Nikolai Weibull wrote:
> On 9/17/07, Tony Mechelynck <[EMAIL PROTECTED]> wrote:
>
>> Nikolai Weibull wrote:
>
>>> Ranting aside, the prefix of number above is now#. I mean, have I
>>> gone to the trouble to declare what "namespace" the variable is in,
>>> shouldn't that be enough? And
>>>
>>> let s:bug#number = 1
>>>
>>> results in a
>>>
>>> E461: Illegal variable name: s:bug#number
>>>
>>> anyway, so there's no conflict in what you mean when you say bug#number.
>
>> s:number is not visible outside the script. It can never cause a sourcing of
>> the script.
>>
>> g:bug#number (or bug#number outside a function) is defined in the script but
>> visible everywhere. If you use it outside the script and it is undefined,
>> does
>> the script get loaded? Or is # just a lambda 'iskeyword' character when it
>> applies to a variable?
>
> No, it doesn't cause the script to get loaded. This to allow you to
> set "options" for an autoloaded script before it gets loaded.
>
> What is "lambda" supposed to mean here?
Hm? What is the exact English equivalent? random, average, any, John Q.
Public, typical, ordinary, ...
>
>> l:number ( or number, used inside a function) is not visible outside that
>> function. What about l:bug#number? I suppose it should be handled like
>> s:bug#number (both valid or both throwing an error).
>
> There are no such variable scopes. That's exactly what I said in the
> message you responded to.
>
>> There are other scopes: w:value, s:value and t:value are visible outside the
>> script, but not everywhere. I suppose w:bug#number, b:bug#number and
>> t:bug#number should also be valid.
>
> Why? None of the variables you talk about make much sense in having
> available. The only really useful scope for a autloaded script's
> variable, that is, a variable of the form script#variable, is global
> (g:).
Oh? Why not have a different buffer-local value for some variable pertaining
to a given autoload script, the way there are buffer-local options? I have a
b:match_words variable in some of my texts; that variable is different for
different 'syntax'es. If matchit was an autoload script, it would be
b:matchit#words or some such.
>
>> - name (with omitted scope) defaults to l:name if in a function, otherwise to
>> g:name
>
> Yes, script#name can never be local to a function. My suggestion is
> that script#name is clear enough in that it is a global variable.
>
> nikolai
Regards,
Tony.
--
A sense of humor keen enough to show a man his own absurdities will
keep him from the commission of all sins, or nearly all, save those
that are worth committing.
-- Samuel Butler
--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---
