Chris, could you try to convert head and dbln files from UNIX to Mac
(newline characters)?
Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.
I also updated and rebuilt from cvs this afternoon, with the same result.
Chris
> Chris, could you try to convert head and dbln files from UNIX to Mac
> (newline characters)?Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.
For the record, Mac OS/X newlines *are* UNIX newlines.
esc abbr octal dec ctrl-char
UNIX \n LF \012 10 ^J
MacOSX \n LF \012 10 ^J
DOS \r\n CR LF \015\012 13 10 ^M^J
MacOS9* \r CR \015 13 ^M
[*] Mac OS9 and earlier versions of Mac OS
I have also seen text which came with \n\r, presumably from buggy
software, either as badly formed DOS or MacOS9 format.
Some software takes it upon itself to translate text files on download
to the local format, causing all sorts of mayhem. (e.g. Internet
Explorer, some FTP & email software) MacOSX versions of Internet
Explorer and MS Office actually automatically translate text files into
the *old* Mac format. I'll leave it up to the masses to decide if this
is a backwards-compatibility decision or a bug.
for more info see the man pages for: dos2unix, unix2dos, file, tr
Hamish
On Mon, 7 Mar 2005 12:19:44 +1300, Hamish <hamish_nospam@yahoo.com> wrote:
> > Chris, could you try to convert head and dbln files from UNIX to Mac
> > (newline characters)?
>
> Radim, I changed the line endings to Mac, but I still received a bus
> error on loading the layer.For the record, Mac OS/X newlines *are* UNIX newlines.
Some software takes it upon itself to translate text files on download
to the local format, causing all sorts of mayhem. (e.g. Internet
Explorer, some FTP & email software) MacOSX versions of Internet
Explorer and MS Office actually automatically translate text files into
the *old* Mac format. I'll leave it up to the masses to decide if this
is a backwards-compatibility decision or a bug.
The only thing I used to edit these files is vim. Otherwise, it went
straight from GRASS on Linux to GRASS on OSX.
Chris Fonnesbeck wrote:
Chris, could you try to convert head and dbln files from UNIX to Mac
(newline characters)?Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.I also updated and rebuilt from cvs this afternoon, with the same result.
Does the error appeares with all vectors?
The vectors mooved from Mac to Linux work?
Can you send here the end of debug (g.gisenv set=DEBUG=5 before module)?
Radim
Chris Fonnesbeck wrote:
Chris, could you try to convert head and dbln files from UNIX to Mac
(newline characters)?Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.I also updated and rebuilt from cvs this afternoon, with the same result.
Anybody successfully used files created on Linux on Mac?
Can users with access to more platforms to test that?
Radim
Could someone running Grass on a Linux machine post files somewhere, and one of us Mac users could test them? I'd be happy to try a few and see if they work on my powerbook.
Nick Cahill
On Mar 7, 2005, at 3:01 AM, Radim Blazek wrote:
Chris Fonnesbeck wrote:
Chris, could you try to convert head and dbln files from UNIX to Mac
(newline characters)?Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.
I also updated and rebuilt from cvs this afternoon, with the same result.Anybody successfully used files created on Linux on Mac?
Can users with access to more platforms to test that?Radim
On Mon, 07 Mar 2005 09:58:40 +0100, Radim Blazek <blazek@itc.it> wrote:
Chris Fonnesbeck wrote:
>>Chris, could you try to convert head and dbln files from UNIX to Mac
>>(newline characters)?
>
> Radim, I changed the line endings to Mac, but I still received a bus
> error on loading the layer.
>
> I also updated and rebuilt from cvs this afternoon, with the same result.Does the error appeares with all vectors?
The vectors mooved from Mac to Linux work?
I have never successfully imported this particular vector into Mac, as
it takes way too long (see previous thread). I just tried moving a
simple vector layer to Linux from Mac and got a segmentation fault.
Can you send here the end of debug (g.gisenv set=DEBUG=5 before module)?
Here are the last several lines
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_line()
D5/5: line type 4 -> 8
D3/5: dig_Rd_P_line()
D5/5: line type -1 -> 0
Bus error
Chris Fonnesbeck wrote:
On Mon, 07 Mar 2005 09:58:40 +0100, Radim Blazek <blazek@itc.it> wrote:
D3/5: dig_Rd_P_line()
D5/5: line type 4 -> 8
D3/5: dig_Rd_P_line()
D5/5: line type -1 -> 0
Bus error
The type '-1' is wrong, it should be 1-8. The problem could be in grass51/lib/vector/Vlib/read_nat.c Vect__Read_line_nat():
/* first byte: 0 bit: 1 - alive, 0 - dead
* 1 bit: 1 - categories, 0 - no category
* 2-3 bit: store type
* 4-5 bit: reserved for store type expansion
* 6-7 bit: not used
*/
....
char rhead, nc;
....
rhead >>= 2;
type = dig_type_from_store ( (int) rhead );
the line can be deleted:
rhead &= 0xFE;
but I don't know where is the problem. Maybe rhead should be unsigned?
Radim
> Anybody successfully used files created on Linux on Mac?
> Can users with access to more platforms to test that?Could someone running Grass on a Linux machine post files somewhere,
and one of us Mac users could test them? I'd be happy to try a few andsee if they work on my powerbook.
Well, start with the Spearfish57 data, this will be created on Linux I
think.
http://grass.ibiblio.org/sampledata/spearfish_grass57data.tar.gz
(I've had trouble with this before, IIRC, had to export to shapefile &
reimport on Mac with v.in.ogr. This was a while back now, so no good
confirmation)
Hamish
Hamish wrote:
Anybody successfully used files created on Linux on Mac?
Can users with access to more platforms to test that?Could someone running Grass on a Linux machine post files somewhere, and one of us Mac users could test them? I'd be happy to try a few and
see if they work on my powerbook.
Well, start with the Spearfish57 data, this will be created on Linux I
think.http://grass.ibiblio.org/sampledata/spearfish_grass57data.tar.gz
Yes.
(I've had trouble with this before, IIRC, had to export to shapefile &
reimport on Mac with v.in.ogr. This was a while back now, so no good
confirmation)
Why don't you report this problem?
Radim
Chris Fonnesbeck wrote:
On Mon, 07 Mar 2005 09:58:40 +0100, Radim Blazek <blazek@itc.it> wrote:
Chris Fonnesbeck wrote:
Chris, could you try to convert head and dbln files from UNIX to Mac
(newline characters)?Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.I also updated and rebuilt from cvs this afternoon, with the same result.
Does the error appeares with all vectors?
The vectors mooved from Mac to Linux work?I have never successfully imported this particular vector into Mac, as
it takes way too long (see previous thread). I just tried moving a
simple vector layer to Linux from Mac and got a segmentation fault.Can you send here the end of debug (g.gisenv set=DEBUG=5 before module)?
Here are the last several lines
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_line()
D5/5: line type 4 -> 8
D3/5: dig_Rd_P_line()
D5/5: line type -1 -> 0
Bus error
Can you try to run v.category type=point on the vector on Linux and then copy the result to Mac if it works? The problem could be dead line, but I don't know why. Thanks.
Radim
Are you sure the file is not corrupted? Try md5sum on 'coor'.
Radim
Chris Fonnesbeck wrote:
On Mon, 07 Mar 2005 09:58:40 +0100, Radim Blazek <blazek@itc.it> wrote:
Chris Fonnesbeck wrote:
Chris, could you try to convert head and dbln files from UNIX to Mac
(newline characters)?Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.I also updated and rebuilt from cvs this afternoon, with the same result.
Does the error appeares with all vectors?
The vectors mooved from Mac to Linux work?I have never successfully imported this particular vector into Mac, as
it takes way too long (see previous thread). I just tried moving a
simple vector layer to Linux from Mac and got a segmentation fault.Can you send here the end of debug (g.gisenv set=DEBUG=5 before module)?
Here are the last several lines
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_line()
D5/5: line type 4 -> 8
D3/5: dig_Rd_P_line()
D5/5: line type -1 -> 0
Bus error
Sorry, ignore previous 2 mails (v.category and md5sum coor).
It fails when it is reading topo file, please try to:
1. run v.build on the file on Mac
2. check 'topo' file with md5sum on both platforms
Thanks
Radim
Chris Fonnesbeck wrote:
On Mon, 07 Mar 2005 09:58:40 +0100, Radim Blazek <blazek@itc.it> wrote:
Chris Fonnesbeck wrote:
Chris, could you try to convert head and dbln files from UNIX to Mac
(newline characters)?Radim, I changed the line endings to Mac, but I still received a bus
error on loading the layer.I also updated and rebuilt from cvs this afternoon, with the same result.
Does the error appeares with all vectors?
The vectors mooved from Mac to Linux work?I have never successfully imported this particular vector into Mac, as
it takes way too long (see previous thread). I just tried moving a
simple vector layer to Linux from Mac and got a segmentation fault.Can you send here the end of debug (g.gisenv set=DEBUG=5 before module)?
Here are the last several lines
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_node()
D3/5: dig_node_alloc_line(): add = 1
D3/5: dig_Rd_P_line()
D5/5: line type 4 -> 8
D3/5: dig_Rd_P_line()
D5/5: line type -1 -> 0
Bus error
Radim Blazek wrote:
Hamish wrote:
Anybody successfully used files created on Linux on Mac?
Can users with access to more platforms to test that?Could someone running Grass on a Linux machine post files somewhere, and one of us Mac users could test them? I'd be happy to try a few and
see if they work on my powerbook.
Well, start with the Spearfish57 data, this will be created on Linux I
think.http://grass.ibiblio.org/sampledata/spearfish_grass57data.tar.gz
Yes.
Nobody using Mac is capable to downolad the mapset and say if it works or not?
Radim
>>>> Anybody successfully used files created on Linux on Mac?
>>>> Can users with access to more platforms to test that?
I have just tested. Everything seems to work ok for me.
Linux/i686: 6.0.0cvs (2 days old)
Mac/G4: 6.0beta1 (Lorenzo's binaries + libraries)
downloaded spearfish57 onto Mac. Areas, lines, and points maps all
display fine.
on linux, made tar.gz of vector/ and dbf/ dirs and unpacked in a mapset
of the same projection. Areas, lines, and points maps all display &
query fine.
didn't try Mac -> Linux.
I can only guess that the problems have to do with transfering between
current 6.0 on linux and older 5.7 binaries on Mac or vice versa (solved
by v.build.all). OR transfering via a web browser or FTP program that
modifies newlines.
??
I guess if spearfish57 vectors really didn't work on Mac, we would have
heard about it by now. hopefully...?
AFAICT it's good to go Radim.
can anybody else test?
Hamish
I just tried the spearfish data and it worked fine for me. Vectors (lines, areas, points) all display; queries OK; v.build ran fine. Are there any other testing issues? Using powerbook G4, Lorenzo's 6.0b2 binaries.
Nick Cahill
On Mar 9, 2005, at 2:57 AM, Radim Blazek wrote:
Radim Blazek wrote:
Hamish wrote:
Anybody successfully used files created on Linux on Mac?
Can users with access to more platforms to test that?Could someone running Grass on a Linux machine post files somewhere, and one of us Mac users could test them? I'd be happy to try a few and
see if they work on my powerbook.
Well, start with the Spearfish57 data, this will be created on Linux I
think.http://grass.ibiblio.org/sampledata/spearfish_grass57data.tar.gz
Yes.
Nobody using Mac is capable to downolad the mapset and say if it works or not?
Radim
On Wednesday 09 March 2005 12:57 am, Radim Blazek wrote:
Radim Blazek wrote:
> Hamish wrote:
>>>> Anybody successfully used files created on Linux on Mac?
>>>> Can users with access to more platforms to test that?
>>>
>>> Could someone running Grass on a Linux machine post files somewhere,
>>> and one of us Mac users could test them? I'd be happy to try a few and
>>>
>>> see if they work on my powerbook.
>>
>> Well, start with the Spearfish57 data, this will be created on Linux I
>> think.
>>
>> http://grass.ibiblio.org/sampledata/spearfish_grass57data.tar.gz
>
> Yes.Nobody using Mac is capable to downolad the mapset and say if it works
or not?
works for me on a G5 running the latest GRASS6 + libs package on MacOX10.3.8
--
Dylan Beaudette
Soils and Biogeochemistry Graduate Group
University of California at Davis
530.754.7341
Fine here - PB G$, grass60 built from source CVS snapshot 2005-2-26.
I wonder if there is a filesystem issue? I wonder if the person with the problem (sorry, didn't keep the early emails on this) is using UFS instead of the more normal (for Macs) HFS+? Either for the application and/or the data. (I'm using HFS+) I see occassional reports of issues with UFS, and I had a strange problem once with postgre where it would only compile on a UFS disk, until the programmers did something to fix it.
On Mar 9, 2005, at 2:57 AM, Radim Blazek wrote:
Radim Blazek wrote:
Hamish wrote:
Anybody successfully used files created on Linux on Mac?
Can users with access to more platforms to test that?Could someone running Grass on a Linux machine post files somewhere, and one of us Mac users could test them? I'd be happy to try a few and
see if they work on my powerbook.
Well, start with the Spearfish57 data, this will be created on Linux I
think.http://grass.ibiblio.org/sampledata/spearfish_grass57data.tar.gz
Yes.
Nobody using Mac is capable to downolad the mapset and say if it works or not?
Radim
-----
William Kyngesburye <kyngchaos@charter.net>
http://webpages.charter.net/kyngchaos/
"Time is an illusion - lunchtime doubly so."
- Ford Prefect
Thanks to all who tested the portability.
Radim
Hamish wrote:
Anybody successfully used files created on Linux on Mac?
Can users with access to more platforms to test that?I have just tested. Everything seems to work ok for me.
Linux/i686: 6.0.0cvs (2 days old)
Mac/G4: 6.0beta1 (Lorenzo's binaries + libraries)downloaded spearfish57 onto Mac. Areas, lines, and points maps all
display fine.on linux, made tar.gz of vector/ and dbf/ dirs and unpacked in a mapset
of the same projection. Areas, lines, and points maps all display &
query fine.didn't try Mac -> Linux.
I can only guess that the problems have to do with transfering between
current 6.0 on linux and older 5.7 binaries on Mac or vice versa (solved
by v.build.all). OR transfering via a web browser or FTP program that
modifies newlines.
??I guess if spearfish57 vectors really didn't work on Mac, we would have
heard about it by now. hopefully...?AFAICT it's good to go Radim.
can anybody else test?
Hamish