No subject

[Alex Tweedly]

Brian Yennie wrote:

> Here's on last stab at using Rev's built-in sort:
>
> sort lines of fld 1 numeric by (lineOffset(word 1 of each, fld 
> 1)*100000 + (word 2 of each))
>
> Woo hoo! Everyone loves cryptic one-liners! =P
>
I don't know if that quite gets it .... if the input were
IPAE 1
LAG 2
IPAE 3
LAG 4

then your code would give
IPAE 1
IPAE 3
LAG 2
LAG 4

while the non-Rev-sorting versions would have kept the data unchanged 
(they preserved blocks of lines with the same first item, rather than 
actually sorted on it).

Howard said:

>>> Yours puts LAG 1 at the end, since it sorts the entire list
>>> aphabetically. So if he wants to keep the original alpha order intact
>>> and just sort within the alpha groups, you need to add a little more
>>> control to the sorts. Otherwise the simple solution is fine.
>>
>>
I have long forgotten what the original question was - so I have no idea 
which would be "right" :-)


-- 
Alex Tweedly       http://www.tweedly.net



-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.8.6 - Release Date: 30/03/2005