Code wrongly replaced decimal separator with '.' and used atof() to scan
the number string which itself may be localized on *iX systems but not
on Windows. Hence on *iX the numbers may had been truncated where on
Windows they were not.

Additionally made this work with Unicode separators if defined for the
locale, removed the awkward conversion to ASCII byte string and
eliminated use of toupper() and isdigit() calls. Also eliminated a
possible buffer overflow.

(cherry picked from commit 9e9f39d1)

Conflicts:
	basic/source/sbx/sbxscan.cxx

replace decimal separator at correct position, rhbz#919020

(cherry picked from commit e96c7a60)

Change-Id: I35d82e8f267ecb925783bf6f2de044a29c08f688
Reviewed-on: https://gerrit.libreoffice.org/2859Reviewed-by: Noel Power <noel.power@suse.com>
Tested-by: Noel Power <noel.power@suse.com>
(cherry picked from commit 96b079e2)
Reviewed-on: https://gerrit.libreoffice.org/2893Reviewed-by: Fridrich Strba <fridrich@documentfoundation.org>
Tested-by: Petr Mladek <pmladek@suse.cz>
Reviewed-by: Petr Mladek <pmladek@suse.cz>