Hello,
I deeply apologize for the multiple emails.
Apparently, you can author a filter in Thunderbird which will take your partially written message and cause it to be sent (!!!) instead of stored in drafts....
-----
Anyways, he comes from slimv on vim where you can apparently eval the paragraph of code I had written previously and the repl would go into the package. He was curious of he could do it in SLIME too.
Thank you!
-pete
On 02/06/2016 10:36 AM, Peter Keller wrote:
Hello,
I deeply apologize for the multiple emails.
Apparently, you can author a filter in Thunderbird which will take your partially written message and cause it to be sent (!!!) instead of stored in drafts....
Anyways, he comes from slimv on vim where you can apparently eval the paragraph of code I had written previously and the repl would go into the package. He was curious of he could do it in SLIME too.
It dawns upon me that Thunderbird had shown me potentially an illusory view of what emails it claimed it received.
So, in case no multiple emails were sent, he has a file that looks like:
(ql:quickload #:foo) (in-package #:foo)
(asdf:load-system #:bar) (in-package #:bar)
and so on.
The systems might or might not be related and he switches between them while he works. So it was a convenience of his to put his slimv cursor on the paragraph, eval the paragraph, then the repl would be in the right package automatically.
Thank you
-pete
So, in case no multiple emails were sent, he has a file that looks like:
(ql:quickload #:foo) (in-package #:foo)
(asdf:load-system #:bar) (in-package #:bar)
and so on.
Hi!
I have no solution for you, just some ideas.
There is slime-enable-evaluate-in-emacs variable in SLIME. Docstring states "*If non-nil, the inferior Lisp can evaluate arbitrary forms in Emacs. The default is nil, as this feature can be a security risk".
Also (swank-repl::repl-eval "(in-package :cl-user)") _might_ help you. I don't know if you need to enable this variable.
Look for "current lisp package" words in slime.el and read the comment there.
It seems that SLIME already tries to deduce current package for the point in the file from nearest previous in-package form. Simple experimenting should show. I'm not a EMACS user, so I can't help.