[gcj] Re: A late question about 'watersheds'..

2009-09-08 Thread Nactive
Well, I had the same problem but in the example input the last case is: 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 The answer is: a b c d e f g h i j k l m n o p q r s t u v w x y z That made it pretty clear for me. On Sep 7, 4:55 am, jz wrote: > hey folks, good job with gcj. > > i

[gcj] Re: A late question about 'watersheds'..

2009-09-07 Thread gislan
On Sun, 6 Sep 2009 19:55:54 -0700 (PDT), jz wrote: > in following case in sample input, > [2 3] > 7 6 7 > 7 6 7 > > now upper 6 and downer 6 has same altitude. > So what i thought was i should follow "NORTH WEST EAST SOUTH" rule, > so upper 6 must be the sink, and the basin map should be Water

[gcj] Re: A late question about 'watersheds'..

2009-09-07 Thread Hawston LLH
I think your misunderstanding is quite understandable, any language can be as confusing. If without reading the sink rule ("lower" ...), the phrase "In case of a tie" could arguably interpreted as "tie with current cell" or "conflict among neighbors", so it is always better to read carefully what t

[gcj] Re: A late question about 'watersheds'..

2009-09-07 Thread jz
ok now i'm clear. if i'm the only one who's confused in it, i must study english harder T_T. thanks for your help, Hawston :) On Mon, Sep 7, 2009 at 4:32 PM, Hawston LLH wrote: > nothing to do with the sequence of the two rules because they are saying > totally different things, the tie mean t

[gcj] Re: A late question about 'watersheds'..

2009-09-07 Thread Robert
On Sep 7, 9:11 am, jz wrote: > ● For each cell, if none of its 4 neighboring cells has a lower altitude > than the current cell's, then the water does not flow, and the current cell > is called a *sink*. I you take a look at the lower 6, you'll see it has no neighbor with a lower altitude (pay a

[gcj] Re: A late question about 'watersheds'..

2009-09-07 Thread Hawston LLH
nothing to do with the sequence of the two rules because they are saying totally different things, the tie mean tie among the neighbors (case of multiple choices, but it does not say how the choice is derived in this statement), and obviously nothing to do with current cell having same altitude as

[gcj] Re: A late question about 'watersheds'..

2009-09-07 Thread jz
thanks everybody. i'm still not 100% clear though, but maybe the former rule comes prior to the latter one? ● For each cell, if none of its 4 neighboring cells has a lower altitude than the current cell's, then the water does not flow, and the current cell is called a *sink*. ● In case of a tie, w

[gcj] Re: A late question about 'watersheds'..

2009-09-06 Thread Hawston LLH
i think it depend on the definition of "flow", either from (high) to (low) or from (high) to (equal or lower). based on the definition given in the question "For each cell, if none of its 4 neighboring cells has a lower altitude than the current cell's, then the water does not flow, and the current

[gcj] Re: A late question about 'watersheds'..

2009-09-06 Thread Dhruva Sagar
According to the rules, water will not flow from the 6 in the first row to the 6 in the second row.Hence the second row will take up 'b' because none of the paths from the second row end up in the first row, it is a completely different/new path. Hope that makes it at least a bit more clear. Than

[gcj] Re: A late question about 'watersheds'..

2009-09-06 Thread tog
Well I think that since 6 is a sink you can not move to the other one ... You cannot move to a place being at the same height (of your current position) - no slope ;) On Mon, Sep 7, 2009 at 8:25 AM, jz wrote: > > hey folks, good job with gcj. > > i couldn't understand something in prob B 'waters