[GRASS5] U.S. Census Tiger Files 2000...

GRASS GRASS developer
1

FWIW:

I just tried a few v.tig.in.lndmk imports with 2000 data and didn't find a problem with one dataset.

Ascii delimited dumps are fine with me.

RIM I don't use and hasn't been useful when I tried.

***** My Other Random Comments

I have been parsing the Tiger files separately using perl scripts driven by data dictionary files which I copy from the Tiger specs. I use these to stuff Postgres with selected data for each record type mostly keyed by TLID. I have had problems doing a v.reclass on v.tig.in.basic data (TLID are too big for C operations) so I build by own reclass map in perl and write .att and .cat file directly.

As a map company I class all chains that say are roads by name in a separate category so I deal with each and every name. Thus I deal with categories that number in the 10,000s or more.

-Tom Rake, The Duff Map Company.

From: "Eric G. Miller" <egm2@jps.net>
To: GRASS5-dev <grass5@geog.uni-hannover.de>
CC: GRASSLIST <GRASSLIST@baylor.edu>
Subject: [GRASS5] U.S. Census Tiger Files 2000...
Date: Fri, 20 Jul 2001 18:33:46 -0700

If anyone would confirm that GRASS 5.0 can't import TIGER 2000 files
properly, I would appreciate it.

Apparently, the Census Bureau has changed the format of some of the
TIGER files slightly for the 2000 series. From my simple trials,
these changes don't affect v.in.tig.basic, but do break v.in.tig.lndmk.

Also, I guess we aren't really supporting the RIM database with GRASS
5.0? Are people using this? I haven't heard much about its use for a
while, so I'm not sure if it's desireable to continue supporting it
directly in v.in.tig.lndmk.

What do people prefer for managing the attributes? If I were to rewrite
this to support 2000 files, would delimited ASCII dumps be acceptable
(keyed via TLID, etc..). I've still a bit of homework to do on the file
formats...

--
Eric G. Miller <egm2@jps.net>
_______________________________________________
grass5 mailing list
grass5@geog.uni-hannover.de
http://www.geog.uni-hannover.de/mailman/listinfo/grass5

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

2

On Mon, Jul 23, 2001 at 05:37:41PM -0400, Tom Rake wrote:

FWIW:

I just tried a few v.tig.in.lndmk imports with 2000 data and didn't find a
problem with one dataset.

Hmm. I couldn't get it to work right with the line or area classes.
Maybe I'll try a different data set.

Ascii delimited dumps are fine with me.

RIM I don't use and hasn't been useful when I tried.

***** My Other Random Comments

I have been parsing the Tiger files separately using perl scripts driven by
data dictionary files which I copy from the Tiger specs. I use these to
stuff Postgres with selected data for each record type mostly keyed by TLID.
I have had problems doing a v.reclass on v.tig.in.basic data (TLID are too
big for C operations) so I build by own reclass map in perl and write
.att and .cat file directly.

Hmm, don't know why TLID's are too big. If an "int" == "long", then it
should accomodate numbers up to 2147483647. Are there TLID's that
exceed this?

As a map company I class all chains that say are roads by name in a separate
category so I deal with each and every name. Thus I deal with categories
that number in the 10,000s or more.

You should rewrite the Tiger routines! You obviously know TIGER files
better than I...

--
Eric G. Miller <egm2@jps.net>