[Freeciv-Dev] Re: [PATCH] Map cleanups (PR#1208)

["Ross W. Wetmore" <rwetmore@xxxxxxxxxxxx>]

I lost you here. This is the only part that looks like a "request", 
but it is confusing.

At 12:39 PM 02/01/07 +0100, Raimar Falke wrote:
>On Mon, Jan 07, 2002 at 12:31:10AM -0500, Ross W. Wetmore wrote:
>> At 03:45 PM 02/01/06 +0100, Raimar Falke wrote:
>> >On Sun, Jan 06, 2002 at 04:32:02AM -0800, jdorje@xxxxxxxxxxxxxxxxxxxxx
wrote:
>> >> Ross W. Wetmore wrote:
>> >> 
>> >> > This series of patches cleans up a number of cosmetic and real issues
>> >> > in map.c and map.h.
[...] 
>> Much of this changes with the topology fixes, since it is these few basic
>> elements that form the valid interface, and the implementation here is
where 
>> the core topology changes will largely go. But first one needs to get the 
>> interfaces in so the code can be updated to use them and there is something
>> to build on.
>> 
>> Note, this is in the 5-10% range of easy things to get out of the way.
>
>I agree that the goal is ok. However I had some problems with the
>iterators of an old corecleanup version. So I have to take a look at
>the current version.
[...]
>The change in mapclean_0* is ok. You have to make a proposal for the
>further changes (patch or description).
[...]
>> In summary, there are 2 minor fixes (will do in next pass), 4 major
religious 
>> wars and one skirmish to be fought over two comments and three functional 
>> changes, and one previously agreed on and requested feature Raimar now
wants 
>> dropped.
>> 
>> Does this summarize the current status?
>
>Besides the requested feature: yes it looks so.
>
>       Raimar
>-- 
> email: rf13@xxxxxxxxxxxxxxxxx
> "The very concept of PNP is a lovely dream that simply does not translate to
>  reality. The confusion of manually doing stuff is nothing compared to the
>  confusion of computers trying to do stuff and getting it wrong, which they
>  gleefully do with great enthusiasm." 
>    -- Jinx Tigr in the SDM

If you mean the map_clean_0 comment and map_clean is what is up for 
discussion and being summarized I thought, map_clean is one patch. It is 
split into three sections to make it easier for people to distinguish 
non-functional or purely local changes from the more serious parts. 
There is no such things as map_clean_0, but there is a current patch up for 
consideration, with a fairly complete listing of all its elements. There
is nothing more to be "requested" or no new patch submitted here as far as 
I can tell. The current one just needs to be dealt with, hopefully in a
way that will prove fruitful to all, but especially the codebase.

If you mean a proposal or patch request for future patches, as in the
various iterator change, then you need to spell out a bit more of what 
you are looking for or set some priorities. 

<ASIDE>
Note I carefully excluded the rotational direction system from this patch, 
so map.c and map.h are still significantly divergent from the corecleanups 
:-).

Look at places like this to see what ordering they use for their direction
enums. It is 2/3 of the way down.

This also shows how the coordinate system for native isometric is typically
handled. It corresponds to the term "compressed native isometric" in the
corecleanups.

http://www.gamedev.net/reference/articles/article747.asp
</ASIDE>

To state some obvious points ...

There will probably be 20-25 patches to merge the corecleanups with CVS.
There were 13 in August of which only 5 were even partially looked at, 
and none of the rotational or topology sections were ever reached. It
bogged down in the bugfixes and preliminaries like the current patch.
Including the 3 month straightest_dir() bugfix debate.

I believe you recently stated that the issue with corecleanups was not
getting patches. If you are waiting for these, there are probably 3-4
areas that can be run in parallel, so you will get a few every week
until things bog down and it becomes clear the effort is going to be as
useful as the earlier one.

The contents of the corecleanups have been available and elements of them
sometimes discussed. This is a working prototype with general documentation
in the code to explain much of what is being done in various parts. This 
is slightly more than a proposal tends to provide. But I assure you the
read is probably far more useful. 

There is an outstanding Freeciv TODO for the cleanups and the topology, 
particularly the isometric parts. Consider this as a combined submission 
to deal with them.

I really would *like* to see discussion of various aspects thrashed out
in advance, then the final patches submitted primarily for code review
and sanity testing before going into CVS. It is a mistake to turn every
patch into a from scratch dogfight bringing every past issue that can be
tied to it into the process as part of a general ongoing guerilla war :-).

This means discussion and requirements precede final implementation - but 
this is not really part of Freeciv policy, just as significant updates or
alternate devpaths are generally killed in favour of small nitfixes. It
will be interesting to see how this goes :-).

Cheers,
RossW
=====