[implementations-list] Vimpulse update
Alessandro Piras
laynor at gmail.com
Wed Jun 17 23:50:54 CEST 2009
Little update to VImpulse.
I'm happy to announce we have a visual mode that (almost) doesn't suck.
1. works like vim in normal visual mode: the character under the cursor
when you hit "v" is always selected, and so is the character under the
cursor when you move.
2. works like vim in linewise mode: the line you're on is selected, the
cursor movement is not restricted to the beginning of the line.
As a consequence "0v$" now selects the whole line, not just the characters
up to the second last in the line, and "ggVG" selects everything, not the
text up to the second last line.
In other words, it doesn't suck.
I'm going to upload the stuff on emacs wiki when I discover how to.
There are 2 things that still bother me (i guess more than 2, but only those
come to mind now):
A) the parens matching highlight sucks big time. example:
if (giancane(cancane(giangiovanni(x), gioacchino(y)),
gianny(x)))
I'd like to hit $ and see if the parens match (in normal mode). But stupid
emacs higlights instead the second last paren. going forward one chare _is
not_ an option, as I would have to hit some stupid arrow key because viper
constrains the movement (and vim too).
So this is the next thing I'm going to look at.
B) Before talking about the annoyance, I must say thanks to the visual block
mode developers, from what I read in the comments, cppjavaperl and Frank
Fisher maybe.
It works great, and it's very usable as is. But it suffers of the same
suckage of the old visual mode:the cursor must be placed 1 column forward of
what you would expect. This leads to the same kind of suckage of the paren
matching: if you want to block select from column n to column m, and m
happens to be the last column in some line, you're forced to touch the arrow
keys. As I know nothing about how the block mode is implemented, this will
take some more time. I'll ask for information if needed.
Cheers,
Alessandro
More information about the implementations-list
mailing list