Bill McCarthy wrote:
> does this LC_NUMERIC actually work on Windows - I recall 
> having problems with LC_ALL using the unix sort of GNU 
> coreutils - is this an environment variable we need to set?

Just to spell out the situation: In a country where ',' is used for the decimal
point, a user might have their environment variables set to indicate that fact.
However, Bram decided to have *Vim* override the environment variables so the 
Vim
process will always use '.' for the decimal point.

You don't have to do anything. Vim configures LC_NUMERIC when running to ensure 
that
the C library function strtod() works in a consistent way when used in Vim 
(i.e. Vim
expressions will always interpret "1.23" as a float, and will never interpret 
"1,23"
as a float).

Bram's suggestion was that those who do use ',' should ensure they have 
LC_NUMERIC
set so "1,23" is a float in most applications, then confirm that in Vim, it's 
"1.23"
that is a float, while "1,23" is not.

The previous float patches were using "do-it-yourself" code to convert a string 
to a
floating point number (which gave rise to the weirdness we discussed last 
April).
However, Bram has now decided to use strtod() because it handles overflow and
underflow as well as other tricky issues.

Re your comment on 'sort', you probably do want to 'set LC_ALL=C' before 
running the
GNU sort on Windows.

John


--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---

Raspunde prin e-mail lui