[GRASS-dev] Re: [GRASS GIS] #548: grass7: v.reclass segfaults on string column

[Markus Neteler]

On Tue, Aug 4, 2009 at 8:54 AM, GRASS GIS<trac at osgeo.org> wrote:
> #548: grass7: v.reclass segfaults on string column
...
>  {{{
>  v.db.connect boundary_county -g
>  }}}
>
>  ?
>
>  I suppose DBF driver. Then it works... The key is the driver used for
>  input vector map.

Good point, indeed I was reading it with @neteler from a DBF mapset
(my bad, sorry). Copying it over into the SQLite DB fails however:

1. I made a copy of "boundary_county" in my own mapset "neteler" (DBF)
   from PERMANENT
2. Then I switched into the sqlite mapset and run:

g.copy vect=boundary_county at neteler,boundary_county
Program received signal SIGSEGV, Segmentation fault.

#gdb:
Program received signal SIGSEGV, Segmentation fault.
0x00007f979f9a3d69 in rtree_load_from_sidx (fp=0x7fffa8a8c5c8,
rootpos=179105, t=0xe67010, off_t_size=4)
    at spindex_rw.c:735
735                                 newnode->branch[j].child = NULL;


(gdb) bt
#0  0x00007f979f9a3d69 in rtree_load_from_sidx (fp=0x7fffa8a8c5c8,
rootpos=179105, t=0xe67010, off_t_size=4)
    at spindex_rw.c:735
#1  0x00007f979f9a4228 in dig_Rd_spidx (fp=0x7fffa8a8c5c8,
Plus=0x7fffa8a8c3b0) at spindex_rw.c:843
#2  0x00007f97a065dfff in Vect_open_sidx (Map=0x7fffa8a8c3a0, mode=1)
at open.c:891
#3  0x00007f97a065ca70 in Vect__open_old (Map=0x7fffa8a8c3a0,
name=0xe62bc0 "boundary_county",
    mapset=0xe5e650 "sqlite", update=1, head_only=1) at open.c:262
#4  0x00007f97a065d295 in Vect_open_update_head (Map=0x7fffa8a8c3a0,
name=0xe62bc0 "boundary_county",
    mapset=0xe5e650 "sqlite") at open.c:522
#5  0x00007f97a065760a in Vect_copy (in=0x7fffa8a891d0
"boundary_county", mapset=0xe635f0 "neteler",
    out=0xe62bc0 "boundary_county") at map.c:237
#6  0x0000000000401b5e in do_copy (n=2, old=0xe62ae0
"boundary_county at neteler", mapset=0xe62f20 "neteler",
    new=0xe62bc0 "boundary_county") at do_copy.c:27
#7  0x00000000004019a5 in main (argc=3, argv=0x7fffa8a8f428) at main.c:93
}}}

It is unrelated hence I don't clutter the trac ticket.
Once this works, I continue with the ticket.

Markus