[Xastir] map problem on MEPIS

[Tom Russo]

On Sat, Feb 06, 2010 at 08:03:38PM -0500, we recorded a bogon-computron collision of the <kk4r at cox.net> flavor, containing:
> 
> Libraries
>   I built the proj.4 library as suggested, but skipped libgeotiff, since 
> it sounded unnecessary for me.
>   In retrospect,  it would have been good to do it -- very easy, why not?

There's no point in building proj.4 and not libgeotiff --- Xastir uses proj.4
only for its geotiff code.  

> Install xastir
>   I got the expected results, except GeoTiff and GDAL capabilities are 
> not available.

GDAL is pretty much unnecessary.  Xastir's use of GDAL is currently so limited 
that you will probably never benefit from it.  It allows use of some esoteric
map types, but only a few of them look good --- only those that have been
special cased, such as raw TIGER/Line files.  And those are so slow you might
as well use the TIGER shapefiles instead.

> festival
>   running "festival --server &" fails due to "no default voice found ..."

You need to configure festival and add a voice package.  Can't give quick 
advice there, though.

> xastir
>   running xastir seems to work great, but there are some messages at the 
> console
>   see below
> 
> Indexing maps...
> draw_geo_image_map:sscanf parsing error
> draw_geo_image_map:sscanf parsing error
> draw_geo_image_map:sscanf parsing error
> draw_geo_image_map:sscanf parsing error

Must be a .geo file with bad info in it.

> Finished indexing maps
> *** Reading WX Alert log files
> *** Done with WX Alert log files
> convert_from_xastir_coordinates:X out-of-range (too 
> low):18446744073677686016
> convert_from_xastir_coordinates:X out-of-range (too high):161465600

This means that some map tried to draw out of range.  It happens most of the 
time when you're zoomed so that the entire world is visible.  But it can
also happen if you have a bad map.

What map are you displaying?  At what zoom level?

> My suggestions for the MEPIS instructions on the web would be to 
> resolve  which automake should be used (there were several versions 
> available).  Apparently something more needs to be done for festival, 
> which may be obvious, but I've never even heard of festival, so it will 
> require a little detective work.  It seems like the bleeding edge 
> version is recommended over STABLE.  Why wouldn't STABLE be recommended 
> for the masses?

The main reason is because there's only one developer who does the releases,
and we've not been doing stable releases as often as there used to be.  That's 
because Curt has to do a lot of work to kick 'em out the door, and he's been 
busy with his day job and non-open-source life.  He hasn't even had time to do 
the development snapshots that used to be done in between releases.  There was 
about a year between release 1.9.4 and 1.9.6, though 1.9.8 came out a few 
months later.

Since Stable releases aren't happening with any real frequency, if you want bug
fixes you pretty much have to track CVS.

-- 
Tom Russo    KM5VY   SAR502   DM64ux          http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236        http://kevan.org/brain.cgi?DDTNM
  In some cultures what I do would be considered normal. 
                                  -- Ineffective daily affirmation