What are your suggestions for an ideal Vim configu

问题:

There are a lot of threads pertaining to how to configure Vim/GVim for Perl development on PerlMonks.org. My purpose in posting this question is to try to create, as much as possible, an ideal configuration for Perl development using Vim/GVim. Please post your suggestions for .vimrc settings as well as useful plugins.

I will try to merge the recommendations into a set of .vimrc settings and to a list of recommended plugins, ftplugins and syntax files.

.vimrc settings

"Create a command :Tidy to invoke perltidy"
"By default it operates on the whole file, but you can give it a"
"range or visual range as well if you know what you're doing."
command -range=% -nargs=* Tidy <line1>,<line2>!
    \perltidy -your -preferred -default -options <args>

vmap <tab> >gv    "make tab in v mode indent code"
vmap <s-tab> <gv

nmap <tab> I<tab><esc> "make tab in normal mode indent code"
nmap <s-tab> ^i<bs><esc>

let perl_include_pod   = 1    "include pod.vim syntax file with perl.vim"
let perl_extended_vars = 1    "highlight complex expressions such as @{[$x, $y]}"
let perl_sync_dist     = 250  "use more context for highlighting"

set nocompatible "Use Vim defaults"
set backspace=2  "Allow backspacing over everything in insert mode"

set autoindent   "Always set auto-indenting on"
set expandtab    "Insert spaces instead of tabs in insert mode. Use spaces for indents"
set tabstop=4    "Number of spaces that a <Tab> in the file counts for"
set shiftwidth=4 "Number of spaces to use for each step of (auto)indent"

set showmatch    "When a bracket is inserted, briefly jump to the matching one"

syntax

• vim-perl: Support for Perl 5 and Perl 6 in Vim

plugins

• delimitMate provides auto-completion for quotes, parens, brackets, etc. in insert mode. It handles apostrophes more intelligently than closepairs.vim does.

• perlhelp.vim: Interface to perldoc

• taglist.vim: Source code browser

ftplugins

• perldoc.vim: perldoc command from vim

CPAN modules

• Vim::X

Debugging tools

I just found out about VimDebug. I have not yet been able to install it on Windows, but looks promising from the description.

回答1:

From chromatic's blog (slightly adapted to be able to use the same mapping from all modes).

vmap ,pt :!perltidy<CR> 
nmap ,pt :%! perltidy<CR>

hit ,pt in normal mode to clean up the whole file, or in visual mode to clean up the selection. You could also add:

imap ,pt <ESC>:%! perltidy<CR>

But using commands from input mode is not recommended.

回答2:

" Create a command :Tidy to invoke perltidy.
" By default it operates on the whole file, but you can give it a
" range or visual range as well if you know what you're doing.
command -range=% -nargs=* Tidy <line1>,<line2>!
    \perltidy -your -preferred -default -options <args>

回答3:

Look also at perl-support.vim (a Perl IDE for Vim/gVim). Comes with suggestions for customizing Vim (.vimrc), gVim (.gvimrc), ctags, perltidy, and Devel:SmallProf beside many other things.

回答4:

Perl Best Practices has an appendix on Editor Configurations. vim is the first editor listed.

回答5:

Andy Lester and others maintain the official Perl, Perl 6 and Pod support files for Vim on Github: https://github.com/vim-perl/vim-perl

回答6:

For tidying, I use the following; either \t to tidy the whole file, or I select a few lines in shift+V mode and then do \t

nnoremap <silent> \t :%!perltidy -q<Enter>
vnoremap <silent> \t :!perltidy -q<Enter>

Sometimes it's also useful to deparse code. As the above lines, either for the whole file or for a selection.

nnoremap <silent> \D :.!perl -MO=Deparse 2>/dev/null<CR>
vnoremap <silent> \D :!perl -MO=Deparse 2>/dev/null<CR>

回答7:

.vimrc:

" Allow :make to run 'perl -c' on the current buffer, jumping to 
" errors as appropriate
" My copy of vimparse: http://irc.peeron.com/~zigdon/misc/vimparse.pl

set makeprg=$HOME/bin/vimparse.pl\ -c\ %\ $*

" point at wherever you keep the output of pltags.pl, allowing use of ^-]
" to jump to function definitions.

set tags+=/path/to/tags

回答8:

Here's an interesting module I found on the weekend: App::EditorTools::Vim. Its most interesting feature seems to be its ability to rename lexical variables. Unfortunately, my tests revealed that it doesn't seem to be ready yet for any production use, but it sure seems worth to keep an eye on.

回答9:

Here are a couple of my .vimrc settings. They may not be Perl specific, but I couldn't work without them:

set nocompatible        " Use Vim defaults (much better!) "
set bs=2                " Allow backspacing over everything in insert mode "
set ai                  " Always set auto-indenting on "
set showmatch           " show matching brackets "

" for quick scripts, just open a new buffer and type '_perls' "
iab _perls #!/usr/bin/perl<CR><BS><CR>use strict;<CR>use warnings;<CR>

回答10:

I have 2.

The first one I know I picked up part of it from someone else, but I can't remember who. Sorry unknown person. Here's how I made "C^N" auto complete work with Perl. Here's my .vimrc commands.

" to use CTRL+N with modules for autocomplete "
set iskeyword+=:
set complete+=k~/.vim_extras/installed_modules.dat

Then I set up a cron to create the installed_modules.dat file. Mine is for my mandriva system. Adjust accordingly.

locate *.pm | grep "perl5" | sed -e "s/\/usr\/lib\/perl5\///" | sed -e "s/5.8.8\///" | sed -e "s/5.8.7\///" | sed -e "s/vendor_perl\///" | sed -e "s/site_perl\///" | sed -e "s/x86_64-linux\///" | sed -e "s/\//::/g" | sed -e "s/\.pm//" >/home/jeremy/.vim_extras/installed_modules.dat

The second one allows me to use gf in Perl. Gf is a shortcut to other files. just place your cursor over the file and type gf and it will open that file.

" To use gf with perl "
set path+=$PWD/**,
set path +=/usr/lib/perl5/*,
set path+=/CompanyCode/*,   " directory containing work code "
autocmd BufRead *.p? set include=^use
autocmd BufRead *.pl set includeexpr=substitute(v:fname,'\\(.*\\)','\\1.pm','i')

回答11:

I find the following abbreviations useful

iab perlb  print "Content-type: text/html\n\n <p>zdebug + $_ + $' + $`  line ".__LINE__.__FILE__."\n";exit;
iab perlbb print "Content-type: text/html\n\n<p>zdebug  <C-R>a  line ".__LINE__.__FILE__."\n";exit;
iab perlbd do{print "Content-type: text/html\n\n<p>zdebug  <C-R>a  line ".__LINE__."\n";exit} if $_ =~ /\w\w/i;
iab perld print "Content-type: text/html\n\n dumper";use Data::Dumper;$Data::Dumper::Pad="<br>";print Dumper <C-R>a ;exit;

iab perlf foreach $line ( keys %ENV )<CR> {<CR> }<LEFT><LEFT>
iab perle while (($k,$v) = each %ENV) { print "<br>$k = $v\n"; }
iab perli x = (i<4) ? 4 : i;
iab perlif if ($i==1)<CR>{<CR>}<CR>else<CR>{<CR>}
iab perlh $html=<<___HTML___;<CR>___HTML___<CR>

You can make them perl only with

au bufenter *.pl iab xbug print "<p>zdebug ::: $_ :: $' :: $`  line ".__LINE__."\n";exit;

回答12:

By far the most useful are

1. Perl filetype pluging (ftplugin) - this colour-codes various code elements
2. Creating a check-syntax-before-saving command "W" preventing you from saving bad code (you can override with the normal 'w').

Installing he plugins are a bit dicky as the version of vim (and linux) put the plugins in different places. Mine are in ~/.vim/after/

my .vimrc below.

set vb
set ts=2
set sw=2
set enc=utf-8
set fileencoding=utf-8
set fileencodings=ucs-bom,utf8,prc
set guifont=Monaco:h11
set guifontwide=NSimsun:h12
set pastetoggle=<F3>
command -range=% -nargs=* Tidy <line1>,<line2>!
    \perltidy
filetype plugin on
augroup JumpCursorOnEdit
 au!
 autocmd BufReadPost *
 \ if expand("<afile>:p:h") !=? $TEMP |
 \ if line("'\"") > 1 && line("'\"") <= line("$") |
 \ let JumpCursorOnEdit_foo = line("'\"") |
 \ let b:doopenfold = 1 |
 \ if (foldlevel(JumpCursorOnEdit_foo) > foldlevel(JumpCursorOnEdit_foo - 1)) |
 \ let JumpCursorOnEdit_foo = JumpCursorOnEdit_foo - 1 |
 \ let b:doopenfold = 2 |
 \ endif |
 \ exe JumpCursorOnEdit_foo |
 \ endif |
 \ endif
 " Need to postpone using "zv" until after reading the modelines.
 autocmd BufWinEnter *
 \ if exists("b:doopenfold") |
 \ exe "normal zv" |
 \ if(b:doopenfold > 1) |
 \ exe "+".1 |
 \ endif |
 \ unlet b:doopenfold |
 \ endif
augroup END