Enhancement/refactor typescript

[herniadlf]

Related issue #10

[greenlucid]

Looks good, some feedback I can give through skimming through the code:

• some .tsx files could be .ts instead since they don't use JSX syntax, such as ./src/types.tsx or ./src/constants/chains.tsx. Making them .ts is cleaner (this is just a nitpick, but technically correct)
• I think we avoid using class, this, and all those OOP features as much as possible, are they really necessary? I didn't dig much, but why is something like this preferable to something like this?:

interface Session {
    currentSessionNumber: number,
    disputeID: number
}

why are those constructors needed?

[herniadlf]

Looks good, some feedback I can give through skimming through the code:

• some .tsx files could be .ts instead since they don't use JSX syntax, such as ./src/types.tsx or ./src/constants/chains.tsx. Making them .ts is cleaner (this is just a nitpick, but technically correct)
• I think we avoid using class, this, and all those OOP features as much as possible, are they really necessary? I didn't dig much, but why is something like this preferable to something like this?:

interface Session {
    currentSessionNumber: number,
    disputeID: number
}

why are those constructors needed?

Thanks for ur feedback @greenlucid ! Both issues have to do with typescript misinterpretations. I took your feedback, did some reading and I think I've corrected it all:

• I moved to .ts those files that don't have react at all.
• I removed the classes from the types, leaving only the interfaces.

Next week i'll be testing all the changes on goerli (i think governor and kleros liquid are there) to test the appeal module. The happy path it's good.

[greenlucid]

What is the type of this amount supposed to be? BigNumber? string? Can you pass decimals here? This amount should be typed, is there any reason this wasn't the case?
Same for winner in the props of the AppealSideBox (is it a number? an address?)

[herniadlf]

@greenlucid Yep, I had doubts about some objects types.. But after I saw your comment I've made a review about all the "any" types and I think that it's okay now. lmk what you think

[greenlucid]

Seems like the bug that made withdrawTransactionList revert had been undiscovered for 3 years:
c6d454f
I guess I'm the first person that ever attempts to withdraw a list

[greenlucid]

There's an issue I haven't been able to solve. On occasion, a transparent iframe will get in the way of the app and prevent you from interacting. If you F12 and delete it from the HTML inspector, the app works normally. Got any idea why that happens? I haven't replicated the issue in live governor.kleros.io, only on your branch. I don't know how to reproduce the bug consistently. It occurred on the following circumstances:

• using firefox
• running the app with yarn start (despite it being handled with npm)

If you can't reproduce it, I could just merge and solve it if it ever becomes an issue

[herniadlf]

There's an issue I haven't been able to solve. On occasion, a transparent iframe will get in the way of the app and prevent you from interacting. If you F12 and delete it from the HTML inspector, the app works normally. Got any idea why that happens? I haven't replicated the issue in live governor.kleros.io, only on your branch. I don't know how to reproduce the bug consistently. It occurred on the following circumstances:

• using firefox
• running the app with yarn start (despite it being handled with npm)

If you can't reproduce it, I could just merge and solve it if it ever becomes an issue

I can reproduce it in master also. It's related to hot reload in local environment. I fixed it following this answer.

I've also fixed:

• index class definitions
• a warning in the poh.svg logo

[sonarqubecloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
20 Code Smells

No Coverage information
0.0% Duplication

[herniadlf]

@greenlucid with the new package-json updates something in the build is broken. If you are okay, I can try to solve the iframe issue as a separate PR, because it's not related with the typescript refactor

[greenlucid]

Let's just see what happens. Sure, then, if the iframe issue won't be noticed in production, then it's fine.