Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-10 Thread Tom Russo
On Sun, Jun 10, 2007 at 05:25:28PM -0400, we recorded a bogon-computron 
collision of the <[EMAIL PROTECTED]> flavor, containing:
>  Hi Tate and Curt,
> 
>  I have Xastir working without rtree currently.  It does fine except it does 
>  take a little while on some redraws.
> 
>  I tried updating to the developer's codebase to see if it was something 
>  unusual in the intermediate releases.  I also deleted the files in the rtree 
>  directory and having cvs replace them.  Alas, nothing has quite worked to 
>  fix the rtree problem.  I could try wiping the whole Xastir src directory 
>  and updating that.  It shouldn't work, but who knows?
> 
>  I've also re-ran the Fedora system updater to see if anything obvious was 
>  old or out of date.
> 
>  Now I did install Shapelib, but it is also included in the Xastir install 
>  now?  Could that be a problem - i.e. should I try removing Shapelib?

I really doubt that cleaning out the xastir code and replacing shapelib will
have any effect on this issue, which appears to be a linker bogon.

I just downloaded the 2.8GB Fedora 7 DVD ISO to my work machine and will
burn a DVD on Monday --- then I'll make a virtual machine running F7 and
see if I can reproduce and diagnose the problem.  I don't know that I can
do anything since the linker error is completely meaningless to me, but 
maybe I can track it down.

>  Two other things I've noticed on unrelated topics:
>  1) There's a ton of .geo radar files out there that simply don't work 
>  because they point to a nonexistent web location at NOAA/NWS.  I guess NOAA 
>  updated their system a while ago (I've been out of aprs for a couple of 
>  years.)  I was able to relocate the correct file and fix the geo file for my 
>  area.  The Xastir wiki and readme.maps still points to these download 
>  locations with the outdated .geo's.

Sign up for a Wiki account and drop us a line --- we'll get you set up with
"wikieditor" privileges so you can fix the wiki.  If you send a patch to 
README.MAPS we can commit the updated info, too.

>  2) My day job involves working with GIS, so I was using ArcGIS to export a 
>  geotiff.  I am using PAMAP data, which is beautiful georeferenced RGB color 
>  tif files with 1-2 foot resolution.  I sure would like one in color, but 
>  ArcGIS can only export 24 Bit True color, 8 Bit Palette, or 8 Bit grayscale. 
>   Xastir can't interpret the 24 bit color, and the 8 bit palette must create 
>  a custom CLUT that Xastir doesn't interpret.  In 8 bit palette mode, the 
>  data exists, comes through georeferenced perfectly, as I can see the 
>  patterns, but the colors are seemingly randomly chosen and assigned.

You can use the GDAL tools to turn the 24 bit color images to 8 bit color
that Xastir can use.  Use "rgb2pct.py" (which requires gdal to be built with
python support).

I have had very bad experiences with ArcGIS-generated GeoTIFFs, because it
seems that ArcGIS doesn't generate proper GeoTIFF metadata tags.  Every time
I've ever gotten GeoTIFFs generated by ESRI products I've had to massage them
with GDAL to get them readable by Xastir or GRASS.  Your milage may vary
(i.e., perhaps the latest versions of ArcGIS are better).

-- 
Tom RussoKM5VY   SAR502   DM64ux  http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://kevan.org/brain.cgi?DDTNM
"And, isn't sanity really just a one-trick pony anyway? I mean all you get is
 one trick, rational thinking, but when you're good and crazy, oooh, oooh,
 oooh, the sky is the limit!"  --- The Tick
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-10 Thread Stephen Peters

Hi Tate and Curt,

I have Xastir working without rtree currently.  It does fine except  
it does take a little while on some redraws.


I tried updating to the developer's codebase to see if it was  
something unusual in the intermediate releases.  I also deleted the  
files in the rtree directory and having cvs replace them.  Alas,  
nothing has quite worked to fix the rtree problem.  I could try  
wiping the whole Xastir src directory and updating that.  It  
shouldn't work, but who knows?


I've also re-ran the Fedora system updater to see if anything obvious  
was old or out of date.


Now I did install Shapelib, but it is also included in the Xastir  
install now?  Could that be a problem - i.e. should I try removing  
Shapelib?


Two other things I've noticed on unrelated topics:
1) There's a ton of .geo radar files out there that simply don't work  
because they point to a nonexistent web location at NOAA/NWS.  I  
guess NOAA updated their system a while ago (I've been out of aprs  
for a couple of years.)  I was able to relocate the correct file and  
fix the geo file for my area.  The Xastir wiki and readme.maps still  
points to these download locations with the outdated .geo's.


2) My day job involves working with GIS, so I was using ArcGIS to  
export a geotiff.  I am using PAMAP data, which is beautiful  
georeferenced RGB color tif files with 1-2 foot resolution.  I sure  
would like one in color, but ArcGIS can only export 24 Bit True  
color, 8 Bit Palette, or 8 Bit grayscale.  Xastir can't interpret the  
24 bit color, and the 8 bit palette must create a custom CLUT that  
Xastir doesn't interpret.  In 8 bit palette mode, the data exists,  
comes through georeferenced perfectly, as I can see the patterns, but  
the colors are seemingly randomly chosen and assigned.


Thanks again guys for making this great software!  It really is awesome!

-Steve



On Jun 10, 2007, at 3:34 PM, Tate Belden wrote:


Curt Mills wrote:

On Sat, 9 Jun 2007, Tate Belden wrote:

Do you have the X11-devel packages installed?
Do you have the rtree-devel package installed?

Good guess on rtree, but all the rtree code is inside Xastir
sources, so there are no extra packages to download.  Same case with
the the Shapefile code these days.


Thanks, that's what it was - a guess.

Steve, you got it working yet?



___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-10 Thread Tate Belden

Curt Mills wrote:

On Sat, 9 Jun 2007, Tate Belden wrote:


Do you have the X11-devel packages installed?
Do you have the rtree-devel package installed?


Good guess on rtree, but all the rtree code is inside Xastir
sources, so there are no extra packages to download.  Same case with
the the Shapefile code these days.



Thanks, that's what it was - a guess.

Steve, you got it working yet?

___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-09 Thread Curt Mills
On Fri, 8 Jun 2007, Tom Russo wrote:

> Without the rtrees, every time you zoom or pan a map, all the shapefiles
> in it are searched linearly for shapes that overlap the screen.  With rtrees,
> a fast lookup can be done to access only those shapes that overlap without
> searching through the entire file.  It can make things a lot faster if you're
> doing a lot of zooming and panning.

Unless you're zoomed out most or all of the way for the Shapefile
anyway, in which case it's just as fast to load the whole Shapefile
without rtree support at all.

If you're using very large (area) Shapefiles or are zoomed in a lot,
rtree support _really_ makes rendering faster.

-- 
Curt, WE7U. archer at eskimo dot com
http://www.eskimo.com/~archer
  Lotto:  A tax on people who are bad at math. - unknown
Windows:  Microsoft's tax on computer illiterates. - WE7U.
The world DOES revolve around me:  I picked the coordinate system!"
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-09 Thread Curt Mills
On Sat, 9 Jun 2007, Tate Belden wrote:

> Do you have the X11-devel packages installed?
> Do you have the rtree-devel package installed?

Good guess on rtree, but all the rtree code is inside Xastir
sources, so there are no extra packages to download.  Same case with
the the Shapefile code these days.

-- 
Curt, WE7U. archer at eskimo dot com
http://www.eskimo.com/~archer
  Lotto:  A tax on people who are bad at math. - unknown
Windows:  Microsoft's tax on computer illiterates. - WE7U.
The world DOES revolve around me:  I picked the coordinate system!"
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-09 Thread Richard Polivka, N6NKO
I have a fresh install of Fedora 7 production. Refer to previous emails 
about 1.9 and Fedora 7.


I downloaded the CVS devel package and compiled. It runs with no issues 
or problems short of a couple of nuisance warnings during compile.


It sounds like this is a bad install of Fedora 7.

73 from 807,

Richard, N6NKO




Tom Russo wrote:

On Sat, Jun 09, 2007 at 01:14:47AM -0600, we recorded a bogon-computron collision of 
the <[EMAIL PROTECTED]> flavor, containing:
  

 Stephen Peters wrote:


Tate,
I just double-checked the ld.so.conf and re-ran ldconfig.  Neither seems to 
change the error message.  One interesting thing, is that I don't have a 
/usr/X11R6/lib directory.  I do have a /usr/lib/X11 directory, but it's 
empty.  Thank you for the suggestion!

It is also very good to know you've got it running on F7.
When I compile without rtree, it does work fine.  I must have something 
weird going on here with that package.

-Steve
  

 Mmmm, strange, yet odd...

 Do you have the X11-devel packages installed?



If he had not, disabling rtree would not have fixed the problem.

  

 Do you have the rtree-devel package installed?



There is no such thing.  rtree is a library that comes with xastir, not a
package that you can install.

  

___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-09 Thread Tom Russo
On Sat, Jun 09, 2007 at 01:14:47AM -0600, we recorded a bogon-computron 
collision of the <[EMAIL PROTECTED]> flavor, containing:
>  Stephen Peters wrote:
> > Tate,
> > I just double-checked the ld.so.conf and re-ran ldconfig.  Neither seems to 
> > change the error message.  One interesting thing, is that I don't have a 
> > /usr/X11R6/lib directory.  I do have a /usr/lib/X11 directory, but it's 
> > empty.  Thank you for the suggestion!
> > It is also very good to know you've got it running on F7.
> > When I compile without rtree, it does work fine.  I must have something 
> > weird going on here with that package.
> > -Steve
> 
>  Mmmm, strange, yet odd...
> 
>  Do you have the X11-devel packages installed?

If he had not, disabling rtree would not have fixed the problem.

>  Do you have the rtree-devel package installed?

There is no such thing.  rtree is a library that comes with xastir, not a
package that you can install.

-- 
Tom RussoKM5VY   SAR502   DM64ux  http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://kevan.org/brain.cgi?DDTNM
"And, isn't sanity really just a one-trick pony anyway? I mean all you get is
 one trick, rational thinking, but when you're good and crazy, oooh, oooh,
 oooh, the sky is the limit!"  --- The Tick
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-09 Thread Tate Belden

Stephen Peters wrote:

Tate,

I just double-checked the ld.so.conf and re-ran ldconfig.  Neither seems 
to change the error message.  One interesting thing, is that I don't 
have a /usr/X11R6/lib directory.  I do have a /usr/lib/X11 directory, 
but it's empty.  Thank you for the suggestion!


It is also very good to know you've got it running on F7.

When I compile without rtree, it does work fine.  I must have something 
weird going on here with that package.


-Steve



Mmmm, strange, yet odd...

Do you have the X11-devel packages installed?
Do you have the rtree-devel package installed?

Kinda shooting in the dark here, but there's implications of either.

Good luck!
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-08 Thread Stephen Peters

Tate,

I just double-checked the ld.so.conf and re-ran ldconfig.  Neither  
seems to change the error message.  One interesting thing, is that I  
don't have a /usr/X11R6/lib directory.  I do have a /usr/lib/X11  
directory, but it's empty.  Thank you for the suggestion!


It is also very good to know you've got it running on F7.

When I compile without rtree, it does work fine.  I must have  
something weird going on here with that package.


-Steve



On Jun 9, 2007, at 12:30 AM, Tate Belden wrote:

I just installed the current CVS of Xastir on 3 different F7 boxen  
with no issues at all. I followed the Fedora CVS howto on the wiki  
site - step by step and had no problems at all.


I suspect - but don't know for sure - that you haven't modified  
the /etc/ld.so.conf file or not run /sbin/ldconfig as directed in  
the howto(?).


http://www.xastir.org/wiki/index.php/HowTo:Fedora


___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-08 Thread Tate Belden

Stephen Peters wrote:

Group,

I was following the How-to's for Fedora at the Wiki.  I followed most of the 
guidelines except used the GUI Package Manager for more of the packages than 
the instructions called for.  I only had to install libgeotiff from source at 
the command line.  Fedora 7 is brand new - updated to recent packages - with 
nothing else except Xastir.

The bootstrap and configure go well with no warnings or errors.  The final 
report from configure is pasted at the bottom.

Now here's the error.  I think this is almost the last step, so I think I'm 
SOOO close.

Any ideas on what to try?

-Steve


make[3]: Entering directory `/usr/local/src/xastir/src'
gcc  -g -O2 -pipe -W -Wall -Wpointer-arith -Wstrict-prototypes 
-Wno-unused-parameter -pthread -Wl,--no-keep-memory -L/usr/local/lib -lgdal 
-lWand -lMagick -o xastir alert.o awk.o bulletin_gui.o color.o datum.o db.o 
dbfawk.o draw_symbols.o fcc_data.o festival.o geo-find.o geocoder_gui.o gps.o 
hashtable.o hashtable_itr.o hostname.o igate.o interface.o interface_gui.o 
io-common.o io-mmap.o lang.o list_gui.o locate_gui.o location.o location_gui.o 
main.o maps.o map_cache.o map_dos.o map_gdal.o map_geo.o map_gnis.o map_pdb.o 
map_shp.o map_tif.o map_tiger.o map_WMS.o messages.o messages_gui.o objects.o 
popup_gui.o rac_data.o rotated.o rpl_malloc.o shp_hash.o snprintf.o sound.o 
track_gui.o util.o view_message_gui.o wx.o wx_gui.o x_spider.o xa_config.o 
compiledate.o -Lrtree -lrtree  -lXm -lXt -lX11 -lWand -lMagick -ldb-4.5 -lrt 
-lcurl -lXm -lXt -lXp -lXext -lm -lSM -lICE -lX11 -lshp -lpcre -lproj -ltiff 
-lgeotiff -lgdal
/usr/bin/ld: rtree/librtree.a(split_l.o)(.text+0x27f): unresolvable R_386_32 
relocation against symbol `CoverSplit'
/usr/bin/ld: final link failed: Nonrepresentable section on output
collect2: ld returned 1 exit status
make[3]: *** [xastir] Error 1
make[3]: Leaving directory `/usr/local/src/xastir/src'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/usr/local/src/xastir/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/local/src/xastir'
make: *** [all] Error 2
[EMAIL PROTECTED] xastir]# 





configure: creating ./config.status
config.status: creating Makefile
config.status: creating callpass/Makefile
config.status: creating config/Makefile
config.status: creating help/Makefile
config.status: creating m4/Makefile
config.status: creating scripts/Makefile
config.status: creating src/Makefile
config.status: creating src/rtree/Makefile
config.status: creating src/shapelib/Makefile
config.status: creating src/shapelib/contrib/Makefile
config.status: creating symbols/Makefile
config.status: creating xastir.spec
config.status: creating xastir-min.spec
config.status: creating xastir-lsb.spec
config.status: creating config.h
config.status: config.h is unchanged
config.status: executing depfiles commands
===

xastir 1.9.0 has been configured to use the following
options and external libraries:

Building with AX25 . : no
Building with Festival . : yes
Building with GPSMan ... : no
Building with GraphicsMagick/ImageMagick ... : ImageMagick
Building with libproj .. : yes
Building with GeoTiff .. : yes
Building with GDAL/OGR . : yes
Building with ShapeLib . : yes
Building with pcre . : yes
Building with dbfawk ... : yes
Building with map caching .. : yes
Building with rtree indexing ... : yes
--
Building with ErrorPopups (Old Method) . : no
Building with libgc (Debug)  : no
Building with profiling (Debug)  : no
Building with Linux Standard Base .. : no

xastir will be installed in /usr/local/bin.

___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Stephen,

I just installed the current CVS of Xastir on 3 different F7 boxen with 
no issues at all. I followed the Fedora CVS howto on the wiki site - 
step by step and had no problems at all.


I suspect - but don't know for sure - that you haven't modified the 
/etc/ld.so.conf file or not run /sbin/ldconfig as directed in the howto(?).


http://www.xastir.org/wiki/index.php/HowTo:Fedora

Following the howto instructions, at the end of ./configure I get:
Building with AX25 . : no
Building with Festival . : yes
Building with GPSMan ... : no
Building with GraphicsMagick/ImageMagick ... : ImageMagick
Building with libproj .. : yes
Building with GeoTiff .

Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-08 Thread Tom Russo
On Fri, Jun 08, 2007 at 09:07:11PM -0700, we recorded a bogon-computron 
collision of the <[EMAIL PROTECTED]> flavor, containing:
>  Tom,
> 
> Thank you for responding.  I applied the patch that you sent, and it didn't 
> cure the problem.  The same error still results after make.  It was a very 
> clever idea though!  

No, not really.  If I had actually dug before making the suggestion, I'd have
realized that split_q.c isn't used by the rtree library in xastir, even though
it's present in the source tree.  So nothing in it can conflict with split_l.c,
and my suggestion was just bollocks.

So in fact, I have NO idea why FC7 (and apparently FC6, or maybe that was just
FC7 Beta) is doing this.  I don't even know what that error message means.

> So, I reupdated from CVS and configure --without-rtree.  Make then finishes, 
> and the application runs.
> 
> What am I missing without rtree?  Wiklipedia tells me that rtrees are ways of 
> indexing spatial data... Seems like most of Xastir would need that!

Rtrees are used in xastir only to speed up shapefile rendering.

Without the rtrees, every time you zoom or pan a map, all the shapefiles
in it are searched linearly for shapes that overlap the screen.  With rtrees,
a fast lookup can be done to access only those shapes that overlap without
searching through the entire file.  It can make things a lot faster if you're
doing a lot of zooming and panning.

-- 
Tom RussoKM5VY   SAR502   DM64ux  http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://kevan.org/brain.cgi?DDTNM
"And, isn't sanity really just a one-trick pony anyway? I mean all you get is
 one trick, rational thinking, but when you're good and crazy, oooh, oooh,
 oooh, the sky is the limit!"  --- The Tick
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-08 Thread Stephen Peters
 Tom,

Thank you for responding.  I applied the patch that you sent, and it didn't 
cure the problem.  The same error still results after make.  It was a very 
clever idea though!  

So, I reupdated from CVS and configure --without-rtree.  Make then finishes, 
and the application runs.

What am I missing without rtree?  Wiklipedia tells me that rtrees are ways of 
indexing spatial data... Seems like most of Xastir would need that!

-Steve



On Friday, June 08, 2007, at 07:26PM, "Tom Russo" <[EMAIL PROTECTED]> wrote:
>On Fri, Jun 08, 2007 at 02:20:01PM -0700, we recorded a bogon-computron 
>collision of the <[EMAIL PROTECTED]> flavor, containing:
>> Group,
>> 
>> I was following the How-to's for Fedora at the Wiki.  I followed most of the 
>> guidelines except used the GUI Package Manager for more of the packages than 
>> the instructions called for.  I only had to install libgeotiff from source 
>> at the command line.  Fedora 7 is brand new - updated to recent packages - 
>> with nothing else except Xastir.
>> 
>> The bootstrap and configure go well with no warnings or errors.  The final 
>> report from configure is pasted at the bottom.
>> 
>> Now here's the error.  I think this is almost the last step, so I think I'm 
>> SOOO close.
>> 
>> Any ideas on what to try?
>
>None, really.  I remember someone saying they'd seen that error in FC6 or FC7,
>and I tried googling the error message --- but I can't find anything that
>helps me figure out why that's happening.  AFAICT it's only happening
>on that system, I can't reproduce it on any of mine, or I'd possibly be
>able to figure it out.  I have no FC system of any vintage, and insufficient
>time to install FC7 into a VMware virtual machine to do the debugging.
>
>The error is in the rtree library, so a temporary work-around is to disable
>rtree support (by specifying --without-rtree on the configure line for xastir).
>
>One fishy thing that I am looking at in the rtree library is that two 
>different files (split_q.h and split_l.h) declare a variable "CoverSplit" as a 
>global variable.  It's possible that these are interfering with each other 
>somehow (why that's not happening with other systems is beyond me, but it 
>might be the explanation).  If it is what's going on, then this *might* fix 
>the problem:
>
>  edit the file src/rtree/split_q.h and change the line:
>
> struct Rect CoverSplit;
>  to 
> struct Rect qCoverSplit;
>
>then edit the file src/rtree/split_q.c and change all the instances where
>that variable is used to  qCoversplit, too.  
>
>To make it easy, I'm attaching a unified diff that will make those two 
>changes.  If this change makes the error messages go away, then I'll commit 
>the change to the repository.  If not, well, it was an attractive straw to 
>grasp at.
>
>To apply the patch, copy the attached patch file to your src/rtree directory
>and type
>   patch -p0 < splitqdiffs
>it will update both files and then you can recompile.
>
>Please let me know what happens.
>
>-- 
>Tom RussoKM5VY   SAR502   DM64ux  http://www.swcp.com/~russo/
>Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://kevan.org/brain.cgi?DDTNM
>"And, isn't sanity really just a one-trick pony anyway? I mean all you get is
> one trick, rational thinking, but when you're good and crazy, oooh, oooh,
> oooh, the sky is the limit!"  --- The Tick
>
>? splitq.diffs
>? splitqdiffs
>Index: split_q.c
>===
>RCS file: /cvsroot/xastir/xastir/src/rtree/split_q.c,v
>retrieving revision 1.4
>diff -u -u -r1.4 split_q.c
>--- split_q.c  3 Jan 2007 19:38:41 -   1.4
>+++ split_q.c  9 Jun 2007 02:22:35 -
>@@ -60,12 +60,12 @@
>   BranchCount = MAXKIDS(n) + 1;
> 
>   /* calculate rect containing all in the set */
>-  CoverSplit = BranchBuf[0].rect;
>+  qCoverSplit = BranchBuf[0].rect;
>   for (i=1; i   {
>-  CoverSplit = RTreeCombineRect(&CoverSplit, &BranchBuf[i].rect);
>+  qCoverSplit = RTreeCombineRect(&qCoverSplit, 
>&BranchBuf[i].rect);
>   }
>-  CoverSplitArea = RTreeRectSphericalVolume(&CoverSplit);
>+  CoverSplitArea = RTreeRectSphericalVolume(&qCoverSplit);
> 
>   RTreeInitNode(n);
> }
>Index: split_q.h
>===
>RCS file: /cvsroot/xastir/xastir/src/rtree/split_q.h,v
>retrieving revision 1.4
>diff -u -u -r1.4 split_q.h
>--- split_q.h  3 Jan 2007 19:38:41 -   1.4
>+++ split_q.h  9 Jun 2007 02:22:35 -
>@@ -39,7 +39,7 @@
> 
> struct Branch BranchBuf[MAXCARD+1];
> int BranchCount;
>-struct Rect CoverSplit;
>+struct Rect qCoverSplit;
> RectReal CoverSplitArea;
> 
> /* variables for finding a partition */
>
>
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-08 Thread Stephen Peters
 Tom,

Thank you for responding.  I applied the patch that you sent, and it didn't 
cure the problem.  The same error still results after make.  It was a very 
clever idea though!  

So, I reupdated from CVS and configure --without-rtree.  Make then finishes, 
and the application runs.

What am I missing without rtree?  Wiklipedia tells me that rtrees are ways of 
indexing spatial data... Seems like most of Xastir would need that!

-Steve



On Friday, June 08, 2007, at 07:26PM, "Tom Russo" <[EMAIL PROTECTED]> wrote:
>On Fri, Jun 08, 2007 at 02:20:01PM -0700, we recorded a bogon-computron 
>collision of the <[EMAIL PROTECTED]> flavor, containing:
>> Group,
>> 
>> I was following the How-to's for Fedora at the Wiki.  I followed most of the 
>> guidelines except used the GUI Package Manager for more of the packages than 
>> the instructions called for.  I only had to install libgeotiff from source 
>> at the command line.  Fedora 7 is brand new - updated to recent packages - 
>> with nothing else except Xastir.
>> 
>> The bootstrap and configure go well with no warnings or errors.  The final 
>> report from configure is pasted at the bottom.
>> 
>> Now here's the error.  I think this is almost the last step, so I think I'm 
>> SOOO close.
>> 
>> Any ideas on what to try?
>
>None, really.  I remember someone saying they'd seen that error in FC6 or FC7,
>and I tried googling the error message --- but I can't find anything that
>helps me figure out why that's happening.  AFAICT it's only happening
>on that system, I can't reproduce it on any of mine, or I'd possibly be
>able to figure it out.  I have no FC system of any vintage, and insufficient
>time to install FC7 into a VMware virtual machine to do the debugging.
>
>The error is in the rtree library, so a temporary work-around is to disable
>rtree support (by specifying --without-rtree on the configure line for xastir).
>
>One fishy thing that I am looking at in the rtree library is that two 
>different files (split_q.h and split_l.h) declare a variable "CoverSplit" as a 
>global variable.  It's possible that these are interfering with each other 
>somehow (why that's not happening with other systems is beyond me, but it 
>might be the explanation).  If it is what's going on, then this *might* fix 
>the problem:
>
>  edit the file src/rtree/split_q.h and change the line:
>
> struct Rect CoverSplit;
>  to 
> struct Rect qCoverSplit;
>
>then edit the file src/rtree/split_q.c and change all the instances where
>that variable is used to  qCoversplit, too.  
>
>To make it easy, I'm attaching a unified diff that will make those two 
>changes.  If this change makes the error messages go away, then I'll commit 
>the change to the repository.  If not, well, it was an attractive straw to 
>grasp at.
>
>To apply the patch, copy the attached patch file to your src/rtree directory
>and type
>   patch -p0 < splitqdiffs
>it will update both files and then you can recompile.
>
>Please let me know what happens.
>
>-- 
>Tom RussoKM5VY   SAR502   DM64ux  http://www.swcp.com/~russo/
>Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://kevan.org/brain.cgi?DDTNM
>"And, isn't sanity really just a one-trick pony anyway? I mean all you get is
> one trick, rational thinking, but when you're good and crazy, oooh, oooh,
> oooh, the sky is the limit!"  --- The Tick
>
>? splitq.diffs
>? splitqdiffs
>Index: split_q.c
>===
>RCS file: /cvsroot/xastir/xastir/src/rtree/split_q.c,v
>retrieving revision 1.4
>diff -u -u -r1.4 split_q.c
>--- split_q.c  3 Jan 2007 19:38:41 -   1.4
>+++ split_q.c  9 Jun 2007 02:22:35 -
>@@ -60,12 +60,12 @@
>   BranchCount = MAXKIDS(n) + 1;
> 
>   /* calculate rect containing all in the set */
>-  CoverSplit = BranchBuf[0].rect;
>+  qCoverSplit = BranchBuf[0].rect;
>   for (i=1; i   {
>-  CoverSplit = RTreeCombineRect(&CoverSplit, &BranchBuf[i].rect);
>+  qCoverSplit = RTreeCombineRect(&qCoverSplit, 
>&BranchBuf[i].rect);
>   }
>-  CoverSplitArea = RTreeRectSphericalVolume(&CoverSplit);
>+  CoverSplitArea = RTreeRectSphericalVolume(&qCoverSplit);
> 
>   RTreeInitNode(n);
> }
>Index: split_q.h
>===
>RCS file: /cvsroot/xastir/xastir/src/rtree/split_q.h,v
>retrieving revision 1.4
>diff -u -u -r1.4 split_q.h
>--- split_q.h  3 Jan 2007 19:38:41 -   1.4
>+++ split_q.h  9 Jun 2007 02:22:35 -
>@@ -39,7 +39,7 @@
> 
> struct Branch BranchBuf[MAXCARD+1];
> int BranchCount;
>-struct Rect CoverSplit;
>+struct Rect qCoverSplit;
> RectReal CoverSplitArea;
> 
> /* variables for finding a partition */
>
>
___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


Re: [Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-08 Thread Tom Russo
On Fri, Jun 08, 2007 at 02:20:01PM -0700, we recorded a bogon-computron 
collision of the <[EMAIL PROTECTED]> flavor, containing:
> Group,
> 
> I was following the How-to's for Fedora at the Wiki.  I followed most of the 
> guidelines except used the GUI Package Manager for more of the packages than 
> the instructions called for.  I only had to install libgeotiff from source at 
> the command line.  Fedora 7 is brand new - updated to recent packages - with 
> nothing else except Xastir.
> 
> The bootstrap and configure go well with no warnings or errors.  The final 
> report from configure is pasted at the bottom.
> 
> Now here's the error.  I think this is almost the last step, so I think I'm 
> SOOO close.
> 
> Any ideas on what to try?

None, really.  I remember someone saying they'd seen that error in FC6 or FC7,
and I tried googling the error message --- but I can't find anything that
helps me figure out why that's happening.  AFAICT it's only happening
on that system, I can't reproduce it on any of mine, or I'd possibly be
able to figure it out.  I have no FC system of any vintage, and insufficient
time to install FC7 into a VMware virtual machine to do the debugging.

The error is in the rtree library, so a temporary work-around is to disable
rtree support (by specifying --without-rtree on the configure line for xastir).

One fishy thing that I am looking at in the rtree library is that two 
different files (split_q.h and split_l.h) declare a variable "CoverSplit" as a 
global variable.  It's possible that these are interfering with each other 
somehow (why that's not happening with other systems is beyond me, but it 
might be the explanation).  If it is what's going on, then this *might* fix 
the problem:

  edit the file src/rtree/split_q.h and change the line:

 struct Rect CoverSplit;
  to 
 struct Rect qCoverSplit;

then edit the file src/rtree/split_q.c and change all the instances where
that variable is used to  qCoversplit, too.  

To make it easy, I'm attaching a unified diff that will make those two 
changes.  If this change makes the error messages go away, then I'll commit 
the change to the repository.  If not, well, it was an attractive straw to 
grasp at.

To apply the patch, copy the attached patch file to your src/rtree directory
and type
   patch -p0 < splitqdiffs
it will update both files and then you can recompile.

Please let me know what happens.

-- 
Tom RussoKM5VY   SAR502   DM64ux  http://www.swcp.com/~russo/
Tijeras, NM  QRPL#1592 K2#398  SOC#236 AHTB#1 http://kevan.org/brain.cgi?DDTNM
"And, isn't sanity really just a one-trick pony anyway? I mean all you get is
 one trick, rational thinking, but when you're good and crazy, oooh, oooh,
 oooh, the sky is the limit!"  --- The Tick
? splitq.diffs
? splitqdiffs
Index: split_q.c
===
RCS file: /cvsroot/xastir/xastir/src/rtree/split_q.c,v
retrieving revision 1.4
diff -u -u -r1.4 split_q.c
--- split_q.c   3 Jan 2007 19:38:41 -   1.4
+++ split_q.c   9 Jun 2007 02:22:35 -
@@ -60,12 +60,12 @@
BranchCount = MAXKIDS(n) + 1;
 
/* calculate rect containing all in the set */
-   CoverSplit = BranchBuf[0].rect;
+   qCoverSplit = BranchBuf[0].rect;
for (i=1; i___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir


[Xastir] New FC7 install troubles with both Release and Stable CVS

2007-06-08 Thread Stephen Peters
Group,

I was following the How-to's for Fedora at the Wiki.  I followed most of the 
guidelines except used the GUI Package Manager for more of the packages than 
the instructions called for.  I only had to install libgeotiff from source at 
the command line.  Fedora 7 is brand new - updated to recent packages - with 
nothing else except Xastir.

The bootstrap and configure go well with no warnings or errors.  The final 
report from configure is pasted at the bottom.

Now here's the error.  I think this is almost the last step, so I think I'm 
SOOO close.

Any ideas on what to try?

-Steve


make[3]: Entering directory `/usr/local/src/xastir/src'
gcc  -g -O2 -pipe -W -Wall -Wpointer-arith -Wstrict-prototypes 
-Wno-unused-parameter -pthread -Wl,--no-keep-memory -L/usr/local/lib -lgdal 
-lWand -lMagick -o xastir alert.o awk.o bulletin_gui.o color.o datum.o db.o 
dbfawk.o draw_symbols.o fcc_data.o festival.o geo-find.o geocoder_gui.o gps.o 
hashtable.o hashtable_itr.o hostname.o igate.o interface.o interface_gui.o 
io-common.o io-mmap.o lang.o list_gui.o locate_gui.o location.o location_gui.o 
main.o maps.o map_cache.o map_dos.o map_gdal.o map_geo.o map_gnis.o map_pdb.o 
map_shp.o map_tif.o map_tiger.o map_WMS.o messages.o messages_gui.o objects.o 
popup_gui.o rac_data.o rotated.o rpl_malloc.o shp_hash.o snprintf.o sound.o 
track_gui.o util.o view_message_gui.o wx.o wx_gui.o x_spider.o xa_config.o 
compiledate.o -Lrtree -lrtree  -lXm -lXt -lX11 -lWand -lMagick -ldb-4.5 -lrt 
-lcurl -lXm -lXt -lXp -lXext -lm -lSM -lICE -lX11 -lshp -lpcre -lproj -ltiff 
-lgeotiff -lgdal
/usr/bin/ld: rtree/librtree.a(split_l.o)(.text+0x27f): unresolvable R_386_32 
relocation against symbol `CoverSplit'
/usr/bin/ld: final link failed: Nonrepresentable section on output
collect2: ld returned 1 exit status
make[3]: *** [xastir] Error 1
make[3]: Leaving directory `/usr/local/src/xastir/src'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/usr/local/src/xastir/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/local/src/xastir'
make: *** [all] Error 2
[EMAIL PROTECTED] xastir]# 




configure: creating ./config.status
config.status: creating Makefile
config.status: creating callpass/Makefile
config.status: creating config/Makefile
config.status: creating help/Makefile
config.status: creating m4/Makefile
config.status: creating scripts/Makefile
config.status: creating src/Makefile
config.status: creating src/rtree/Makefile
config.status: creating src/shapelib/Makefile
config.status: creating src/shapelib/contrib/Makefile
config.status: creating symbols/Makefile
config.status: creating xastir.spec
config.status: creating xastir-min.spec
config.status: creating xastir-lsb.spec
config.status: creating config.h
config.status: config.h is unchanged
config.status: executing depfiles commands
===

xastir 1.9.0 has been configured to use the following
options and external libraries:

Building with AX25 . : no
Building with Festival . : yes
Building with GPSMan ... : no
Building with GraphicsMagick/ImageMagick ... : ImageMagick
Building with libproj .. : yes
Building with GeoTiff .. : yes
Building with GDAL/OGR . : yes
Building with ShapeLib . : yes
Building with pcre . : yes
Building with dbfawk ... : yes
Building with map caching .. : yes
Building with rtree indexing ... : yes
--
Building with ErrorPopups (Old Method) . : no
Building with libgc (Debug)  : no
Building with profiling (Debug)  : no
Building with Linux Standard Base .. : no

xastir will be installed in /usr/local/bin.

___
Xastir mailing list
Xastir@xastir.org
http://lists.xastir.org/cgi-bin/mailman/listinfo/xastir