org-log-into-drawer should behave similarly to org-clock-into-drawer

org-log-into-drawer should behave similarly to org-clock-into-drawer

[Nikolay Kudryavtsev]

Hello.

Currently when org-clock-into-drawer is set to t, it moves all old 
drawer-less clock records to a new drawer. But org-log-into-drawer does 
not do the same for log entries. I think it's logical for those two 
variables to follow the same convention.

-- 
Best Regards,
Nikolay Kudryavtsev

Re: org-log-into-drawer should behave similarly to org-clock-into-drawer

[Nicolas Goaziou]

Hello,

Nikolay Kudryavtsev <nikolay.kudryavtsev@gmail.com> writes:

> Currently when org-clock-into-drawer is set to t, it moves all old
> drawer-less clock records to a new drawer. But org-log-into-drawer
> does not do the same for log entries. I think it's logical for those
> two variables to follow the same convention.

You might be right, but two different features following two different
conventions doesn't sound terribly wrong either.

In any case, we are arguing about a default value for a defcustom. So,
unless it is proven that the current default is more complex to grasp
for beginners, I don't think we should bother much.

WDYT?


Regards,

-- 
Nicolas Goaziou

Re: org-log-into-drawer should behave similarly to org-clock-into-drawer

[Nikolay Kudryavtsev]

There are two things worth noting:
1. The original value for org-log-into-drawer is nil.
2. There seems to be no value that would enable movement of old records 
into drawers.

So assuming somebody just discovers this variable(like me), decides that 
drawers is what he wants, now he has to move all those old entries 
manually. That's not too much work, but we already have a better 
behavior example in org-clock-into-drawer.

It's still a rather minor issue, by all means...

-- 
Best Regards,
Nikolay Kudryavtsev

Re: org-log-into-drawer should behave similarly to org-clock-into-drawer

[Nicolas Goaziou]

Nikolay Kudryavtsev <nikolay.kudryavtsev@gmail.com> writes:

> So assuming somebody just discovers this variable(like me), decides
> that drawers is what he wants, now he has to move all those old
> entries manually. That's not too much work, but we already have
> a better behavior example in org-clock-into-drawer.

If default value for `org-log-into-drawer' is non-nil, state change
notes are invisible by default, which can be very confusing for
newcomers yet having to discover drawers.

This is the historical reason for SCHEDULED and DEADLINE not being
properties in the properties drawer, AFAIR.

> It's still a rather minor issue, by all means...

I agree. I understand the consistency issue, but in this particular
case, I'd rather not enforce drawers by default.

Regards,

-- 
Nicolas

Re: org-log-into-drawer should behave similarly to org-clock-into-drawer

[Nikolay Kudryavtsev]

I'm not suggesting t to be the default value. Here I totally agree with nil.

The main consistency problem is that t would not move log entries, like 
t moves them for org-clock-into-drawer.

-- 
Best Regards,
Nikolay Kudryavtsev

Re: org-log-into-drawer should behave similarly to org-clock-into-drawer

[Nicolas Goaziou]

Nikolay Kudryavtsev <nikolay.kudryavtsev@gmail.com> writes:

> The main consistency problem is that t would not move log entries,
> like t moves them for org-clock-into-drawer.

There is a big difference here.

Clocks are perfectly defined in Org syntax, so one can collect them
without trouble.

However, there is no such thing as a state change note. It is a mere
list item, even customizable, so it can be difficult to tell what it
looks like. Org doesn't even try to collect them: it is very cheesy (see
for example line 204 in org-habit.el for an attempt to find such notes).

Regards,

-- 
Nicolas

Re: org-log-into-drawer should behave similarly to org-clock-into-drawer

[Nikolay Kudryavtsev]

Oh, I never thought about it like this. Not having a consistent syntax 
is a good reason not to include code like this in the core. Thanks.

-- 
Best Regards,
Nikolay Kudryavtsev