The patches in this repository change how Vim handles lisp-mode indentation and how Vim handles yanking an area in the text.
Lisp mode indentation is turned on by using set lisp, either in the Vim commandline (i.e. at the : prompt). It can also be set in your vimrc (you can take a look at my vimrc at akeep/vimrc.
The big change is that in the traditional lisp mode all non-keywords are setup to indent by the length of the item in the function location. For example, if I write a function: my-great-function and I want to call it with arguments on multpile lines I get:
(my-great-function a
b c
d)
Or I can get:
(my-great-function
a b c d)
But what I really want is:
(my-great-function a
b c
d)
In the normal lisp mode, we can use lispwords to control this. So, for
instance, we can do set lispwords+=my-great-function and get the solution we
want. But, this quickly gets tiresome, and you end up having to update the
lispwords each session, or update your vimrc each time you use it.
This patch reverses this, and adds a hack to lispwords. Instead of assuming we
always want to tab things in the length of the first function, we now tab in by
the indent amount at each call. We have added a ! to the front of lispwords
to indicate that these are the things we want to indent based on the length of
the first element. My default, in the scheme.vim of my akeep/vimrc project, I
have if, or, and, +, -, *, /, <, =, and > in there. This
is set with set lispwords=!if,or,and,+,-,*,/,<,=,>.
By default Vim yank leaves you at the mark that you yanked to, rather than the location you ynaked from. For instance, if you have the following mark set:
| marks | code |
|---|---|
| m | (let ([x e] ---) |
| --- | |
| ---) | |
| > | (foo ---) |
Where m is your mark and > is your cursor, using Vim yank will leave your
cursor at m. Our patch allows you to tell Vim to leave your cursor at >.
This is set by using set yankleavecursor=1. If you don't like this, but you
still want to apply the patch, just leave the yankleavecursor unset or set to 0
and you will get Vim's normal behavior with this.
These patches have been tested with Vim 7.3 on Linux and with MacVim on Mac OS X, and applied successfully to both. You should be aware that building Vim from source can be a bit of a pain if you are trying to make sure you get things like gvim, etc. The trick here is really that you need to tell the configure script to look for the GUI interface (like X11, Gnome, Gtk+, etc.) that you want it to try.
If you are using a different version of Vim, and the patches fail for you, and you cannot get them to work, let me know. (If you figure out how to fix it and want to send me the fix, that is also good.)
I have included the patches in two different ways:
- A unified patch with all the changes in it, based on the vim-7.3 tar ball
- Individual file patches organized by the directory that they fall in. (This is the original version of the patches and included here in case some parts give you fits.)
If you are using the combined patch (vim-patches.patch) you should be able to do the following within the vim73 directory:
patch -p2 < vim-patches.patch
Where the -p2 strips off the first two levels of directories in the patch.
If you are individually patching the files you can do something like:
patch normal.c < normal.c.patch
For each of the individual files.