Project ID: 8c1db389-9532-4ead-9ee9-2f4b19a5334c

if you deployed the template then you arent running your code, you would be running the template's code

Hmm, but I had to connect my repo for it. How can I get this solved to let it run my code?

show me a screenshot of the source in your service settings please

Ah, it is connected to the cloned repo:

Is it enough to just change the Repo here?

is that source repo correct?

Yep, it is the cloned one railway created where it uploaded the template code.

it is a private repo though (ATM)

right but if its the template code, then its not your code

so disconnect both the source and upsteam then connect the correct repo that contains your code

And in order to get it started do I have to create a requirements.txt as well as a Procfile too?

yes otherwise railway isnt going to know what to install or run

and is the root directory the root of the repo or where the .py files are in (Since it says "where we should look for your code")?

yes thats correct, you likely would not need to touch that

Can I also create the requirements.txt out of railway? Did not yet use one when being on replit.

the requirements.txt would need to exist within your github repo

I know. Just trying to get it created since so far was developing in the cloud. I will check back with replit to get it created there and then upload it into the repo.

Sry for asking all that nearly OT stuff: How are railway deployments working in general and in particular? Is there any documentation or good readup for this by chance? Would like to know a little bit about what is going on behind the curtain of all that magic. 🙂

saw that

sry

will repost:

Sry for asking all that nearly OT stuff: How are railway deployments working in general and in particular? Is there any documentation or good readup for this by chance? Would like to know a little bit about what is going on behind the curtain of all that magic. 🙂

heres a high level overview -
but if you have any specfic questions i can do my best to answer them

Thank you very much. Maybe I will get back to this.

Unfortunately the freeze on replit did not work entirely. The dotenv package is not listed in my requirements.txt nor is the os package. Now my bot fails with ModuleNotFoundError: No module named 'dotenv'
Is there any default I can use or any other way to get this one solved?

dont use dotenv, its not needed on railway since the service variables are injected directly into the container

and the os package is stdlib so that shouldn't be in the requirements.txt file anyway

oh IC. So accessing the variables via os.getenv("TOKEN") would be sufficient without importing anything?

you would still need to import os, but you would not need dotenv

sry next question already: I have a JSON data file in my data directory which is in parallel hierarchy to my .py file inside the src directory. But it cannot be found during loading.
So this is how the directory structure looks like (screenshot). And this is how I try to access the JSON file within my .py file (2nd screenshot).
But it can't be found. So in which directory is the running program looking for it? I.e. in which directory are we in when the application is started?

In the root since in Procfile it says: web: python src/cardBot.py?

you have not touched the root directory in the service settings right?

can you share your repo so i can take a closer look?

sure - I can make it public maybe…. So here you are: https://github.com/RrTe/RedemptionCardBot

well you see, there is no json file in the data folder

Huh? I uploaded it. Just a sec ….

Ouf! I forgot to commit the upload! Douh!

Ok, so at least ../data/carddata.json can still be not found. Will try now with: data/carddata.json

That seems to work - but now I am getting an error I never got before:

File "src/cardBot.py", line 51, in ) -> list[app_commands.Choice[str]]: TypeError: 'type' object is not subscriptable

No clue whats wrong here. That code ran fine so far..

what version of python is replit running your code with?

3.10.13

So do I have to omit the -> part?

no, let's try getting railway to run your code with python 3.10 since 3.8 is the default

IC

add a runtime.txt file with 3.10 in it

in root directory?

Finally I got it done! Great! Thank you so much! The bot is up and running now.
So if I understand correctly it will be up and running there 24/7 now?

24/7 until you run out of credits, then you will need to upgrade to the hobby plan

what happens if I run out of credits? Will it be shut down or slowed down?

shut down

😦 too bad. But thank you again for your great help! Awesome!

why too bad?

is the hobby plan not something you would be upgrading to when the time comes?

Need to check it out.