allTokens = []
weakTokens = []
tokenDatabase = []
insecureForms = []
print(' %s Phase: Crawling %s[%s1/6%s]%s' %
(lightning, green, end, green, end))
dataset = photon(target, headers, level, threadCount)
allForms = dataset[0]
print('\r%s Crawled %i URL(s) and found %i form(s).%-10s' %
(info, dataset[1], len(allForms), ' '))
print(' %s Phase: Evaluating %s[%s2/6%s]%s' %
(lightning, green, end, green, end))
evaluate(allForms, weakTokens, tokenDatabase, allTokens, insecureForms)
if weakTokens:
print('%s Weak token(s) found' % good)
for weakToken in weakTokens:
url = list(weakToken.keys())[0]
token = list(weakToken.values())[0]
print('%s %s %s' % (info, url, token))
if insecureForms:
print('%s Insecure form(s) found' % good)
for insecureForm in insecureForms:
url = list(insecureForm.keys())[0]
action = list(insecureForm.values())[0]['action']
form = action.replace(target, '')
if form:
def solve_problem(image, features, topic):
"""
------------------------------------------------------------
STEP 2. Predict the Feature
1. Crop the features as collection of images
2. Feed feature images to the model
3. Get predicted value for each feature
------------------------------------------------------------
"""
# Get all features as an image
feature_col = get_features_as_img(image, features)
feature_col = np.array(feature_col)
# Formatting the input
feature_col = feature_col.reshape(feature_col.shape + (1, ))
feature_col = feature_col.astype(np.float32)
input_shape = list(feature_col.shape)
# Load the TFLite model and allocate tensors
interpreter = tf.lite.Interpreter(model_path="core/model/" +
MODEL_ARCHITECTURE)
interpreter.resize_tensor_input(0, input_shape, strict=True)
interpreter.allocate_tensors()
# Get input and output tensors
input_details = interpreter.get_input_details()
output_details = interpreter.get_output_details()
# Feed the data into the model
interpreter.set_tensor(input_details[0]['index'], feature_col)
interpreter.invoke()
# Get predicted value
features_prob = interpreter.get_tensor(output_details[0]['index'])
features_pred = []
for index in range(len(feature_col)):
predicted_feat = np.argmax(features_prob[index])
features_pred.append(predicted_feat)
# Assign the predicted value to each component
for index in range(len(features)):
features[index]["symbol_id"] = features_pred[index].item()
features[index]["symbol"] = math_symbol[features_pred[index]]["symbol"]
# Display predicted features from model in terminal
if DISPLAY_PREDICTED_FEATURES:
print("--------------------------------------------------------------")
print("Predicted Features :")
print(features_pred)
print("--------------------------------------------------------------")
"""
------------------------------------------------------------
STEP 3. Turn Features Into Mathematical Terms
1. Get the position order of each features
2. Turn those features into mathematical terms
------------------------------------------------------------
"""
# Get position order for each feature
features = arrange_position(features)
# Turn features into mathematical terms
terms = get_math_terms(features, topic)
display_terms = terms.copy()
"""
------------------------------------------------------------
STEP 4. Evaluate the Problem
------------------------------------------------------------
"""
# Choose the topic and evaluate the problem
result = evaluate(terms, topic)
"""
------------------------------------------------------------
STEP 5. Display the Question and Answer
------------------------------------------------------------
"""
# Display the question and answer
qna_dict = display_qna(display_terms, result, topic)
return qna_dict