[GRASS5] Setting a removal date for LZW

GRASS GRASS developer
1

I'd like to set a date in the next week or two where we make the switch
from using LZW to DEFLATE in reading/writing FP rasters and G3D files.

I've been using a version of libgis compiled with the new libz wrapper
functions for the last couple weeks and have not had any problem with
floating point rasters. I have not yet tested it with G3D, though I
don't expect any real problems since the new functions should be drop in
replacements for the old and only need to know where to start
reading/writing and how many bytes. I will try to start testing G3D
stuff in the next week, but I have very little experience using these
files and associated programs. One or two of the r3 modules will
have to be changed with regards to renaming the option to use LZW to
DEFLATE.

While I've had no problems, the only way to really know these functions
are working correctly is to try them out under various operating
conditions by a wider number of users (than 1). I don't forsee any
problems provide zlib is present, since these functions hardly do
anything too tricky (the library takes care of the hard part).

Ciao,
--
Eric G. Miller <egm2@jps.net>

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

2

Hi Eric,

On Wed, Dec 06, 2000 at 06:35:21PM -0800, Eric G . Miller wrote:

I'd like to set a date in the next week or two where we make the switch
from using LZW to DEFLATE in reading/writing FP rasters and G3D files.

Yes, good idea. As we have the startup problems fixed, I will be
a tester for you!

I've been using a version of libgis compiled with the new libz wrapper
functions for the last couple weeks and have not had any problem with
floating point rasters. I have not yet tested it with G3D, though I
don't expect any real problems since the new functions should be drop in
replacements for the old and only need to know where to start
reading/writing and how many bytes. I will try to start testing G3D
stuff in the next week, but I have very little experience using these
files and associated programs. One or two of the r3 modules will
have to be changed with regards to renaming the option to use LZW to
DEFLATE.

I am using G3D, so I'll test that as well.

While I've had no problems, the only way to really know these functions
are working correctly is to try them out under various operating
conditions by a wider number of users (than 1). I don't forsee any
problems provide zlib is present, since these functions hardly do
anything too tricky (the library takes care of the hard part).

Shall I wait or directly convert?

Ciao
Markus

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

3

On Thu, Dec 07, 2000 at 09:02:06AM +0000, Markus Neteler wrote:

Hi Eric,

On Wed, Dec 06, 2000 at 06:35:21PM -0800, Eric G . Miller wrote:
> I'd like to set a date in the next week or two where we make the switch
> from using LZW to DEFLATE in reading/writing FP rasters and G3D files.
Yes, good idea. As we have the startup problems fixed, I will be
a tester for you!

Okay, I'd like to propose Dec. 16 or 17 for change. If we do it much
later, I won't be available for the week of Dec. 21-28, so if there are
problems, I won't be able to do anything until around Jan. 1.

> I've been using a version of libgis compiled with the new libz wrapper
> functions for the last couple weeks and have not had any problem with
> floating point rasters. I have not yet tested it with G3D, though I
> don't expect any real problems since the new functions should be drop in
> replacements for the old and only need to know where to start
> reading/writing and how many bytes. I will try to start testing G3D
> stuff in the next week, but I have very little experience using these
> files and associated programs. One or two of the r3 modules will
> have to be changed with regards to renaming the option to use LZW to
> DEFLATE.
I am using G3D, so I'll test that as well.

I think I still have to set up the relevant g3d modules (I forget).

> While I've had no problems, the only way to really know these functions
> are working correctly is to try them out under various operating
> conditions by a wider number of users (than 1). I don't forsee any
> problems provide zlib is present, since these functions hardly do
> anything too tricky (the library takes care of the hard part).

Shall I wait or directly convert?

Please read my previous email on this subject, as it outlines steps you
should take (especially the one's about backing up your data and
uncompressing FP rasters *before* making the switch -- you could also
test out the grass module I wrote for after the fact upgrades).

--
Eric G. Miller <egm2@jps.net>

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'

4

On Thu, Dec 07, 2000 at 07:14:08AM -0800, Eric G . Miller wrote:

On Thu, Dec 07, 2000 at 09:02:06AM +0000, Markus Neteler wrote:
> Hi Eric,
>
> On Wed, Dec 06, 2000 at 06:35:21PM -0800, Eric G . Miller wrote:
> > I'd like to set a date in the next week or two where we make the switch
> > from using LZW to DEFLATE in reading/writing FP rasters and G3D files.
> Yes, good idea. As we have the startup problems fixed, I will be
> a tester for you!

Okay, I'd like to propose Dec. 16 or 17 for change. If we do it much
later, I won't be available for the week of Dec. 21-28, so if there are
problems, I won't be able to do anything until around Jan. 1.

Sounds good to me.
  

> > I've been using a version of libgis compiled with the new libz wrapper
> > functions for the last couple weeks and have not had any problem with
> > floating point rasters. I have not yet tested it with G3D, though I
> > don't expect any real problems since the new functions should be drop in
> > replacements for the old and only need to know where to start
> > reading/writing and how many bytes. I will try to start testing G3D
> > stuff in the next week, but I have very little experience using these
> > files and associated programs. One or two of the r3 modules will
> > have to be changed with regards to renaming the option to use LZW to
> > DEFLATE.
> I am using G3D, so I'll test that as well.

I think I still have to set up the relevant g3d modules (I forget).

> > While I've had no problems, the only way to really know these functions
> > are working correctly is to try them out under various operating
> > conditions by a wider number of users (than 1). I don't forsee any
> > problems provide zlib is present, since these functions hardly do
> > anything too tricky (the library takes care of the hard part).
>
> Shall I wait or directly convert?

Please read my previous email on this subject, as it outlines steps you
should take (especially the one's about backing up your data and
uncompressing FP rasters *before* making the switch -- you could also
test out the grass module I wrote for after the fact upgrades).

For the others: here is the reference:

r.lzw2z.tgz:
http://www.geog.uni-hannover.de/cgi-bin/minorweb.pl?A=READ&L=grass5/2000/11/19/13

Test procedure:
http://www.geog.uni-hannover.de/cgi-bin/minorweb.pl?A=READ&L=grass5/2000/11/28/4

Hope not to be the only testing person (solaris, SGI etc. wanted).

Markus

PS: No complaints yet on beta10... Already published for several hours.

----------------------------------------
If you want to unsubscribe from GRASS Development Team mailing list write to:
minordomo@geog.uni-hannover.de with
subject 'unsubscribe grass5'