Re: The continuing saga of trying to get TG to work . . .

Chris Metzler wrote:
> Thanks for replying Jon.  I was getting ready to repost, or make a
> pathetic plea for suggestions, or something.
> 
> On Fri, 7 Apr 2006 16:13:39 -0000 (Local time zone must be set--see zic 
> 	manual page)
> Jon Stockill wrote:
> 
>>On Thu, March 23, 2006 8:13 am, Chris Metzler wrote:
>>
>>
>>>Finally, time to create the scenery.  The server beeped a bunch,
>>>showing unprintable characters in the location just like hgtchop
>>>had.  When it was done, there was no Scenery/ directory.  In the
>>>Status/ directory though, there were a zillion files of the form
>>>(tile_number).failed, one for every file in the region.
>>
>>All the beeping sounds very much like you're dumping data that should be
>>going to a file to stdout instead. That would explain the lack of
>>output.
> 
> 
> I understand what you're saying, but I don't know if that's right.
> It's embedded in the middle of text status messages being written
> to stdout.
> 
> It happens right from the beginning, in fact, when I run hgtchop.
> I'm messing with just one SRTM1 file -- N38W086.hgt.zip -- and
> this is what happens:
> 
> stax:~/Projects/TGSandbox-627> for i in data/srtm-1-US/N*.hgt.zip;
> do ../TerraGear/source/src/Prep/DemChop/hgtchop 1 $i
> work/SRTM-United_States-1/; done Extracting
> data/srtm-1-US/N38W086.hgt.zip to /tmp Archive:
> data/srtm-1-US/N38W086.hgt.zip inflating: /tmp/N38W086.hgt Proceeding
> with /tmp/N38W086.hgt Loading HGT data file: /tmp/N38W086.hgt
>   Name = N38W086.hgt
>   Origin = -309600, 136800
> HGT file spans tile boundaries (ok)
>   dx = 3  dy = 7
> -86:, 38:
> width = 0.25 height = 0.125
> min = -309600,136800  max = -308700,137250
> start_x = 0  span_x = 900
> start_y = 0  span_y = 450
> array_file = work/SRTM-United_States-1//w090n30/w086n38/1548288.arr.gz
> -86:
, 38:
> width = 0.25 height = 0.125
> min = -308700,136800  max = -307800,137250
> start_x = 900  span_x = 900
> start_y = 0  span_y = 450
> array_file = work/SRTM-United_States-1//w090n30/w086n38/1548289.arr.gz
> -86:, 38:
> width = 0.25 height = 0.125
> min = -307800,136800  max = -306900,137250
> start_x = 1800  span_x = 900
> start_y = 0  span_y = 450
> array_file = work/SRTM-United_States-1//w090n30/w086n38/1548290.arr.gz
> 		[ rest omitted ]
> 
> I don't know if they're coming over OK in this email, but you should
> see/hear some non-alphanumeric characters in there.  I don't know what
> that's about; but if I proceed like it's normal and go all the way
> through to running fgfs-tools-server and fgfs-tools-client, I get stuff
> like this:
> 
> stax:~/Projects/TGSandbox/work-502>
> ~/Projects/TerraGear/source/src/BuildTiles/Parallel/fgfs-tools-server . .
> w090n30 Work base: . Output base: .
> Area width: 10 degrees
> Area height: 10 degrees
> lons = w090 lats = n30
> start_lon = -90  start_lat = 30
> socket is connected to port = 32786
> starting pass = 1 with lat = 30.0625 lon = -89.875
> 
> [ skip to tiles from the 1deg x 1deg area SRTM1 I d/l'd ]
> 
> Bucket = -86:, 38: (1)
> client returned = -1548288
> logged bad tile = 1548288
> Bucket = -85:, 38: (1)
> client returned = -1548290
> logged bad tile = 1548290
> [ snip ]
> Bucket = -86:, 38: (1)
> client returned = -1548304
> logged bad tile = 1548304
> [ rest snipped ]
> 
> Note the extra non-alphanumeric crap in there.  The status directory
> fills with ".failed" files -- some of which I would expect to see, for
> the portions of the 10deg x 10deg area for which I do not have DEM
> data.  But not the ones for which I do.
> 
> I have no idea what to do at this point.  I've been playing with
> this for three weeks now -- never mind airports or lakes or roads,
> just trying to build a single 1deg x 1deg area with landcover alone,
> and get weirdness.
> 
> Any suggestions?  I could sure use them.

Try redirecting the output to a file. I suspect you're seeing a mix of 
stdout (the data) and stderr (the status messages). Seperating the two 
may give you more idea what's going on.

Jon

_______________________________________________
Terragear-devel mailing list
[email protected]
http://mail.flightgear.org/mailman/listinfo/terragear-devel
2f585eeea02e2c79d7b1d8c4963bae2d