FIX: City borders are not set correctly

[jamespetts]

FIX: City borders are not set correctly

[jamespetts]

Thank you very much for your observation - that is interesting. May I
ask - on which branch have you observed this?

James.

On 01/03/2014 11:34, dschr wrote:

Hello!
Thanks for simutrans-experimental, this is a great improvement on
simutrans.
To the issue:
I have observed that city borders are not set correctly, which leads
to an asymmetrical city growth, i.e. some overland streets are full of
houses after a short while, while other, far closer parts of the city,
does not evolve at all.
I also observed that the townhall is often in one corner of the
city-border-box.

Greetings

—
Reply to this email directly or view it on GitHub
#15.

[dschr]

I cloned simutrans experimental two days ago from the git repository and compiled it on a Linux 64 bit machine. I already saw this behaviour at the beginning of january 2014.

[dschr]

Oh and i used the british experimental pakset and the given configuration files from here:http://bridgewater-brunel.me.uk/downloads/simutrans-experimental-config.zip

[dschr]

[jamespetts]

Thank you very much for the details - that is helpful. I am having some
trouble reproducing this, however, albeit I use a Windows machine. I
assume that you are using the master branch, which corresponds to the
latest release?

On 01/03/2014 12:06, dschr wrote:

bildschirmfoto - 01 03 2014 - 13 05 11
https://f.cloud.github.com/assets/6823951/2301304/d1484158-a139-11e3-9057-cad5c4bab6d5.png

—
Reply to this email directly or view it on GitHub
#15 (comment).

[dschr]

Yes, i checked out the master branch.

[jamespetts]

Hmm, very odd indeed. Can you try the 112.5-merge branch to see whether
that is any different? I am not quite sure why you are seeing this issue
whereas I am not.

On 01/03/2014 13:39, dschr wrote:

Yes, i checked out the master branch.

—
Reply to this email directly or view it on GitHub
#15 (comment).

[dschr]

112.5-merge won't compile on my machine and gives out the following error while compiling:
if(!gb || gb->get_is_factory() && gb->get_fabrik() == NULL)
^
common.mk:50: recipe for target 'build/default/simworld.o' failed
make: *** [build/default/simworld.o] Error 1

[dschr]

`--> make -v
GNU Make 4.0
Built for x86_64-unknown-linux-gnu

[jamespetts]

Thank you very much to that. I will forward this to Bernd, who is the
one who has done most of the work in merging the latest Standard
releases into Experimental for this branch, who hopefully will be able
to solve it;

James.

On 01/03/2014 14:03, dschr wrote:

112.5-merge won't compile on my machine and gives out the following
error while compiling:
if(!gb || gb->get_is_factory() && gb->get_fabrik() == NULL)
^
common.mk:50: recipe for target 'build/default/simworld.o' failed
make: *** [build/default/simworld.o] Error 1

—
Reply to this email directly or view it on GitHub
#15 (comment).

[dschr]

Thank you!
If you need some further information or more details, feel free to ask

Greetings

[BerndGabriel]

@dschr: were you able to solve the issue in the meantime? If not, what is the compiler's error message? Which gcc do you use? I had no trouble compiling with gcc 4.5.2 for windows x86 and x64 (for x64 it finds some other errors regaring pointer vs integer sizes and printf formats)

[dschr]

Hello Bernd,

no, unfortunately the border-issue is not solved yet. However i am able to compile 112.5-merge now properly, but the program aborts on execution. The error message is the following:

Use work dir /home/daniel/sim/default/
Reading low level config data ...
Parsed simuconf.tab for directory layout; multiuser = 1
FATAL ERROR: loadsave_t::start_tag() - expected "env_t", got "umgebool>
0
true
383
264
128
0
true
4
-16
<32>"
Aborting program execution ...

For help with this error or to file a bug report please see the Simutrans forum at
http://forum.simutrans.com
zsh: abort (core dumped) /home/daniel/sim/default/simutrans-experimental

I compiled it with:
gcc-Version 4.8.2 20140206 (prerelease) (GCC)

[jamespetts]

All that you need to do is delete settings-experimental.xml or
settings-experimental-debug.xml.

On 08/03/2014 13:14, dschr wrote:

Hello Bernd,

no, unfortunately the border-issue is not solved yet. However i am
able to compile 112.5-merge now properly, but the program aborts on
execution. The error message is the following:

Use work dir /home/daniel/sim/default/
Reading low level config data ...
Parsed simuconf.tab for directory layout; multiuser = 1
FATAL ERROR: loadsave_t::start_tag() - expected "env_t", got "umgebool>
0
true
383
264
128
0
true
4
-16
"
Aborting program execution ...

For help with this error or to file a bug report please see the
Simutrans forum at
http://forum.simutrans.com
zsh: abort (core dumped) /home/daniel/sim/default/simutrans-experimental

I compiled it with:
gcc-Version 4.8.2 20140206 (prerelease) (GCC)

—
Reply to this email directly or view it on GitHub
#15 (comment).

[dschr]

where can i find these files? I searched through my simutrans-directory for xml-files, but there were none.

[jamespetts]

They are not in the simutrans directory, but in your home directory.

[dschr]

Thank you James for your response, i were able to execute it now.
So what i could observe is that on a map with one city, the city borders are set correctly. I weren't able to test it for more than one city, as the program got stuck while initialising the map if i chose more than one city to create. As it works for one city, i suggest that at least the border issue is solved.

[jamespetts]

Which branch are you using? The way-improvements branch should allow you
to generate multiple cities without hanging.

On 08/03/2014 16:48, dschr wrote:

Thank you James for your response, i were able to execute it now.
So what i could observe is that on a map with one city, the city
borders are set correctly. I weren't able to test it for more than one
city, as the program got stuck while initialising the map if i chose
more than one city to create. As it works for one city, i suggest that
at least the border issue is solved.

—
Reply to this email directly or view it on GitHub
#15 (comment).

[dschr]

Hello,
yes, with the way-improvement branch i was able to create maps with more than one city and the city borders are as they should be. Great work!

[jamespetts]

Splendid - thank you for checking. I am planning on merging the
way-improvements branch back into devel-new shortly.

James.

On 09/03/2014 09:57, dschr wrote:

Hello,
yes, with the way-improvement branch i was able to create maps with
more than one city and the city borders are as they should be. Great work!

—
Reply to this email directly or view it on GitHub
#15 (comment).