Michał Muskała
Michał Muskała
I'll look into properly handling indentation myself Thank you both for working on this! ❤️
@maennchen I noticed you implemented conversion of single-line tripple-quoted strings into a regular string. I'm not sure we should actually do that - in general we try to preserve the...
Support for indentation was implemented in https://github.com/WhatsApp/erlfmt/pull/362 and I cut a 1.5.0 release with support for the new OTP 27 features
This is somewhat expected - erlfmt generally expects macro definitions and arguments to be valid Erlang AST, with some extensions - this is not true of a stand-alone `A:8` expression....
I think there's another way to solve the particular scenario at hand that should be considered. In particular, the reason for the process to start trapping exits and delaying shutdowns...
> Do you have a suggestion to how that mechanism would look like? Today you either terminate on the exit signal if trap exit is disabled or react on this...
I believe detaching "mailboxes" from processes and making them a firs-class object would be generally a beneficial idea - in particular it would make implementing a pool of workers trivial...
Because the derived or compile-time encoded keys need to be the same across all the `escape` options since they won't be re-escaped during encoding. In particular for `escape: :html_safe` the...
Could it be that fun2ms translates the map update syntax into `map_put` calls?
The string encoding should be as defined in the RFC if you use the `escape: :unicode_safe` option. The 3rd point would likely be best handled by a pre-pass that encodes...