def update_any_entry():
# print("X", selected_row_data[0], name.get(), blood_group.get(), city.get(), contact.get())
database.update_table(selected_row_data[0], name.get(), blood_group.get(),
city.get(), contact.get())
database_list.delete(0, tk.END)
for row in database.show():
database_list.insert(tk.END, row)
def parse(url):
""" Parses the data fetched from fetch_url
If its not empty, frequency, POS, sentiment of the text are
returned back to front-end.
"""
soup = fetch_url(url)
result = {}
if soup:
text = text_from_html(soup)
text = re.sub(' +', ' ', text)
result_list = pos_text(text)
#print "result_list", result_list
word_tuple = []
for x in result_list:
res = hash_word(x)
word_tuple.append(res)
print "word_tuple", word_tuple
url_hashed = hash_word(quote(url))
sent = get_intent(url)
update_table((url_hashed, sent), "string")
for y in word_tuple:
update_table(y, "tuple")
for (x, y) in result_list:
result[x] = y
return result
else:
print "Alert the url could not be found"
return ""
def submit():
if request.method == 'POST':
data = request.form.to_dict()
# write_to_csv(data)
update_table(data)
return redirect('/thankyou.html')
else:
return 'Something Went Wrong!!!'
def differenceoftimes(self, data, suffix=''):
self.sum_time = (time.time() - self.time_dict["start_time"])
database.update_table(self.xmodule_runtime.user_id,
self.xmodule_runtime.xqueue['default_queuename'],
"web_count", self.sum_time)
# table = PrettyTable(['ID', 'name', 'club', 'wins', 'defeats', 'hits_got', 'hits_given'])
# for fencer in index:
# table.add_row([fencer.ID,fencer.name,fencer.club,fencer.wins,fencer.defeats,fencer.hits_got,fencer.hits_given])
# print(table)
#создаём базу
create_db()
#вбрасываем первых ребят
first_update(index)
#вызов функции должен быть перенесен в main.py
matching(index)
#прописываем подравшихся драчунов
update_table(update)
#дёргаем текущую версию базы
for x in ask_table():
print(x)
#делаем рейтинг
dwarfing(index)
#смотрим табличку
table = PrettyTable(
['ID', 'name', 'club', 'wins', 'defeats', 'hits_got - hits_given'])
for fencer in index:
table.add_row([
fencer.ID, fencer.name, fencer.club, fencer.wins, fencer.defeats,
int(fencer.hits_got) - int(fencer.hits_given)
def detect(self, path, student_name, course_name, block_id):
image_string = self.decode_json(path, course_name, block_id,
student_name)
frame = self.decode_image(image_string)
log.info("Line 2")
# Load the Haar-Cascade and
# retrieve Cascade to detect eyes from the same
eye_cascade_path = "/edx/app/edxapp/venvs/proctor/lib/python2.7/site-packages/suspicious_images/haarcascade_eye.xml"
eye_cascade = cv2.CascadeClassifier(eye_cascade_path)
# A boolean variable which stores if the person was cheating in the frame
cheat_bool = 0
# reduce the noise in image by blurring and the blurred image to
# original image so as to subtract local mean color image.
blurred = cv2.GaussianBlur(frame, (21, 21), 0)
weight_frame = cv2.addWeighted(frame, 1.75, blurred, -0.5, 0)
# Equalise histogram to improve the contrast (Removing glare)
# - Need to convert it to YUV image
yuv = cv2.cvtColor(frame, cv2.COLOR_BGR2YUV)
yuv[:, :, 0] = cv2.equalizeHist(yuv[:, :, 0])
# reconvert the image to grayscale to be feeded to Clahe
bgr_frame = cv2.cvtColor(yuv, cv2.COLOR_YUV2BGR)
gray = cv2.cvtColor(bgr_frame, cv2.COLOR_BGR2GRAY)
# again apply Contrast Limited Adaptive Histogram Equalisation
# can only work on grayscale images
clahe = cv2.createCLAHE(clipLimit=40.0, tileGridSize=(8, 8))
gray = clahe.apply(gray)
# load the haar-cascade to get the region of interest - eye_region
eyes = eye_cascade.detectMultiScale(gray,
scaleFactor=1.1,
minNeighbors=3)
# How many times to load the haar-cascade to increase effectiveness
sample_tries = 2
while sample_tries:
for (x, y, width, height) in eyes:
cv2.rectangle(frame, (x, y), (x + width, y + height),
(0, 255, 0), 2)
break
sample_tries -= 1
# if already eyes are recongnized - break out of loop
if len(eyes):
break
# If no eye_region is found, that would mean that either the person's face is not
# straightly oriented in front of webcam or he is cheating - In any case mark as suspicious
if len(eyes) == 0:
cheat_bool = 1
else:
# crop the right portion of the eyes to enable accurate detection of pupil,
# blur that region again to reduce noise.
gray_cropped_right = gray[eyes[0][1]:eyes[0][1] + eyes[0][3],
eyes[0][0] + eyes[0][2] / 2:eyes[0][0] +
eyes[0][2]]
cv2.GaussianBlur(gray_cropped_right, (3, 3), 16)
# use HoughCircles method to detect the circular shape in right part of eye (cropped earlier)
circles_right = cv2.HoughCircles(gray_cropped_right,
cv2.cv.CV_HOUGH_GRADIENT,
dp=5,
minDist=60,
param1=30,
param2=10,
minRadius=5,
maxRadius=20)
try:
# find the pupil-positions corresponding to the circle region
pupil_position = np.uint16(np.around(circles_right))
# form a small circle around the pupil-position
for (a, b, radius) in pupil_position[0]:
cv2.circle(frame, (a + x + width / 2, b + y), 2,
(255, 0, 0), 2)
except:
pass
# If the user is found to be cheating, the corresponding images can be saved in the directory
if cheat_bool == 1:
# TODO Increament this variable in database using some course_name/student_name
student_id = database.get_id(student_name)
database.update_table(student_id, course_name, "img_count", 1)
path_img = self.get_path(path, course_name, block_id,
student_name) + "/susp_img"
if not os.path.exists(path_img):
os.makedirs(path_img)
path_img += "/"
cv2.imwrite(path_img + str(len(os.listdir(path_img))) + ".jpg",
frame)