[GRASS5] v.in.mif

[David D Gray]

Heiko Kehlenbrink wrote:

> Hi all,
> 
> i had just imported the 9023.MIF/MID files into GRASS from 
> www.english-nature.org.uk/pubs/gis .
> 
> I used  this Region:
> 
>                         REGION FACILITY
> LOCATION: britain                                  MAPSET: test
> 
> CURRENT REGION: N=732954  S=0  RES=1  ROWS=732954
>                E=730403  W=0  RES=1  COLS=730403
> PROJECTION: 99 (Transverse Mercator)
> ZONE:       0
> DATUM:      osgb36
> ELLIPSOID:  airy
> 
> First thing i had to do was rename the files from 9023.MIF and 9023.MID 
> to 9023.mif and 9023.mid.
> Than i had to add a logfile output(otherwise v.in.mif ended without 
> producing anything, not even an error), maybe it would be a good idea to 
> pipe the logging-output to nirwana
> by default?
> 
> 


Heiko

This is currently being revised, along with it's very similar sister 
module v.in.shape, to be more modular and have better import routines, 
and these improvements should be available by PR4. For now some remarks:


> GRASS:~ > v.in.mif  input=/home/heiko/grassdaten/britain/9023NA.mif 
> output=9023MIF logfile=/tmp/miflog.txt
> 
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.
> WARNING: Duplicate ring added.


I haven't seen these warnings are actually associated with any errors, 
and are probably spurious. It perhaps happens if a polygon is scanned 
after all its neighbours.


> WARNING: You must run `v.support option=build' on
>  this (these) map(s) to build the topology.
> 
> GRASS:~ > v.support map=9023MIF option=build
> V.SUPPORT:
> 
> Selected information from vector header
> Organization:
> Map Name:      9023MIF
> Source Date:
> Orig. Scale:   200
> No snapping will be done
> 
>    Reading Vector file 100%
>    Building areas 100%
>    Building islands 100%
>    Attaching labels
> PNT_TO_AREA failed: (487645.986637, 421993.662174) (Category 27)
> PNT_TO_AREA failed: (472080.339655, 415630.182221) (Category 28)
> PNT_TO_AREA failed: (396831.840312, 347753.296154) (Category 36)
> WARNING: area 62 label: 31 matched another label: 42.
> PNT_TO_AREA failed: (506536.608376, 386967.204960) (Category 43)
> PNT_TO_AREA failed: (413584.035007, 333139.904697) (Category 49)
> PNT_TO_AREA failed: (652335.198233, 303559.020389) (Category 58)
> PNT_TO_AREA failed: (646401.438483, 309376.050000) (Category 60)
> PNT_TO_AREA failed: (646401.438483, 309376.050000) (Category 61)
> WARNING: area 183 label: 66 matched another label: 67.
> WARNING: area 180 label: 78 matched another label: 79.
> PNT_TO_AREA failed: (541797.631173, 190794.412779) (Category 81)
> PNT_TO_AREA failed: (608065.746866, 154680.930000) (Category 92)
> PNT_TO_AREA failed: (571502.343881, 162589.318863) (Category 93)
> PNT_TO_AREA failed: (571502.343881, 162589.318863) (Category 95)
> PNT_TO_AREA failed: (453538.667104, 142008.500280) (Category 104)
> WARNING: area 176 label: 114 matched another label: 117.
> WARNING: area 174 label: 119 matched another label: 121.
> WARNING: area 8 label: 123 matched another label: 124.
> WARNING: area 172 label: 128 matched another label: 130.
> WARNING: area 143 label: 206 matched another label: 207.
> WARNING: area 143 label: 207 matched another label: 208.
> Number of lines:   424
> Number of nodes:   306
> Number of areas:   188
> Number of isles:   92
> Number of atts :   196
> Number of unattached atts :   13
> Snapped lines  :   0
> 
> 


The centroid location code is probably buggy. Code was inserted to find 
the centroids of actual imported polygons, based on mistrust of the 
values in the MIF file. That had reasons at the time, but is perhaps 
over-cynical, the new version will just scan the `Centre' values.


> After that v.in.mif worked for me , produced a Logfile containing:
> Mapinfo Data File:
> 
> VERSION: 300
> DELIMITER: ,
> Projection info:  Earth Projection 8, 79, "m", -2, 49, 0.9996012717, 
> 400000, -100000 Bounds (-7845061.1011, -15524202.1641) (8645061.1011, 
> 4470074.53373)
> 
> Field data -- 2 columns
> 
> Field 0 is nacode of type Integer
> Field 1 is naname of type Char
> 
> 
> I also tried to import the Field 1 data with :
> 
> GRASS:~ > v.in.mif  input=/home/heiko/grassdaten/britain/9023NA.mif 
> output=9023MIF logfile=/tmp/miflog.txt attribute=naname
> 
> or
> 
> GRASS:~ > v.in.mif  input=/home/heiko/grassdaten/britain/9023NA.mif 
> output=9023MIF logfile=/tmp/miflog.txt attribute=Field1
> 
> but failed.
> 


Likely there is a bug in the mid import stage. Again there's no point to 
maintaining this as changes should be made to the new version.

David