[Advanced-java] GIS, autocad, dxf? List Archives?

[Steven J. Owens]

Hi folks,

     I need a little advice on working with either autocad or dxf
files for GIS purposes.

     Also, I tried to check the list archives for this topic, at
http://lists.xcf.berkeley.edu/lists/advanced-java/, but any search I
entered got me an ht://Dig error, "Unable to read configuration file".


     I'm working on a pro-bono project for a non-profit, to build a
fairly simple GIS system of a campground, roughly three miles by three
miles.  They run a fairly complex event yearly, with a lot of campers.
A surveyor has been donating his time (I guess he got tired of the
painfully inaccurate maps :-), surveying the campground and entering
the data all into Autocad.  

     What I need to do is make a simple GIS web application to enable
the prospective campers to examine the campsite in detail, and figure
out where they want to camp.  I'm thinking in terms of something
fairly straightforward; they view the entire map in one screen, click
on a section, get a zoom-in, click again, get an even closer zoom-in,
and can generally pan around a bit.

     The web aspects of this are fairly straightforward.  The part I'm
concerned with is parsing the autocad or dxf files, and generating the
structured data, as well as the images.  

     I've never worked with autocad, or with GIS-oriented stuff, and
I'm not sure what to expect to find in the data.  The surveyor is not
a technologically oriented kind of guy, and I'm not even sure if he
could tell me if he's using some specific autocad map format or
add-on, or not.

     To illustrate how I need to use this data:

     I don't want to generate each map view dynamically, since I don't
really have requirements that demand it.  It seems to make a lot more
sense to pre-generate the map images.  Take the map image, lay a 10x10
grid over it, generate individual image files of each grid square
square at double the magnification.  Repeat the process on each grid
square.

     The user gets a page with a single big image in a client-side
image map, with the image map hot spots defined in the same 10x10
grid.  They click on a hotspot, their browser loads the appropriate
zoomed-in grid square image, with its own set of 10x10 hotspots.  I
can even pre-generate the client-side image map tags, though I might
keep them in a simple database just for the sake of flexibility.

     I might not use a 10x10 ratio, come to think of it, that might
be too much of a zoom jump for usability.  Perhaps 4x4 at each level,
and do three or more levels of zoom.

     To make this all work, I need to be able to open the autocad file
the surveyor sends me, and get individual block outlines.

     Specifically, I need to ultimately come up with a series of X,Y
coordinate points that define each block outline.  These X,Y
coordinate points need to be relative to the 0,0 for each individual
chunk, though if I can start by getting the outlines on the main,
whole map, I can at least modify from there.

     I also need a label that identifies the block.  Supposedly the
block outlines can be labeled, and I think he's labelling his, but
looking at the sample files I have so far, I'm not seeing them.  I
don't know if there's a "standard" format for this labelling, or if
he's just squeezing the label into some sort of attached text property
on each block.

     I also need, at some point, to generate the chunked map images,
but in a pinch, I can probably get the surveyor to do that.  If I can
find a tool that parses the file formats and extracts the data, I
suspect it'll have provisions for generating map images as well.

-- 
Steven J. Owens
puff@darksleep.com

"I'm going to make broad, sweeping generalizations and strong,
 declarative statements, because otherwise I'll be here all night and
 this document will be four times longer and much less fun to read.
 Take it all with a grain of salt." - Me at http://darksleep.com