[GRASS5] Final additions to 5.4.0

Last Saturday I merged a few changes from the previous week in the 5.3 CVS HEAD into the 5.4 release branch. A few bugfixes and I added the new tcltkgrass modules that Michael put in, as I thought they couldn't do any harm.

There've been a lot of changes to the 5.3 CVS HEAD though this week, which really makes it quite confusing. Unless anybody makes a special case for including any of these, I will leave them out, and tidy up the INSTALL and NEWS.html and release 5.4.0 next week. If there were any simple bugfixes necessary for compilation on all platforms that are unlikely to have any side effects, they could go in. But there's a good few substantial-looking changes that will have to stay out.

I thought a good plan would be for the version of all files in the released version of 5.4.0 to be those that would be directly moved into the 5.7 CVS by Bernhard. Is there any reason not to do this?

Paul

Last call (almost) for changes to NEWS.html: see current version below.
To make changes and corrections, check out from CVS with the releasebranch sticky tag:
cvs co -r releasebranch_5_4 grass/NEWS.html
is one way to do this (in a different directory from the normal CVS HEAD version of grass).

What's new in GRASS 5.4.0 compared to 5.3.0

[GRASS 5.4.0 released xx November 2004]

Modules fixed/updated (major changes only)

      d.zoom: fixes (Hamish Bowman)
      g.mlist: sort output (Hamish Bowman)
      nviz: some bugfixes (Glynn Clements, Hamish Bowman)
      PNGdriver: can now also write PPM images and optionally supports incremental writing of PNG after each display
      operation (Glynn Clements)
      ps.map: floating point legend fixes (Hamish Bowman)
      rgb.hsv.sh: bugfix (Glynn Clements)
      r.buffer: nautical miles support (Hamish Bowman)
      r.colors: new format for rules files (Hamish Bowman, Glynn Clements)
      r.univar: faster C implementation now operational and script version renamed to r.univar.sh (Hamish Bowman)
      v.in.garmin.sh: now operational (Hamish Bowman)

Source Code:

      New modules now built by default: d.info, d.nviz, r.grow2, r.in.mat, r.out.mat, r.series
      Alternate Build system now default (Glynn Clements, Paul Kelly)
      Support for zlib compression in integer/CELL maps (GRASS_INT_ZLIB environment variable) (Glynn Clements)
      Cygwin and Mac OSX shared library compilation working (Paul Kelly)
      Longer names for database, location and mapset allowed in start-up screen (Markus Neteler, Glynn Clements)
      Doxygen comments in libraries for improved Programmers' Manual (Markus Neteler)
      Raster map titles standardised to 1024 characters in all relevant modules and libraries (Hamish Bowman)
      Various compiler warnings fixed (Glynn Clements)
      r3.mapcalc: now built using 2-D r.mapcalc code instead of duplicate version (Glynn Clements)
      g.pnmcomp: Added for internal use (Glynn Clements)

On Sat, Oct 30, 2004 at 01:45:07PM +0100, Paul Kelly wrote:

Last Saturday I merged a few changes from the previous week in the 5.3 CVS
HEAD into the 5.4 release branch. A few bugfixes and I added the new
tcltkgrass modules that Michael put in, as I thought they couldn't do any
harm.

Thanks for this!

There've been a lot of changes to the 5.3 CVS HEAD though this week, which
really makes it quite confusing. Unless anybody makes a special case for
including any of these, I will leave them out, and tidy up the INSTALL and
NEWS.html and release 5.4.0 next week. If there were any simple bugfixes
necessary for compilation on all platforms that are unlikely to have any
side effects, they could go in. But there's a good few substantial-looking
changes that will have to stay out.

No opinion here. I think 5.4.0 should not be postponed (there could be
5.4.1 if needed).

I thought a good plan would be for the version of all files in the
released version of 5.4.0 to be those that would be directly moved into
the 5.7 CVS by Bernhard. Is there any reason not to do this?

This is not a good idea. In fact the files from 5.3-CVS-Head should
be copied as 5.7-CVS depends on them. As you have written above, 5.4.0
does not contain all recent changes (some of them were done for 5.7-CVS).
This would mean extra work to merge in the missing changes.

Markus