Week 50. User code gets emptied at parsing
I located the problem: the code from the user was being deleted inside the exercise.py so, what I need to do now is finding the point where it gets emptied.
I located the problem: the code from the user was being deleted inside the exercise.py so, what I need to do now is finding the point where it gets emptied.
So, with the discovery of the previos week, I now can debug with more ease.
Even after using the invisibility_of_element_located function, or the selenium methods to wait for an element to be clickable, some errors still happened.
I found the reason for getting that timeout errors the previous week.
The main menu, where the exercises are collected, changed to react a while ago. So now that I have a basic functionality test working, it’s time to change th...
As I commented the previous week, one thing I need to keep in mind for these integration is that the RADI container must be up before trying to execute the n...
After the issue with ACE editor was solved in the previos week, my main task this week was to start implementing the test exercise by exercise.
Finally this week the problem of inserting code in the ACE editor is resolved!
While the error when inserting code in ACE editor seems close to be resolved, this week I returned to the exercise implementation.
As studied in the previous weeks, a lot of exercises share topic so now I reproduced the shell script of the previous weeks with the rest of topics.
With the fidings of the previous weeks, maybe two or three scripts per exercise were not really needed.
The objective of these test is checking wether, executing the user code, the robot changes it position or not.
The past week I was traveling, so I wasn’t able to write something for the blog.
So, it looks like there is some problem with the PYTHONPATH in my new exercise, and because of that the modules cannot be “seen” when the code from user is e...
Afther checking everything was ok with raising the exercise, it was time to test it.
Once the manager.py and the flow is understood, it was time to really start implementing the exercise.
This two weeks were dedicated to completely understanding the manager.py.
As I was implementing the exercise I made a lot of errors.
This week, we are going to start with the implementation of a image processing exercise.
I kept looking through the week to see if there is any other way of inserting text using Selenium or external functions. But there isn’t a lot of information...
As I already knew the edits is a WebElement, I looked for ways of interacting with it using Selenium and asked my co-tutor about it. It seems that there is a...
The send_keys function didn’t work, so now I have to look for another way of doing it.
Once I have the topic part, the next step is interacting with the ACE editor that the platform uses.
As I added the subprocess line to the Selenium script I noticed that it got stuck and never gived an answer unless I stopped it manually.
I already made the .sh script.
As i was trying to test the .sh script made the previous week… i got another error.
The new approach was trying to make a script at the first level of the container.
Once I had the script already working inside the container it was time to test it outside of it, in the same script as the test were being made.
Once the error of the previos week got resolved, i could keep trying to make my script work.
The problem with rospy continues, i can’t seem to find from where it comes so I need to ask for help to the rest of developers of unibotics.
First approach. Trying to make a python script that returns the position of the robot to me.
I already had a little notion of what a ros topic is, so now it’s time to start coding and finishing to understand how they really work.
I already could raise the RADI container and I knew how to use Selenium to raise a test server. So a few steps of the integration were already done.
First thing I need to do in my functionality test: Make the RADI container available. In order to do that, normally, i would type docker start RADI in my lin...
Once the frontend tests are finished and working, the next step is to start with the functionality tests for the production exercises.
Although the test code may seem simple or repetitive at first glance, there is a lot behind it. That’s why I think it’s good to make this blog post, even if ...
Once the test are made, it was time to make one file that contained them all.
This week I ended developing and evolving the frontend test por the unibotics-webserver repository.
I added automatic tests using Selenium for the frontend of unibotics. One test for each button separately.
The test for color_filter exercise are done. They are yet basic tests that check if the buttons on the page are clickable or not. There are many improvements...
This weeks the work has been centered around implementing automatic test to the unibotic platform. That way, when a deployment is done, we could automaticall...
As I continue to study the code of some files in the unibotics repository I helped solve a little issue!
Finally, I was able to solve the previous problems I faced while trying to make a D1 Deployment.
The task for this week was getting familiar with the front-end of the unibotics repository.
This week I had two main task:
This first week the task was simple: review the Python syntax so I can use this language on the rest of the TFG.
First post