[GRASS-dev] Re: bug in Vect_cidx_find_next() ?

[Moritz Lennert]

Radim Blazek wrote:
> Yes,
> it seems to be bug, I think that on row 296 in cindex.c
> the requested cat should also be checked, not only the type.
> But I am tired now, please check it twice.

Changing that line into

if ( ci->cat[cat_index][0] == cat && ci->cat[cat_index][1] & type_mask )

works. Thanks !

But I don't really understand how this could happen, as I thought the 
following lines (286-288) should take care of this before:

while ( cat_index > start_index ) {
     if ( ci->cat[cat_index-1][0] != cat ) {
        break;

So it might actually be a problem in the determination of cat_index 
which is higher than it should be.

But just as you, I'm too tired now to look any deeper into this...
;-)

Moritz

> Bravo Moritz!
> 
> Radim
> 
> On 9/28/06, Moritz Lennert <mlennert at club.worldonline.be> wrote:
>> Hello,
>>
>> Trying to rework d.vect.chart I am using the Vect_cidx_find_next() 
>> function.
>>
>> However, using a map with areas I get behaviour which seems incorrect
>> and which I cannot explain.
>>
>> I have the following category index (result of Vect_cidx_dump(Map, 
>> stdout)):
>>
>> .
>> .
>> [Field 0]
>> .
>> .
>> Field      1  number of unique cats:     589  number of cats:    1185
>> number of types: 3
>> ------------------------------------------------------------------------------------------ 
>>
>>              type |     count
>>                 8 |       592
>>                 4 |         1
>>                64 |       592
>>   category | type | line/area
>>      11001 |    8 |      1895
>>      11001 |   64 |       127
>>      11002 |    8 |      1810
>>      11002 |   64 |        42
>>      11004 |    8 |      1874
>>      11004 |   64 |       106
>>      11005 |    8 |      1923
>>      11005 |   64 |       155
>>      11007 |    8 |      1866
>>
>> etc.
>>
>> Now, when I launch
>>
>> Vect_cidx_find_next( Map, field_index=1, cat=11002, type=15,
>> start_index=3, &stype, &sid )
>>
>> it returns 4 and line id=1874.
>>
>> This is wrong as the fourth item in the index, which does have line id
>> 1874, has cat 11004 and not cat 11002. This is confirmed by launching
>>
>> Vect_cidx_get_cat_by_index ( Map, field_index, 4, &testcat, &testtype,
>> &testid)
>>
>> which returns testcat = 11004.
>>
>> Am I right in assuming that there is a bug in Vect_cidx_find_next, or am
>> I misunderstanding something here ?
>>
>> I don't have this problem with a pure point file, which only has one 
>> type:
>>
>> Field      1  number of unique cats:     589  number of cats:     593
>> number of types: 1
>> ------------------------------------------------------------------------------------------ 
>>
>>              type |     count
>>                 1 |       593
>>   category | type | line/area
>>      11001 |    1 |        97
>>      11002 |    1 |        12
>>      11002 |    1 |       593
>>      11004 |    1 |        76
>>      11005 |    1 |       125
>>      11007 |    1 |        68
>>
>>
>> So this seems to be linked to the type handling.
>>
>> Any ideas,
>>
>> Moritz
>>
> 
> _______________________________________________
> grass-dev mailing list
> grass-dev at grass.itc.it
> http://grass.itc.it/mailman/listinfo/grass-dev