This is the final version, the previous one didn't work so well with "inf", now it does. I've put it on my GitHub, get it here https://github.com/alexandros301/vanillaS2f

On Tue, Jul 7, 2015 at 12:20 PM, Alexandros Drymonitis <adrcki@gmail.com> wrote:


On Tue, Jul 7, 2015 at 6:35 AM, Chris McCormick <chris@mccormick.cx> wrote:
Hello,

On 07/07/15 03:31, Alexandros Drymonitis wrote:
> On Mon, Jul 6, 2015 at 8:56 PM, Cyrille Henry <ch@chnry.net
> <mailto:ch@chnry.net>> wrote:
>     look like it's more a s2i than a s2f
> True! Now it's real s2f

Having tried this strategy myself I discovered there are many pitfalls.
The following are valid floating point representation that your
abstraction will not parse:

-2.58934e+27
Covered!

I think there are also valid values like "-0.0" and "inf" that can come
from floating point numbers rendered as strings and could cause some
troubles.
Covered "inf", but not "-0.0", it will just output -0