Re: [lmi] Group quotes, part deux

[Greg Chicares]

On 2016-05-19 21:24, Vadim Zeitlin wrote:
[...]
> I'm really sorry about this problem

My fault, not yours.

>  Indeed, thanks a lot to Kim for finding this. Out of curiosity, how did
> you do it? The difference is relatively small (although far from
> insignificant, of course), so how did you manage to not miss it?

[If I may answer for Kim] Design test cases to magnify anomalies.
It's a rare skill that often goes unrecognized.

Rather than send you an input file that requires proprietary
product files, let me just paste the test case from a "roster"
spreadsheet. If 'icedove' sends literal tabs, then you can
paste it back into a spreadsheet and everything will line up.
Even if that doesn't work, you'll sense the logic (column
headings will be obvious to you at this point):

1000000 564.63  100000  370.38  1100000 935.01          3.7038

1000000 333.40  50000   293.31  1050000 626.71          5.8662

1000000 407.66  100000  337.67  1100000 745.33          3.3767

1000000 116.42  100000  225.85  1100000 342.27          2.2585

930000  83.65   50000   195.65  980000  279.30          3.9130

985000  238.10  100000  285.74  1085000 523.84          2.8574

889000  79.96   100000  191.29  989000  271.25          1.9129

871000  155.89  50000   223.67  921000  379.56          4.4734

953000  351.34  50000   285.83  1003000 637.17          5.7166

904000  114.70  50000   208.97  954000  323.67          4.1794

                                                        

9532000 2445.75 750000  2618.36 10282000        5064.11         3.8258

        0.2566          3.4911          0.4925


Expected 3.49, got 3.83 .