Re: [Vim-latex-devel] Re: [Vim-latex-cvs] vimfiles/plugin imaps.vim,1.7,1.8

[Benji F. <be...@me...>]

Mikolaj Machowski wrote:
> On Mon, Dec 09, 2002 at 01:30:44AM +0100, Luc Hermitte wrote:
>=20
>>* On Sun, Dec 08, 2002 at 03:12:10PM -0800, Srinath Avadhanula <srinath=
@fastmail.fm> wrote:
>>
>>>Option 1:
>>>    change things like '\onlyslides{??}??' to
>>>    "\\onlyslides{\xab\xbb}\xab\xbb"
>>
>>Hard to read IMHO. Hard to maintain.
>>
>>>Option 2:
>>>    Change it to something like "\\onlyslides{<++>}<++>"
>>>    Here the placeholder (or marker) characters are "<+" and "+>".
>>
>>Easier to read, but imaps may not work with a new language using '<+'
>>and '+>'. I think you should choose something expressive and odd enough
>>to not beeing used by anything else.
>=20
>=20
> But it is very hard to find something odd and common to various
> encodings. Why not simple '<' and '>'? Easy to read; easy to maintain;
> everywhere won't be problems with encoding; shorter than '<+', '+>'.
>=20
> Mikolaj

      I guess I will not actually implement anything today...

      IIUC, templates are supposed to include prompts like

\label{fig:=ABlabel=BB}

1. If we use < and > then we will only occasionally have trouble when=20
the file contains

	A silly example:  $3 < 5 > 4$

or some such.  I type <C-J> and get "< 5 >" in Select mode.  OK,=20
provided that <C-J> jumps to the "<label>" and does not delete "< 5 >";=20
is this what happens, or does the place holder get deleted if I hit=20
<C-J> again?

      More realistic:  lots of people are too lazy to use $\langle$ and=20
$\rangle$, so $<f, g>$ is likely to come up.

     If I try this with an HTML file, things get ugly.  Need I say more?

2. The current system has problems.  Here is a new one.  I try

:set enc=3Dlatin1 fenc=3Dlatin1

so that things look right as I edit plugin/imaps.vim (and read=20
doc/latex-suite.txt).  Now, when I copy from gvim and paste into=20
Mozilla, gvim gets an X-windows error and crashes.

3. The "<++>" proposal might work, but I will make two predictions:
(a) Some crazed mathematician will decide that "<+f, g+>" is the right=20
notation for some sort of inner product.
(b) If there is not already an XML variant that uses "<+tags=20
like=3Dthis+>", there will be soon.

Conclusions:
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

      The place holders can already be configured.  We have to encourage=20
the use of this option, and we have to change the defaults.  We have to=20
allow for place holders with more than one character.

      I think that "<++>" is a reasonable default.

      I think that the IMAP() function should be changed.  Currently:

	:call IMAP(lhs, rhs, ft)

and rhs optionally contains the place-holder characters.  If the user=20
decides to change these characters, (s)he must rewrite all the calls to=20
IMAP().  I propose changing this to

	:call IMAP(ft, lhs, rhs1, ph1, rhs2, ph2, ...)

with a variable number of arguments.  The rhsx arguments are plain text.=20
  The phx arguments are either "<" or ">" and indicate that the=20
corresponding place holder should be inserted.  Or simpler:  if place=20
holders always come in pairs, and I am not missing a third kind of=20
special character, the phx arguments are also plain text, to be enclosed=20
in "<>" pairs.  For example,

	:call IMAP('tex', 'bf`', '\textbf{', '', '}', '')

would expand "bf`" to "\textbf{<>}<>".

					--Benji