def authenticateCredentials(userName, password):
accessID = dc.encryptWithSecretKey(userName)
password = dc.encryptWithSecretKey(password)
status = isAccessIdAuthentic(accessID)
if status != False:
dbPass = status["password"]
if password == dbPass:
return 0
else:
return 2
else:
return 1
def isAccessTokenAuthentic(accessID, accessToken):
accountDetails = isAccessIdAuthentic(accessID)
if accountDetails != False:
try:
userName = accountDetails["userName"]
password = accountDetails["password"]
encodingKey = accountDetails["encodingKey"]
credentials = userName + accessID + password + encodingKey
internal_accessToken = dc.encryptWithSecretKey(credentials)
except Exception as e:
print(str(e))
return False
if internal_accessToken == accessToken:
return True
else:
print(credentials)
print(accessToken)
print(internal_accessToken)
return False
def loginPage():
loginForm = forms.LoginForm()
if request.method == 'POST':
if loginForm.validate_on_submit():
print("INFO: ML_GUI: Form Data Validated.")
userName = loginForm.userName.data
password = loginForm.password.data
status = authenticateCredentials(userName, password)
if status == 0:
print("Username/password combination validated.")
accessID = dc.encryptWithSecretKey(userName)
encodingKey = dc.generateHash(accessID)
accessToken = generateAccessToken(userName, password, accessID,
encodingKey)
workspaces_url = url_for("mlGUI_bp.showWorkSpace",
userName=userName,
accessID=accessID,
accessToken=accessToken)
#workspaces_url = "192.168.0.192:5000" + workspaces_url
#print(workspaces_url)
return redirect(workspaces_url)
elif status == 1:
print("Invalid username.")
return render_template('login.html',
form=loginForm,
status="Invalid usename," + userName +
" does not exist in database.")
elif status == 2:
print("Incorrect password.")
return render_template(
'login.html',
form=loginForm,
status="Incorrect password given for user: "******"login.html", form=loginForm)
else:
return make_response("Invalid http method", 404)
def generateAccessToken(userName, password, accessID, encodingKey):
password = dc.encryptWithSecretKey(password)
credentials = userName + accessID + password + encodingKey
accessToken = dc.encryptWithSecretKey(credentials)
print(credentials)
return accessToken
def createNewWorkflow(accessID, workFlowName, baseModelID):
# declare as global to update the original variable.
global REQUEST_ID
# generate a new id for each request.
REQUEST_ID = str(uuid.uuid4())
userWorkflows = MLC.USER_WORKFLOWS_PATH
workflowID = dc.encryptWithSecretKey(workFlowName)
# STEP-1 Create new workspace.
try:
if not os.path.exists("./" + userWorkflows + "/" + str(accessID) +
"/" + str(workflowID)):
Path("./" + userWorkflows + "/" + str(accessID) + "/" +
str(workflowID)).mkdir(parents=True, exist_ok=True)
else:
pass
except Exception as e:
print("ERROR: HyAPI_ML_INTERNAL: " + str(e))
return False
# STEP-2 Populate workspace with template
templateFile = MLC.TF_WORKSPACE_TEMPLATE
#print(templateFile)
userTfWorkflow = MLDC.getUserTF_workflow(accessID, workflowID)
#The template will be extracted to the newly created User's Tensoflow workflow directory
if extractZip(templateFile, userTfWorkflow):
print(
"INFO: HyAPI_ML_INTERNAL: New workflow directory clone successful."
)
else:
print(
"ERROR: HyAPI_ML_INTERNAL: Error cloning new workflow from template."
)
return False
#TODO: implement other models
# STEP-3 Extract selected model in the "pre-trained-model" folder
if str(baseModelID) == "01":
baseModel = MLC.MODEL_1_ZIP_PATH
baseModelName = MLC.MODEL_NAME_1
usrBaseModelPath = MLDC.getBaseModelDirectory(accessID, workflowID)
if extractZip(baseModel, usrBaseModelPath):
print(
"INFO: HyAPI_ML_INTERNAL: Pre-Trained-Model copied successfully."
)
else:
print(
"ERROR: HyAPI_ML_INTERNAL: Error copying pre-trained-model to new workflow."
)
return False
# STEP-4 create and return response to client
print("INFO: HyAPI_ML_INTERNAL: New workflow created successfuly.")
#create workflow details as a dict to add in the json db
workflowDetails = {
"workflow_Name": workFlowName,
"workflow_ID": workflowID,
"process_ID": "Never Run",
"process_Status": "Never Run",
"baseModel": baseModelName,
"dataSet": "empty",
"timeStamp": fh.getTimeStamp()
}
if MLDC.saveWorkflowDetails(accessID, workflowID, workflowDetails):
return True
else:
print(
"ERROR: HyAPI_ML_INTERNAL: Error saving workflow details to db."
)
return False
else:
print(
"ERROR: HyAPI_ML_INTERNAL: Wrong parameter for baseModel(acceptable values are 01 & 02)"
)
return False
def createNewWorkflow():
# declare as global to update the original variable.
global REQUEST_ID
# generate a new id for each request.
REQUEST_ID = str(uuid.uuid4())
accessID = getAccessId()
if fh.isAccessIdAuthentic(accessID):
#This is the folder in workspace where all the files generated by services shoudl be stored.
#userTfDir = dc.getUserTfDir(accessID)
#Get the File IDs from the request parameters
newWorkFlowDetails = request.json
#Get the data from the json included in the request.
try:
workFlowName = newWorkFlowDetails["WorkflowName"]
try:
#TODO: Provision for multiple base models
#baseModel = newWorkFlowDetails["BaseModel"]
#baseModelID = baseModel
baseModel = "01"
baseModelID = "01"
# 01,02,03,04
except Exception as e:
print("ERROR: HyAPI_ML: " + str(e))
baseModel = "x"
except Exception as e:
print("ERROR: HyAPI_ML: " + str(e))
workFlowName = "x"
#######################################################
if baseModel != "x" and workFlowName != "x":
userWorkflows = MLC.USER_WORKFLOWS_PATH
workflowID = dc.encryptWithSecretKey(workFlowName)
# STEP-1 Create new workspace.
try:
if not os.path.exists("./" + userWorkflows + "/" +
str(accessID) + "/" + str(workflowID)):
Path("./" + userWorkflows + "/" + str(accessID) + "/" +
str(workflowID)).mkdir(parents=True, exist_ok=True)
else:
pass
except Exception as e:
print(str(e))
status = "Error creating new workflow directory."
print("ERROR: HyAPI_ML: " + status)
httpCode = MLC.HTTP_SERVER_ERROR
res = createReqDetails_newWorkflow(workFlowName, workflowID,
baseModelID, REQUEST_ID,
status, httpCode)
return make_response(res, httpCode)
# STEP-2 Populate workspace with template
templateFile = MLC.TF_WORKSPACE_TEMPLATE
#print(templateFile)
userTfWorkflow = MLDC.getUserTF_workflow(accessID, workflowID)
#The template will be extracted to the newly created User's Tensoflow workflow directory
if extractZip(templateFile, userTfWorkflow):
print(
"INFO: HyAPI_ML: New workflow directory clone successful.")
else:
status = "Error cloning new workflow from template."
print("ERROR: HyAPI_ML: " + status)
httpCode = MLC.HTTP_SERVER_ERROR
res = createReqDetails_newWorkflow(workFlowName, workflowID,
baseModelID, REQUEST_ID,
status, httpCode)
return make_response(res, httpCode)
# STEP-3 Extract selected model in the "pre-trained-model" folder
if str(baseModel) == "01":
baseModel = MLC.MODEL_1_ZIP_PATH
baseModelName = MLC.MODEL_NAME_1
usrBaseModelPath = MLDC.getBaseModelDirectory(
accessID, workflowID)
if extractZip(baseModel, usrBaseModelPath):
print(usrBaseModelPath)
print(
"INFO: HyAPI_ML: Pre-Trained-Model copied successfully."
)
else:
status = "Error copying pre-trained-model to new workflow."
print("ERROR: HyAPI_ML: " + status)
httpCode = MLC.HTTP_SERVER_ERROR
res = createReqDetails_newWorkflow(workFlowName,
workflowID, baseModelID,
REQUEST_ID, status,
httpCode)
return make_response(res, httpCode)
# STEP-4 create and return response to client
status = "New workflow created successfuly."
print("INFO: HyAPI_ML: " + status)
httpCode = MLC.HTTP_CREATED
#create workflow details as a dict to add in the json db
workflowDetails = {
"workflow_Name": workFlowName,
"workflow_ID": workflowID,
"process_ID": "Never Run",
"process_Status": "Never Run",
"baseModel": baseModelName,
"timeStamp": fh.getTimeStamp()
}
if MLDC.saveWorkflowDetails(accessID, workflowID,
workflowDetails):
res = createReqDetails_newWorkflow(workFlowName,
workflowID, baseModelID,
REQUEST_ID, status,
httpCode)
return make_response(res, httpCode)
else:
status = "Error saving workflow details to db."
print("ERROR: HyAPI_ML: " + status)
httpCode = MLC.HTTP_SERVER_ERROR
res = createReqDetails_newWorkflow(workFlowName,
workflowID, baseModelID,
REQUEST_ID, status,
httpCode)
return make_response(res, httpCode)
else:
status = "Wrong parameter for baseModel(acceptable values are 01 & 02)"
print("ERROR: HyAPI_ML: " + status)
httpCode = MLC.HTTP_BAD_REQUEST
res = createReqDetails_newWorkflow(workFlowName, workflowID,
baseModelID, REQUEST_ID,
status, httpCode)
return make_response("Error", 400)
else:
status = "WorkflowName & BaseModel values are required in input json. One or both are missing in the request."
print("ERROR: HyAPI_ML: " + status)
httpCode = MLC.HTTP_BAD_REQUEST
res = createReqDetails_newWorkflow("N/A", "N/A", "N/A", REQUEST_ID,
status, httpCode)
return make_response(res, httpCode)