[issue1580] Use shorter float repr when possible

[Preston Briggs]

Preston Briggs <prest...@google.com> added the comment:

>> It'd probably have to be touched up a bit.

> This may be an understatement. :-)

Probably so.  Nevertheless, it's got to be easier
than approaching the problem from scratch.
And considering that this discussion has been
going on for over a year now, it might be a way to
move forward.

> In the first 50 lines of the 3897-line dtoa.c file, I see this warning:
>
> /* On a machine with IEEE extended-precision registers, it is
>  * necessary to specify double-precision (53-bit) rounding precision
>  * before invoking strtod or dtoa.  [...] */
>
> It seems to me that figuring out how and when to do this, and
> writing and testing the appropriate configure/pyport/whatever
> code, is already a nontrivial task.

I would consider compiling the library with flags appropriate to forcing
64-bit IEEE arithmetic if possible.  Another possibility is to explore
gdtoa.c
which is supposed to include code for extended precision, quad precision,
etc.

Added file: http://bugs.python.org/file13176/unnamed

_______________________________________
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue1580>
_______________________________________
<div class="gmail_quote">&gt;&gt; It&#39;d probably have to be touched up a 
bit.<br>
<br>
&gt; This may be an understatement. :-)<br><br>Probably so.  Nevertheless, 
it&#39;s got to be easier<br>than approaching the problem from scratch.<br>And 
considering that this discussion has been<br>going on for over a year now, it 
might be a way to<br>
move forward.<br>
<br>&gt; In the first 50 lines of the 3897-line dtoa.c file, I see this 
warning:<br>
&gt; <br>
&gt; /* On a machine with IEEE extended-precision registers, it is<br>
&gt;  * necessary to specify double-precision (53-bit) rounding precision<br>
&gt;  * before invoking strtod or dtoa.  [...] */<br>
&gt; <br>
&gt; It seems to me that figuring out how and when to do this, and<br>&gt; 
writing and testing the appropriate configure/pyport/whatever<br>
&gt; code, is already a nontrivial task.<br>
<br>I would consider compiling the library with flags appropriate to 
forcing<br>64-bit IEEE arithmetic if possible.  Another possibility is to 
explore gdtoa.c<br>which is supposed to include code for extended precision, 
quad precision, etc.<br>
<br></div>

_______________________________________________
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com