[Libreoffice-bugs] [Bug 130725] ImpSvNumberInputScan::StringToDouble may produce inaccurate result

bugzilla-daemon at bugs.documentfoundation.org bugzilla-daemon at bugs.documentfoundation.org
Mon Feb 24 17:57:34 UTC 2020


https://bugs.documentfoundation.org/show_bug.cgi?id=130725

--- Comment #16 from Jonny Grant <jg at jguk.org> ---
(In reply to Mike Kaganski from comment #15)
> (In reply to Oliver Brinzing from comment #14)
> > (In reply to Jonny Grant from comment #13)
> > > Speaking frankly, I give up. Google Calc doesn't suffer this bug.
> > 
> > Are you sure ?
> 
> Jonny Grant mixes the two problems here in the crusade to achieve Casio
> precision. The real problem solved here in this specific bug is incorrect
> conversion between string containing a decimal number, and its closest IEEE
> 754 double precision representation. And this problem indeed is
> LibreOffice-specific. But Jonny Grant tries to expand the scope of the bug,
> which is a wrong thing to do; and then, indeed, the statement that GSheets
> or Excel or whatever do things differently is wrong.

There may be a misunderstanding. I don't seek casio. I seek my accounant, my
broker, my auditor to be able to see 0.001 without them suffering because
LibreOffice isn't catering for users. It's a very simple problem, and a simple
fix?

Mike, Many thanks for filing my bug before I could. It's good you filed it. But
the accurate bug description is on the one I filed.
https://bugs.documentfoundation.org/show_bug.cgi?id=130728

The screenshot is attached to that. I'll attach the screenshot here for
completeness.

Easy steps to reproduce, just open any new Google Docs Spreadsheet, and type
the cells as per screengrab

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/libreoffice-bugs/attachments/20200224/5db1d894/attachment.htm>


More information about the Libreoffice-bugs mailing list