Offer a version of async-start as a function

[Malabarba]

As part of adding optional asynchronous operations to Paradox, I check whether async is installed and ask the user to install it.

The problem is: since async was installed after Paradox, the macro async-start doesn't get expanded on byte-compile, and later leads to an error (byte-compiled code can't call macros).

I've found a hacky solution (quoting the macro call and calling eval on it), but it would be nice to have a proper one. From what I understand of the async-start code, there's no reason for it to be a macro, it could work just as well as a function.

Would it be possible to offer it (or offer an alternative) as a function?

[Malabarba]

The issue, for reference :
Malabarba/paradox#28

(sorry about the accidental close and reopen)

[jwiegley]

I have no problem with that at all, especially if it doesn't need to be a macro. Would you like to make the change?

[thierryvolpiatto]

John Wiegley notifications@github.com writes:

I have no problem with that at all, especially if it doesn't need to
be a macro. Would you like to make the change?

Please be sure to not break existing code if you do that.
IOW test your changes in a separate branch before commiting, in
particular check if async-bytecomp.el works fine, otherwise I will get
tons of bugreports now we use async as dependency for helm installation
from melpa.

Thanks.

Thierry
Get my Gnupg key:
gpg --keyserver pgp.mit.edu --recv-keys 59F29997

[jwiegley]

Yes, very true @thierryvolpiatto.

[Malabarba]

I'll give it a shot next time I have a moment. I can implement it as a new function, no need to change the current one and risk breaking stuff. :)

[thierryvolpiatto]

Artur Malabarba notifications@github.com writes:

I'll give it a shot next time I have a moment. I can implement it as a
new function, no need to change the current one and risk breaking
stuff. :)

Thanks.

Thierry
Get my Gnupg key:
gpg --keyserver pgp.mit.edu --recv-keys 59F29997