set(color='#e69138', linewidth=4)
# ax.legend((bp['medians'][0], bp['medians'][2]), ('Checked Coverage', 'Statement Coverage'))
ax.legend((bp['medians'][0], bp['medians'][2]),
('Checked Coverage', 'Statement Coverage')).remove()
plt.show()
save_path = str(get_project_root()) + results_folder + '/' + str(
project_list) + '_box-plot'
fig.savefig(save_path, dpi=100)
def get_correlation_file():
return str(
get_project_root()) + results_folder + '/correlation_final_lang.txt'
# Django settings for yyblog project.
import os
import os.path
import util
PROJECT_ROOT = util.get_project_root()
SITE_TITLE = "Yu Yang's Blog"
DEBUG = True
TEMPLATE_DEBUG = DEBUG
ADMINS = (
#('Your Name', '*****@*****.**'),
('yangyu', '*****@*****.**'),
)
MANAGERS = ADMINS
DATABASES = {
'default': {
'ENGINE': 'django.db.backends.sqlite3', # Add 'postgresql_psycopg2', 'mysql', 'sqlite3' or 'oracle'.
'NAME': os.path.join(PROJECT_ROOT, 'data.db'), # Or path to database file if using sqlite3.
# The following settings are not used with sqlite3:
'USER': '',
'PASSWORD': '',
'HOST': '', # Empty for localhost through domain sockets or '127.0.0.1' for localhost through TCP.
'PORT': '', # Set to empty string for default.
}
}
def visualize_statement_percent_wise():
path = str(get_project_root()) + results_folder
directory = os.path.join(path)
for root, dirs, files in os.walk(directory):
for file in files:
if file.endswith(".json"):
print(file)
contents = read_json_file(path + '/' + file)
for stmt_tests in contents[0]['tests']:
for test_suites in stmt_tests['test_suites']:
try:
if 0 <= test_suites['statement_coverage'][
'score'] <= 10:
statement_0_10.append(
len(test_suites['statement_coverage']
['tests']))
if 11 <= test_suites['statement_coverage'][
'score'] <= 20:
statement_10_20.append(
len(test_suites['statement_coverage']
['tests']))
if 21 <= test_suites['statement_coverage'][
'score'] <= 30:
statement_20_30.append(
len(test_suites['statement_coverage']
['tests']))
if 31 <= test_suites['statement_coverage'][
'score'] <= 40:
statement_30_40.append(
len(test_suites['statement_coverage']
['tests']))
if 41 <= test_suites['statement_coverage'][
'score'] <= 50:
statement_40_50.append(
len(test_suites['statement_coverage']
['tests']))
if 51 <= test_suites['statement_coverage'][
'score'] <= 60:
statement_50_60.append(
len(test_suites['statement_coverage']
['tests']))
if 61 <= test_suites['statement_coverage'][
'score'] <= 70:
statement_60_70.append(
len(test_suites['statement_coverage']
['tests']))
if 71 <= test_suites['statement_coverage'][
'score'] <= 80:
statement_70_80.append(
len(test_suites['statement_coverage']
['tests']))
if 81 <= test_suites['statement_coverage'][
'score'] <= 90:
statement_80_90.append(
len(test_suites['statement_coverage']
['tests']))
if 91 <= test_suites['statement_coverage'][
'score'] <= 100:
statement_90_100.append(
len(test_suites['statement_coverage']
['tests']))
if 0 <= test_suites['checked_coverage'][
'score'] <= 10:
checked_0_10.append(
len(test_suites['checked_coverage']
['tests']))
if 11 <= test_suites['checked_coverage'][
'score'] <= 20:
checked_10_20.append(
len(test_suites['checked_coverage']
['tests']))
if 21 <= test_suites['checked_coverage'][
'score'] <= 30:
checked_20_30.append(
len(test_suites['checked_coverage']
['tests']))
if 31 <= test_suites['checked_coverage'][
'score'] <= 40:
checked_30_40.append(
len(test_suites['checked_coverage']
['tests']))
if 41 <= test_suites['checked_coverage'][
'score'] <= 50:
checked_40_50.append(
len(test_suites['checked_coverage']
['tests']))
if 51 <= test_suites['checked_coverage'][
'score'] <= 60:
checked_50_60.append(
len(test_suites['checked_coverage']
['tests']))
if 61 <= test_suites['checked_coverage'][
'score'] <= 70:
checked_60_70.append(
len(test_suites['checked_coverage']
['tests']))
if 71 <= test_suites['checked_coverage'][
'score'] <= 80:
checked_70_80.append(
len(test_suites['checked_coverage']
['tests']))
if 81 <= test_suites['checked_coverage'][
'score'] <= 90:
checked_80_90.append(
len(test_suites['checked_coverage']
['tests']))
if 91 <= test_suites['checked_coverage'][
'score'] <= 100:
checked_90_100.append(
len(test_suites['checked_coverage']
['tests']))
except KeyError:
pass
data_to_plot = [
statement_0_10, statement_10_20, statement_20_30, statement_30_40,
statement_40_50, statement_50_60, statement_60_70, statement_70_80,
statement_80_90, statement_90_100
]
fig = plt.figure(1, figsize=(9, 6))
# Create an axes instance
ax = fig.add_subplot(111)
ax.set_ylabel('No. of tests with % coverage score')
ax.set_xlabel(
'Indicates whether or not, a bug detecting test is included in \n generated test suite'
)
# ax.set_xticks(ind + width / 2)
# ax.set_xticklabels(tuple(['False', 'True', 'False', 'True']))
# Create the boxplot
bp = ax.boxplot(data_to_plot)
# bp['medians'][0].set(color='#3d85c6', linewidth=4)
# bp['medians'][1].set(color='#3d85c6', linewidth=4)
#
# bp['medians'][2].set(color='#e69138', linewidth=4)
# bp['medians'][3].set(color='#e69138', linewidth=4)
ax.legend().remove()
plt.show()
# save_path = str(get_project_root()) + results_folder + '/' + str(project_list) + '_box-plot'
# fig.savefig(save_path, dpi=100)
data_to_plot = [
checked_0_10, checked_10_20, checked_20_30, checked_30_40,
checked_40_50, checked_50_60, checked_60_70, checked_70_80,
checked_80_90, checked_90_100
]
fig = plt.figure(1, figsize=(9, 6))
# Create an axes instance
ax = fig.add_subplot(111)
ax.set_ylabel('No. of tests with % coverage score')
ax.set_xlabel(
'Indicates whether or not, a bug detecting test is included in \n generated test suite'
)
# ax.set_xticks(ind + width / 2)
# ax.set_xticklabels(tuple(['False', 'True', 'False', 'True']))
# Create the boxplot
bp = ax.boxplot(data_to_plot)
# bp['medians'][0].set(color='#3d85c6', linewidth=4)
# bp['medians'][1].set(color='#3d85c6', linewidth=4)
#
# bp['medians'][2].set(color='#e69138', linewidth=4)
# bp['medians'][3].set(color='#e69138', linewidth=4)
ax.legend().remove()
plt.show()
def visualize_whole_project():
with open(get_configured_csv_path()) as csv_file:
file_name_to_save = os.path.basename(csv_file.name).split(".")[0]
reader = csv.DictReader(csv_file, delimiter=',')
sum_checked_buggy = 0
sum_checked_fixed = 0
sum_coverage_buggy = 0
sum_coverage_fixed = 0
sum_mutation_buggy = 0
sum_mutation_fixed = 0
project_name = set()
for row in reader:
sum_checked_buggy = sum_checked_buggy + float(row['checked_buggy'])
sum_checked_fixed = sum_checked_fixed + float(row['checked_fixed'])
sum_coverage_buggy = sum_coverage_buggy + float(row['statement_buggy'])
sum_coverage_fixed = sum_coverage_fixed + float(row['statement_fixed'])
sum_mutation_buggy = sum_mutation_buggy + float(row['mutation_buggy'])
sum_mutation_fixed = sum_mutation_fixed + float(row['mutation_fixed'])
project_name.add(row['project'])
checked_coverage_increase_per = compute_percent_increase(sum_checked_buggy, sum_checked_fixed)
statement_coverage_increase_per = compute_percent_increase(sum_coverage_buggy, sum_coverage_fixed)
mutation_coverage_increase_per = compute_percent_increase(sum_mutation_buggy, sum_mutation_fixed)
checked_coverage_increase = [sum_checked_buggy, sum_checked_fixed]
statement_coverage_increase = [sum_coverage_buggy, sum_coverage_fixed]
mutation_coverage_increase = [sum_mutation_buggy, sum_mutation_fixed]
checked_coverage_label = ['', 'inc:' + str(round(checked_coverage_increase_per, 3))]
statement_coverage_label = ['', 'inc: ' + str(round(statement_coverage_increase_per, 3))]
mutation_coverage_label = ['', 'inc: ' + str(round(mutation_coverage_increase_per, 3))]
project_name = ','.join([s for s in project_name])
ind = np.arange(2)
width = 0.3
fig, ax = plt.subplots()
checked_coverage_bars = ax.bar(ind - width / 2, checked_coverage_increase, width)
statement_coverage_bars = ax.bar(ind + width / 2, statement_coverage_increase, width)
mutation_coverage_bars = ax.bar(ind + width * 3 / 2, mutation_coverage_increase, width)
# ax.set_ylabel('sum of coverage score of all projects')
ax.set_xlabel(project_name)
# ax.set_title('Overall result: coverage increase of bug detecting tests')
ax.set_xticks(tuple(ind + width / 120))
ax.set_xticklabels(tuple(['buggy version', 'fixed version']))
# formula_legend = Rectangle((0, 0), 1, 1, fc="w", fill=False, edgecolor='none', linewidth=0)
# ax.legend((checked_coverage_bars[0], statement_coverage_bars[0], mutation_coverage_bars[0]),
# ('Checked Coverage', 'Statement Coverage', 'Mutation Coverage'), loc='lower left')
ax.legend().remove()
autolabel(ax, checked_coverage_bars, checked_coverage_label)
autolabel(ax, statement_coverage_bars, statement_coverage_label)
autolabel(ax, mutation_coverage_bars, mutation_coverage_label)
# save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save) + '_' +
# str(max(project_ids))
save_path = str(get_project_root()) + results_folder + '/' + str(file_name_to_save) + '_whole_result'
fig.savefig(save_path, dpi=100)
plt.show()
