[Docutils-users] title underlines, indented bulleted lists and emacs contrib repeat-last-character

[Martin B. <bl...@ir...>]

Hello!

This long email contains 3 questions, one emacs function and one comment.

Comment
=3D=3D=3D=3D=3D=3D=3D

First off, I would like to compliment you for providing this excellent Py=
thon
package.  I hope it makes it into the standard Python distribution one da=
y.  I
had started implementing a similar package but very basic in comparison, =
for
converting my travel journal input into a set of HTML pages but now I'm h=
oping
to use docutils for it.  The functionality you're providing is extremely =
useful.
Thank you!  I have already converted a few co-workers that are already st=
arting
to use it for writing short technical memos.

Now, I have a few questions about reST:


Question 1
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

The following 3 lines would generate a warning:

Some Section Title That Is Quite Long And The User Is Lazy So Writes
------------------------------

Generates the following warning:
t:25: (WARNING/2) Title underline too short.


Since you can detect that a section underline is not long enough (i.e doc=
utils
generates a specific warning), why not allow it as a valid section marker=
?  I
would imagine requiring a minimum length of, say, 20 characters would wor=
k fine.



Emacs Contribution Attached
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D

I'm lazy, so I came up with an emacs function to "fit" the title underlin=
es to
the appropriate length (see attached file), due to the previous question'=
s
problem.  The function is also a bit more general than that, so you can u=
se it
for make up separators in code, for example (it fills up to the fill-colu=
mn if
the preceding line is empty).  Please feel free to include it in the
distribution.

I personally bind it on C-M-=3D:

  (global-set-key-safe [(control meta =3D)] 'repeat-last-character)


Question 2
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

When I convert

  * a bulleted list of items like this;

  * where the first bullet character does not start on column zero;

It gets converted into a bulleted list inside an "indented block".  In HT=
ML,
this generates:

<blockquote>
<ul>
  <li> ... </li>
  <li> ... </li>
</ul>
</blockquote>

Which is not what the input intends to convey.  In fact, I can hardly ima=
gine a
case where one would like a "new" indented block of a bulleted list.

Why not make allow top-level bulleted lists to start indented?  It looks =
much
clearer in the input.  So far, this is my one and only gripe with reST.

Thanks! :-)
Cheers,