Bram Moolenaar wrote:
> Here is an update for the floating point patch.
I've done a bit more testing with the latest floating point patch. Vim's
floating
point doesn't need to cover all extremes, but the following shows some bad
results.
:help floating-point-format (in eval.txt)
It says you can have negative number: &-0.123
:let a = &-0.123
Result: E112: Option name missing: &-0.123
:let a = -&0.123
This works but is inelegant.
str2float('1.75') works by converting '1' to 1 and '75' to 75, then combining
them
(1 + 75/10^2). That gives a couple of problems. First, should probably document
that
a 32-bit machine won't keep 15 digits of precision in something like
str2float('0.123456789012345').
Second, the lhs and rhs of the '.' can each overflow an integer, giving chaos.
:let a=str2float('0.3111222333')
:echo a
Result: -0.118374
In the next example, Vim calculates (-1 + 0.75).
:let a=str2float('-1.75')
:echo a
Result: -0.250000
Typo in eval.c ("sting" --> "string"):
Convert the sting "text" to a floating point number.
John
--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---
