Last 12 weeks · 0 commits
1 of 6 standards met
I scetched some tests that follow pattern that I use to define auto-binded method in ES6 classes. I tried to highlight the problem I have encountered using previous version of React Hot Loader. The tests could be wrong, I just wanted to add some illustration to my feature request. It would be awesome if the new version of Hot Reloader could handle this case.
Repository: gaearon/react-proxy. Description: Proxies React components without unmounting or losing their state Stars: 456, Forks: 45. Primary language: JavaScript. Languages: JavaScript (100%). Latest release: v2.0.8 (9y ago). Open PRs: 9, open issues: 16. Last activity: 5y ago. Community health: 28%. Top contributors: gaearon, akorchev, Restuta, BenoitZugmeyer, colinmeinke, technicallyfeasible, knpwrs, nhunzaker, wkwiatek, Yomguithereal and others.
Please excuse a somewhat vague report as I’m just investigating a rather involved build system. Our codebase has (a) a custom component base class, which is generally discouraged; and (b) HOCs that use the new functionality. This lead to a very strange breakage when I tried to migrate some components that currently use our internal to instead extend (or ) directly: our React Native app crashed in react-proxy’s update() function with the error . It appears that when extending the React classes, the component ends up being not a class, but instead an object with the shape (per Chrome debugger) It’s not clear to me why this does not happen with our custom base class: I suppose treats it differently, being a full-blown ES6 class as opposed to what React provides. It’s also not clear to me exactly where I should be filing this bug: , , , …? Any thoughts welcomed.
should fix https://github.com/gaearon/react-hot-loader/issues/329 in react-hot-loader@next @gaearon can you take a look?