emallson/narrow-reindent.el

Defines a minor mode `narrow-reindent-mode’ that reindents the region narrowed to. The region is then indented again after widening the buffer.

narrow-reindent.el

Defines a minor mode `narrow-reindent-mode’ that reindents the region narrowed to. The region is then indented again after widening the buffer.

Known Issues

• The new indentation is saved if the narrowed buffer is saved. (Issue #1)

LICENSE

This code is © 2015 J David Smith and is released to the public under the terms of the GPLv3 license.

narrow-reindent.el's Issues

`evil-backward-char` triggers `narrow-reindent--after-narrow` with buggy contents

Problem: eval-backward-char calls evil-narrow-to-line, which calls narrow-to-region, which in turn triggers the advice with bad parameters.

Reproduction: Install narrow-reindent and evil-mode. Turn both on at once.

Solution: Unknown at this time.

Saving while narrowed saves reindentation

Desired Behavior: Saving should store the file with correct indentation.

Actual Behavior: The file is stored with the narrowed region left-aligned.

Planned Fix: Add a before-save/after-save hook of some sort.

Applying the same idea to Org-mode

I am trying to turn my Org-mode into Workflowy-like experience. An interesting thing in Workflowy is that it allows for indefinite depth of headings, where you frequently zoom-in to subtrees. In Org-mode this is inelegant.

Suppose I have a tree:

* a
** aa
*** aaa
**** aaaa
*** aab
** ab
** ac
* b
** ba

If I want to narrow to entry called aaa, I get:

*** aaa
**** aaaa

Instead I want to get is:

* aaa
** aaaa

Imagine what would happen if I have an entry 10 levels deep. Narrowing is supposed to be "concentrating on a local context, ignoring all outside of it", so why do I retain information that it's 10 levels deep? And also it's just noise.

So I'm thinking, narrow-indent could be generalized to include Org-mode too, where it removes stars before narrowing and puts them back before widening.

I need this feature, so I'll possibly try to implement it soon independently, although can't promise anything. But I would like to discuss it here. Also, maybe this could develop into some smart-narrow concept further.