Sam Verschueren

Expose lower-level constructs

20

Hi everyone 👋 ! I was wondering if it would make sense to expose lower-level constructs like the managers and the models. That would give users of `isomorphic-git` more flexibility...

Support refspecs argument in fetch command

2

Hi everyone 👋 ! I'm looking into how the `git fetch` command works, and comparing that with the `fetch` command in `isomorphic-git`. It's a complex beast so I'm far from...

Only show error stack if `--verbose` flag is set

20

We discussed this briefly in a commit in `dev-time-cli`. At the moment, when an error occurs and it is not handled by the developer, the entire error stack is shown....

SVG support

21

Is SVG support planned or was it left out on purpose? Just wondering as I'm looking into dropping `node-gm` in favour of a library that does not depend on external...

Rule proposal: No assignment to global object

16

Recently someone did a PR where he put something on the `global` object. Something like ```js global.foo = 'bar'; ``` This doesn't feel like a good thing to do. You...

`extension.fixImport` shows error

1

I binded `extension.fixImport` to a keyboard shortcut, but every time I press it, it shows me this error.

Bragg 1 supported passing in `params` from a request. This is not yet possible anymore with `bragg@2`. Should be fixed.

In order to upgrade from bragg 1 to bragg 2, it would be nice if bragg 1 understood requests from `[email protected]` for instance. That way interaction between multiple Lambda functions...

It might be better to support the catch-all so we can ditch the mapping template. https://forums.aws.amazon.com/ann.jspa?annID=4040

At this moment, the errors are always logged and the stack trace is logged if `error.expose` is `false`. https://github.com/SamVerschueren/bragg/blob/master/context.js#L28-L32 Instead, we should use the `EventEmitter` and let the user subscribe...