Some .rdf files are XML, but the .rdf extension seems to be used
also for other things.  So... use the 'header' mechanism.

(Hadn't seen that one because of my additional bindings.)

Add missing spaces, remove excess spaces, and
replace groups of indentation spaces with tabs.

When not in a UTF-8 locale, each character is just a single byte.

Instead of always stepping back four bytes and then tentatively
moving forward again (which is wasteful when most codes are just
one or two bytes long), inspect the preceding bytes one by one
and begin the move forward at the first valid starter byte.

This reduces the backwards searching time by close to 40 percent.

This reduces the backwards searching time by a good 20 percent.

Also when toggling the help lines back on or the extra editing space
back off, keep the cursor near the bottom of the screen if it /was/
there, instead of centering it -- if smooth scrolling is on, that is.

This supplements the fix for https://savannah.gnu.org/bugs/?50933.

When the editing space gets narrower, either by toggling line numbers on
or making nano's window smaller, and when the cursor is near the bottom,
it can be pushed offscreen.  In that case, don't get it back onscreen by
centering the current line but by putting it on the bottom row.  This
makes for a smoother experience.

This fixes http://savannah.gnu.org/bugs/?50933.

And hide the cursor again as soon as the user scrolls.
Achieve this through making the 'didfind' variable global.

Also, remove a superfluous call of wnoutrefresh(), as bottombars()
already does that.

This fixes https://savannah.gnu.org/bugs/?50918

.
Reported-by: David Lawrence Ramsey <pooka109@gmail.com>

Also, trim some comments and avoid an unused-variable warning.

Also, reshuffle two initializations to go sit with the other ones.

The bottomwin has just been refreshed by update_the_statusbar(),
and it should not be the responsibility of the prompt routines
to keep the edit window up to date, as they don't write anything
there.

Things have morphed over time and display_buffer() no longer actually
displays the buffer -- it just displays the title bar, precalculates
the multiline color info, and schedules a refresh of the edit window.

Also, avoid an unused-variable warning when configured with
--enable-tiny --enable-help --enable-multibuffer.

Searching for an empty string should be impossible, it should never
happen, but it is bit too hard to verify this at the moment.

If the length of the haystack is smaller than the length of the needle,
this means that also the length of the tail will be smaller -- because
pointer will be bigger than or equal to haystack -- so the pointer gets
readjusted to be a needle length before the end of the haystack, which
means that it ends up /before/ the haystack: thus the while loop will
never run.

On average, this saves some 200 nanoseconds per line.

Because it is slightly faster.

It is not an index, it is not an offset from anything,
it is a direct pointer.

And because I think it looks slightly better.

This avoids https://savannah.gnu.org/bugs/?49912 while at the same time
avoiding to draw the edit window twice in a row -- the first drawing
would use a wrong margin, which results in a visible and irritating
shift left or right of the content upon the second drawing.

This fixes https://savannah.gnu.org/bugs/?50877.

We should measure the length of each /translated/ shortcut description,
not of the original English one.

This fixes https://savannah.gnu.org/bugs/?50899.

This addresses https://savannah.gnu.org/bugs/?50897

.
Reported-by: Hannu Nyman <hannu.nyman@iki.fi>

This elides a counter and a comparison from the central loop,
and thus makes the search a tiny bit faster.

And one-letter variables I cannot "see" -- they are too small.