Persistent named sessions

[mizlan]

Is your feature request related to a problem? Please describe.

Right now, you can either have a persistent default @ session (-S), or non-persistent named sessions (-s <name>)

Describe the solution you'd like

I want the ability to persist named sessions. This would help where nnn is embedded in text editors, for example.

Describe alternatives you've considered

I have not considered alternatives to having persistent named sessions as a concept, but there are definitely some options to be weighed when considering how to implement this (details linked in issue below).

Additional context

Please view mcchrish/nnn.vim#43 (comment) to understand the context of this request and the proposed solutions.

Consider contributing

I am more than eager to write this patch, but I need some pointers on how to proceed.

[mizlan]

Here's the deal. Currently, when both -S flag and -s <name> flag are passed, only the later one has any effect. This is because the value of session is overwritten during option parsing:

nnn/src/nnn.c

Lines 7766 to 7772 in 36e1544

	case 's':
	if (env_opts_id < 0)
	session = optarg;
	break;
	case 'S':
	session = "@";
	break;

Also good to note is that nnn only checks if the session should be persistent when the program exits:

nnn/src/nnn.c

Lines 7189 to 7190 in 36e1544

	if (session && *session == '@' && !session[1])
	save_session(TRUE, NULL);

As a result, if we wanted to, we could have a named persistent session in the form of passing both flags in either order: nnn -S -s <name>.

[jarun]

I am already working on it. In the same line so we are good.

[mizlan]

The behavior isn't exactly as desired. There is still an error message when I try to create a session that is nonexistent (I would instead expect for it to be created).

[jarun]

It is created. The msg comes the first time you load it because it's non-existent.

[jarun]

Maybe we can get rid of it now that we know we will create it if it doesn't exist.

[mizlan]

Yeah, I was referring to the error message. I agree with removing it since -s <name> will always succeed, but its behavior may differ.

[jarun]

On second thoughts, I think we should have it to indicate that the session file doesn't exist.
It would be correct to indicate in cases where, say user had created the session earlier but somehow it was removed. He may want to be notified in that case. And it's also useful for us to debug later.

I am not sure why it's confusing. When you are creating the session for the first time, it doesn't exist and that's what the msg says. If you use option -S it will be created when you quit the program as documented.

[mizlan]

I see where you are coming from but I think there should be some method of disabling that, especially if the user is not directly controlling the session as in the case with nnn.vim. It would be odd for the user to see "failed!" or other messages when they expect it to be more discreet.

[Kabouik]

How about rephrasing the error so that it looks like a simple warning/feedback about the newly created session?

[mizlan]

Sure. My main concern is mostly the 1-second delay that occurs with the error message. So if that could be removed that would be much better already.

[jarun]

My main concern is mostly the 1-second delay that occurs with the error message.

It's necessary information.

How about rephrasing the error so that it looks like a simple warning/feedback about the newly created session?

I think it's just fine for a 1-time message. Let's not overthink. We can document it if we want. I do not want 3 people to spend time discussing it.

[mizlan]

I'll let you try it out after I'm done with the nnn.vim patch. I personally think its annoying, since a nice part of nnn is its fast startup time.

[jarun]

OK.

[jarun]

Closed at 54d760b.