After upgrading my copy of the CSApprox plugin
http://www.vim.org/scripts/script.php?script_id=2390 to version 4.00 I
noticed that if a global Funcref is set to a script-local function in
its defining script, the Funcref will be stored as pointing to
s:<something>, not to <SNR>nn_<something>, i.e., the context is lost,
and any attempt to invoke that Funcref from outside its defining script
causes an error, because Vim tries to invoke the script-local function
in the context of the caller.
I admit that my colorscheme has one fancy feature: it alternates the
StatusLine and WildMenu highlights by means of a CursorHold,CursorHoldI
autocommand. In Console mode with CSApprox installed, in a terminal
capable of (88 or) 256 colours, this means setting highlights for these
two highlight groups only, using a CSApprox function to get the ctermfg
and ctermbg values without (for performance) running the whole plugin as
the CursorScheme autocommand event would. A further complication is that
the colorscheme is sourced early, by a :colorscheme statement in the
vimrc, while the plugin/CSApprox.vim script is only sourced afterwards,
together with other global plugins.
The following doesn't work (the first :let statement triggers an error):
if exists('g:CSApprox_approximator_function')
let s:ctbg1 = g:CSApprox_approximator_function(0, 255, 0) " green
let s:ctbg2 = g:CSApprox_approximator_function(255, 0, 0) " red
let s:ctfg = g:CSApprox_approximator_function(0, 0, 0) " black
else
let s:ctbg1 = 'darkgreen'
let s:ctbg2 = 'black'
let s:ctfg = 'white'
endif
The following, which uses the contents of the contents of the Funcref,
works (it causes an early sourcing of
autoload/csapprox/per_component.vim and autoload/csapprox/common.vim,
with no ill effects AFAICS):
try
let s:ctbg1 = csapprox#per_component#Approximate(0, 255, 0) " green
let s:ctbg2 = csapprox#per_component#Approximate(255, 0, 0) " red
let s:ctfg = csapprox#per_component#Approximate(0, 0, 0) " black
catch
let s:ctbg1 = 'darkgreen'
let s:ctbg2 = 'black'
let s:ctfg = 'white'
endtry
Best regards,
Tony.
--
Real computer scientists despise the idea of actual hardware. Hardware
has limitations, software doesn't. It's a real shame that Turing
machines are so poor at I/O.
--
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php
