rails-devel — List for discussion of Rails development

Re: [Rails-devel] Bugs fixed

[Mark S. <mar...@gm...>]

Erik,

I just went in to replicate the Tile Laying bug in Phase 3 of 1851, and
stumbled across a few other issues:

1. In the Map Display, Cincinati at the top of the board is cut off. I think
the whole position for 1851 is set too high in the Map Frame
2. The L&N Starting Home Location in B16 should be the Northern City, not
the Southern City (based upon the board photo at:
http://boardgamegeek.com/image/126915?size=large.This can be fixed by
updating the CompanyManager.xml File under data/1851, line 38 (for the L&N
Public Company) to be <Home hex=B16", city="1">
3. When attempting to step through the game, I started L&N and GM&O in OR 1,
in SR 2 I sold GM&O so that it's stock value was below L&N, then in OR 2,
L&N Ran first, after completing that run, GM&O was not given a chance to
operate at all.

On Sun, Oct 12, 2008 at 9:58 AM, Erik Vos <eri...@hc...> wrote:

> Having been made aware of some new (well, 5 months old) entries in the bugs
> list, I have gone through that list and made some fixes. Some bug reports I
> cannot reproduce and need further explanation or will be rejected. One or
> two are actually feature requests.
>
> 1945743 Blocked hex sides are not shown on the map.
> True, but this is a feature request, which I have created: 2161464, closing
> this bug report.
> It should be possible to draw thick lines over those hex sides, and IMO
> this
> should be done in a way that those lines also show on top of laid tiles
> (unlike in the real game).
>
> 1945754 Can't see where it shows who has the priority deal.
> Wasn't sure if this is actually a question or a feature request, decided to
> interpret it as both.
> - Answer to the question: On the Game status map ("PD" after the priority
> player name).
> - Request: it has been asked before to also show the PD player in the map
> window (FR 1518278). Then I rejected it, but as it was actually easy to
> implement, I have added it now (also to prevent future requests!).
>
> 1994548 1851 allows negative company treasury amounts.
> I cannot reproduce this problem. Please show how this can be reproduced (if
> still the case).
>
> 1945691 Can't buy trains across from other companies.
> I cannot reproduce this problem. Please show how this can be reproduced (if
> still the case).
>
> 1945678 1851 Company paying double for shares bought from the pool.
> I cannot reproduce this problem. Please show how this can be reproduced (if
> still the case).
>
> 1945675 1851 No multiple yellow track lays from phase 3.
> Confirmed. It has worked in the past, so some recent change must have
> broken
> it.
> As Mark Smith has volunteered to sort out this and other tile laying
> issues,
> I'll leave this one to him.
>
> Fixed one more functional bug:
>
> When a company has a train but no revenue (because no route), the
> allocation
> question (pay/split/withold) is still asked. This question is now
> suppressed
> if the revenue is $0.
>
> A few problems causing the game to abort have also been fixed.
> All fixes have been committed.
>
> Erik Vos
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

[Rails-devel] Bugs fixed

[Erik V. <eri...@hc...>]

Having been made aware of some new (well, 5 months old) entries in the bugs
list, I have gone through that list and made some fixes. Some bug reports I
cannot reproduce and need further explanation or will be rejected. One or
two are actually feature requests.

1945743 Blocked hex sides are not shown on the map.
True, but this is a feature request, which I have created: 2161464, closing
this bug report.
It should be possible to draw thick lines over those hex sides, and IMO this
should be done in a way that those lines also show on top of laid tiles
(unlike in the real game).

1945754 Can't see where it shows who has the priority deal.
Wasn't sure if this is actually a question or a feature request, decided to
interpret it as both.
- Answer to the question: On the Game status map ("PD" after the priority
player name).
- Request: it has been asked before to also show the PD player in the map
window (FR 1518278). Then I rejected it, but as it was actually easy to
implement, I have added it now (also to prevent future requests!).

1994548 1851 allows negative company treasury amounts.
I cannot reproduce this problem. Please show how this can be reproduced (if
still the case).

1945691 Can't buy trains across from other companies.
I cannot reproduce this problem. Please show how this can be reproduced (if
still the case).

1945678 1851 Company paying double for shares bought from the pool.
I cannot reproduce this problem. Please show how this can be reproduced (if
still the case).

1945675 1851 No multiple yellow track lays from phase 3.
Confirmed. It has worked in the past, so some recent change must have broken
it.
As Mark Smith has volunteered to sort out this and other tile laying issues,
I'll leave this one to him.

Fixed one more functional bug:

When a company has a train but no revenue (because no route), the allocation
question (pay/split/withold) is still asked. This question is now suppressed
if the revenue is $0.

A few problems causing the game to abort have also been fixed.
All fixes have been committed.

Erik Vos

Re: [Rails-devel] Infinite Recursion in StockRound.java

[Erik V. <eri...@hc...>]

No, that is another bug that must have crept in recently. I just discovered
it as well, and have fixed it. Not yet committed, though (there is more
coming).
 
It should be: 
 
public int getCurrentPlayerIndex() {

return currentPlayer.getIndex();

}

Erik.


  _____  

From: Mark Smith [mailto:mar...@gm...] 
Sent: Sunday 12 October 2008 14:49
To: Rails Dev Mailing List
Subject: [Rails-devel] Infinite Recursion in StockRound.java


Down in StockRound, I came across what looks a lot like an Infinite
Recursion routine:


 1049     /*----- METHODS TO BE CALLED TO SET UP THE NEXT TURN -----*/


 1050 
 1051     /**


 1052      * @return The index of the player that has the turn.


 1053      */


 1054     public int getCurrentPlayerIndex() {


 1055         return getCurrentPlayerIndex();


 1056     }



What is this supposed to do? It does not make sense they way this is.

Mark

[Rails-devel] Infinite Recursion in StockRound.java

[Mark S. <mar...@gm...>]

Down in StockRound, I came across what looks a lot like an Infinite
Recursion routine:

 1049     /*----- METHODS TO BE CALLED TO SET UP THE NEXT TURN -----*/
 1050
 1051     /**
 1052      * @return The index of the player that has the turn.
 1053      */
 1054     public int getCurrentPlayerIndex() {
 1055         return getCurrentPlayerIndex();
 1056     }

What is this supposed to do? It does not make sense they way this is.

Mark

Re: [Rails-devel] Haven't got the mail list stuff straight

[Mark S. <mar...@gm...>]

I did start the chain to the Rails-devel list...apparently in the process of
replying it did not get all, so the rails devel list address got lost along
the way. I added it back to the "TO" line here.

I can indeed synchronize with the repository and find new changes.

I still need to write up my notes on getting XCode to run well with it.

On Sun, Oct 12, 2008 at 4:37 AM, Erik Vos <eri...@hc...> wrote:

>  BTW, if you want to "earn" commit rights by providing good code
> improvements, you might better provide these to the rails-devel maillist or
> directly to Brett Lentz, as he is managing access. I'm just the guy who
> turns out to have written most of the code....
> If you want, I'm willing to witness 2 points for good patches....
>
> I'll include the missing which_price change later today.
>
> Can you synchronize with the repository, to quickly find new changes?
>
> Erik.
>
>  ------------------------------
> *From:* Mark Smith [mailto:mar...@gm...]
> *Sent:* Sunday 12 October 2008 00:16
>
> *To:* Erik Vos
> *Subject:* Re: [Rails-devel] Haven't got the mail list stuff straight
>
>  Sure go ahead. I don't have commit rights to the CVS yet.
>
> On Sat, Oct 11, 2008 at 5:08 PM, Erik Vos <eri...@hc...> wrote:
>
>>  Mark,
>>
>> That is exactly what I did, except that I completely overlooked the other
>> use of WHICH_START_PRICE under your point 2!
>> I clearly need to regain some routine....   Thanks for pointing that one
>> out.
>> Shall I make that fix too or do you want to try it?
>>
>> Erik.
>>
>>
>>  ------------------------------
>> *From:* Mark Smith [mailto:mar...@gm...]
>> *Sent:* Saturday 11 October 2008 21:18
>> *To:* Erik Vos
>> *Subject:* Re: [Rails-devel] Haven't got the mail list stuff straight
>>
>>   OK, just to provide you my viewpoint of what needs to change, this is
>> what I suggest:
>>
>> 1. in the StartRoundWIndow.java routine,
>>     a. Add line just after setting compName "String playerName =
>> activeItem.getPlayerName();
>>     b. modify the line with "WHICH_START_PRICE" to be
>> "LocalText.getText("WHICH_START_PRICE", new String [] {playerName,
>> compNazme}),
>> 2. in the GameStatus.java routine,
>>    a. Add a line after setting the companyName to null, to be "String
>> playerName = "";
>>    b. Add a line after setting the 'cert' to be "playerName =
>> buy.getPlayerName ();
>>    c. modify the line with "WHICH_START_PRICE" to be
>> "LocalText.getText("WHICH_START_PRICE", new String [] {playerName,
>> compNazme}),"
>> 3.in the LocalisedText.properties, change the WHICH_START_PRICE line to
>> "WHICH_START_PRICE={0} please select the {1} start price:
>>
>> How does that line up with yours?
>>
>> Mark
>>
>> On Sat, Oct 11, 2008 at 12:28 PM, Erik Vos <eri...@hc...> wrote:
>>
>>>
>>> I would suggest a "minor" improvement to the UI when setting a Share
>>> Price...
>>>
>>> In the Dialog Box asking for the Share Price, given it is "Tom" who is
>>> setting the share price, say "Tom set the Share Price for B&O". It would
>>> make it clear in Hot Seat Games, WHO is expected to set the price. I will
>>> look to see if I can determine how to do that as and provide feedback.
>>>
>>> I fully agree. I apologise for already having done it - I overlooked your
>>> last sentence until I had already committed this (very simple) change. I
>>> hope I can restrain myself on your next excellent proposal... ;-)
>>>
>>>
>>
>

Re: [Rails-devel] Recommended Patch to GameManager.java

[Mark S. <mar...@gm...>]

Sometimes it really helps having a fresh pair of eyes look over the code.
:-)

On Sat, Oct 11, 2008 at 5:14 PM, Erik Vos <eri...@hc...> wrote:

>  You're right again. Yesterday I noticed this one too when I restarted the
> game the first time after so many months....
> I really don't understand how this one can have slipped through....
>
> I have fixed it already by initializing specialProperties in PrivateCompany
> to an empty ArrayList rather than to null. Matter of taste, I guess. Mine is
> not the best method if conserving memory is a concern, but there aren't that
> many privates anyway.
>
> Erik.
>
>  ------------------------------
> *From:* Mark Smith [mailto:mar...@gm...]
> *Sent:* Saturday 11 October 2008 21:40
> *To:* Rails Dev Mailing List
> *Subject:* [Rails-devel] Recommended Patch to GameManager.java
>
>  In the process of getting the Rails Devel Source Code to properly load
> and compile and run on my XCode, I came across a problem in the
> setGameParameters that generated a Null Pointer Exception. I feel the best
> way to resolve this is the following change:
>
>   316     private void setGameParameters () {
>   317
>   318         for (PublicCompanyI company : companyManager.getAllPublicCompanies()) {
>   319             hasAnyParPrice = hasAnyParPrice || company.hasParPrice();
>   320             canAnyCompanyBuyPrivates = canAnyCompanyBuyPrivates || company.canBuyPrivates();
>   321             canAnyCompanyHoldShares = canAnyCompanyHoldShares || company.canHoldOwnShares();
>   322         }
>   323
>   324 loop:   for (PrivateCompanyI company : companyManager.getAllPrivateCompanies()) {
>   325             for (SpecialPropertyI sp : company.getSpecialProperties()) {
>   326                 if (sp instanceof SpecialTokenLay
>   327                         && ((SpecialTokenLay)sp).getToken() instanceof BonusToken) {
>   328                     bonusTokensExist = true;
>   329                     break loop;
>   330                 }
>   331             }
>   332
>   333         }
>   334     }
>
> TO:
>
>   316     private void setGameParameters () {
>   317         List<SPecialPropertyI> specials;
>
>   318         for (PublicCompanyI company : companyManager.getAllPublicCompanies()) {
>   319             hasAnyParPrice = hasAnyParPrice || company.hasParPrice();
>   320             canAnyCompanyBuyPrivates = canAnyCompanyBuyPrivates || company.canBuyPrivates();
>   321             canAnyCompanyHoldShares = canAnyCompanyHoldShares || company.canHoldOwnShares();
>   322         }
>   323
>   324 loop:   for (PrivateCompanyI company : companyManager.getAllPrivateCompanies()) {
>
>                     specials = company.getSpecialProperties ();
>                     if (specials != null) {
>   325                 for (SpecialPropertyI sp : specials) {
>   326                     if (sp instanceof SpecialTokenLay
>   327                         && ((SpecialTokenLay)sp).getToken() instanceof BonusToken) {
>   328                         bonusTokensExist = true;
>   329                         break loop;
>   330                     }
>   331                 }
>   332             }
>   333         }
>   334     }
>
> I never really liked try-catch blocks, and in this case if something in the inner loop has a Null Pointer Exception it needs to be thrown. This will test and skip any specials that are null.
>
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>
>

Re: [Rails-devel] Recommended Patch to GameManager.java

[Erik V. <eri...@hc...>]

You're right again. Yesterday I noticed this one too when I restarted the
game the first time after so many months....
I really don't understand how this one can have slipped through....
 
I have fixed it already by initializing specialProperties in PrivateCompany
to an empty ArrayList rather than to null. Matter of taste, I guess. Mine is
not the best method if conserving memory is a concern, but there aren't that
many privates anyway.
 
Erik.


  _____  

From: Mark Smith [mailto:mar...@gm...] 
Sent: Saturday 11 October 2008 21:40
To: Rails Dev Mailing List
Subject: [Rails-devel] Recommended Patch to GameManager.java


In the process of getting the Rails Devel Source Code to properly load and
compile and run on my XCode, I came across a problem in the
setGameParameters that generated a Null Pointer Exception. I feel the best
way to resolve this is the following change:


  316     private void setGameParameters () {


  317 
  318         for (PublicCompanyI company :
companyManager.getAllPublicCompanies()) {


  319             hasAnyParPrice = hasAnyParPrice || company.hasParPrice();


  320             canAnyCompanyBuyPrivates = canAnyCompanyBuyPrivates ||
company.canBuyPrivates();


  321             canAnyCompanyHoldShares = canAnyCompanyHoldShares ||
company.canHoldOwnShares();


  322         }
  323 
  324 loop:   for (PrivateCompanyI company :
companyManager.getAllPrivateCompanies()) {


  325             for (SpecialPropertyI sp : company.getSpecialProperties())
{


  326                 if (sp instanceof SpecialTokenLay
  327                         && ((SpecialTokenLay)sp).getToken() instanceof
BonusToken) {


  328                     bonusTokensExist = true;
  329                     break loop;


  330                 }
  331             }
  332 
  333         }


  334     }

TO:

  316     private void setGameParameters () {


  317         List<SPecialPropertyI> specials;

  318         for (PublicCompanyI company :
companyManager.getAllPublicCompanies()) {


  319             hasAnyParPrice = hasAnyParPrice || company.hasParPrice();


  320             canAnyCompanyBuyPrivates = canAnyCompanyBuyPrivates ||
company.canBuyPrivates();


  321             canAnyCompanyHoldShares = canAnyCompanyHoldShares ||
company.canHoldOwnShares();


  322         }
  323 
  324 loop:   for (PrivateCompanyI company :
companyManager.getAllPrivateCompanies()) {


                    specials = company.getSpecialProperties ();
                    if (specials != null) {
  325                 for (SpecialPropertyI sp : specials) {


  326                     if (sp instanceof SpecialTokenLay
  327                         && ((SpecialTokenLay)sp).getToken() instanceof
BonusToken) {


  328                         bonusTokensExist = true;
  329                         break loop;


  330                     }
  331                 }
  332             }


  333         }
  334     }

I never really liked try-catch blocks, and in this case if something in the
inner loop has a Null Pointer Exception it needs to be thrown. This will
test and skip any specials that are null.

[Rails-devel] Recommended Patch to GameManager.java

[Mark S. <mar...@gm...>]

In the process of getting the Rails Devel Source Code to properly load and
compile and run on my XCode, I came across a problem in the
setGameParameters that generated a Null Pointer Exception. I feel the best
way to resolve this is the following change:

  316     private void setGameParameters () {
  317
  318         for (PublicCompanyI company :
companyManager.getAllPublicCompanies()) {
  319             hasAnyParPrice = hasAnyParPrice || company.hasParPrice();
  320             canAnyCompanyBuyPrivates = canAnyCompanyBuyPrivates
|| company.canBuyPrivates();
  321             canAnyCompanyHoldShares = canAnyCompanyHoldShares ||
company.canHoldOwnShares();
  322         }
  323
  324 loop:   for (PrivateCompanyI company :
companyManager.getAllPrivateCompanies()) {
  325             for (SpecialPropertyI sp : company.getSpecialProperties()) {
  326                 if (sp instanceof SpecialTokenLay
  327                         && ((SpecialTokenLay)sp).getToken()
instanceof BonusToken) {
  328                     bonusTokensExist = true;
  329                     break loop;
  330                 }
  331             }
  332
  333         }
  334     }

TO:

  316     private void setGameParameters () {
  317         List<SPecialPropertyI> specials;

  318         for (PublicCompanyI company :
companyManager.getAllPublicCompanies()) {
  319             hasAnyParPrice = hasAnyParPrice || company.hasParPrice();
  320             canAnyCompanyBuyPrivates = canAnyCompanyBuyPrivates
|| company.canBuyPrivates();
  321             canAnyCompanyHoldShares = canAnyCompanyHoldShares ||
company.canHoldOwnShares();
  322         }
  323
  324 loop:   for (PrivateCompanyI company :
companyManager.getAllPrivateCompanies()) {
                    specials = company.getSpecialProperties ();
                    if (specials != null) {
  325                 for (SpecialPropertyI sp : specials) {
  326                     if (sp instanceof SpecialTokenLay
  327                         && ((SpecialTokenLay)sp).getToken()
instanceof BonusToken) {
  328                         bonusTokensExist = true;
  329                         break loop;
  330                     }
  331                 }
  332             }
  333         }
  334     }

I never really liked try-catch blocks, and in this case if something
in the inner loop has a Null Pointer Exception it needs to be thrown.
This will test and skip any specials that are null.

Re: [Rails-devel] Haven't got the mail list stuff straight

[Erik V. <eri...@hc...>]

I would suggest a "minor" improvement to the UI when setting a Share
Price...

In the Dialog Box asking for the Share Price, given it is "Tom" who is
setting the share price, say "Tom set the Share Price for B&O". It would
make it clear in Hot Seat Games, WHO is expected to set the price. I will
look to see if I can determine how to do that as and provide feedback.

I fully agree. I apologise for already having done it - I overlooked your
last sentence until I had already committed this (very simple) change. I
hope I can restrain myself on your next excellent proposal... ;-)
 
More generally, I think there is a need for a better way to indicate which
player has the turn. Now we only have the yellow background for the player
name.
 
Even more generally, I personally consider the current UI as more or less
provisional. I have worked a lot on it because without a UI there is no game
to play, but my personal emphasis is to get the back-end game engine right.
I'm not a UI designer. A while ago someone proposed to create an
Eclipse-based UI, and we have made room for that by moving all existing UI
code below the ui.swing tree, but nothing has happened since. 

I can understand not wanting to change it again, if you have a solution
ready at hand. It just seemed a bit awkward they way you have it organized
currently.  

Perhaps, yes. As said, it was not easy to accomplish, and this code has gone
through several phases. It's very well possible  certainly don't exclude
that it can be improved, but frankly I don't quite see your point. As far as
I know the price must be set as soon as a player acquires the private
together with the B&O president's share. That is what currently happens.
It's a separate action because the price setting player may be different
from the last acting one (as in the case you reported).

Since I have resolved the issue of getting it to load, compile and run in
debug mode on my Mac OSX, XCode environment, I will write up the steps I
took to get it that far, and include a XCode version of the build.xml (a bit
different from your eclipse version). 

I will look into some of the tile lay bugs reported in the bugs list to see
what could be done about that. This will allow me to understand more about
how your code handles the map, tile lays, promotions, and token placements.


I haven't looked into the bugs list for a while, and I see there's a lot of
new entries (which is good). A while ago I fixed some 18AL tile laying bugs,
but I don't know if that fix covers the reported issue. Please go ahead
looking into those bugs.
 
Meanwhile I will start looking at some of the other reported bugs. I haven't
worked seriously on this project for about half a year, and I'm trying to
collect the courage for a restart. I've found before that fixing some bugs
may well increase my appetite to start off with some new developments....

Erik.

As for incorporating my Map/Tile drawing product, I still need to work out a
detailed plan, but my initial thoughts would be to start by implementing a
version of my "Tile Drawer" that could be done and not impact the existing
map drawing abilities. The only possible downside is that my Tile
Definitions files are somewhat different than those your system uses. And I
do know that you do make use of the TIle Designer results, which I do not.
But there may be a happy middle ground.

Mark  



On Fri, Oct 10, 2008 at 3:02 PM, Erik Vos <eri...@hc...> wrote:


I have fixed the (my) stupid 1851 cash assigment error, and the B&O private
mishandling in 1830 (just two missing statements).

The B&O private was actually one of the hardest things to get right when we
completed 1830,
and it apparently has since fallen victim to one of my many code
reorganizations.
(I'm not really in favour of changing that code yet another time...)

Thanks for the good catch.

Erik.


> -----Original Message-----
> From: Brett Lentz [mailto:wak...@gm...]
> Sent: Thursday 09 October 2008 01:54
> To: Mark Smith
> Cc: rai...@li...
> Subject: Re: [Rails-devel] Haven't got the mail list stuff straight
>
> On Wed, 2008-10-08 at 19:40 -0500, Mark Smith wrote:
> > Brett,
> >
> > For the Rails Developement Mailing List, what is the proper e-mail
> > address to address messages to so that it gets added to the mailing
> > list?
> >
> > I have looked over the Rails Bug List, and have at least one simple
> > fix to the following:
> >
> > In the Bug Reports on Source Forge there is a report [1945662] 1851
> > starting cash wrong. The Solution is straight forward by correcting
> > the 1851 data file under 'data/1851/Game.xml' to update the Players
> > Tags as follows:
> >
> > Currently reads as:
> > 15  <Component name="Bank" class="rails.game.Bank">
> >
> >
> >    16               <Bank amount="8040"/>
> >    17               <Players number="3" cash="400" certLimit="10"/>
> >
> >
> >    18               <Players number="4" cash="430" certLimit="12"/>
> >
> >    19               <Players number="5" cash="500" certLimit="15"/>
> >
> >
> >    20               <Money format="$@"/>
> >    21       </Component>
> >
> >
> > To Be:
> >
> >    15       <Component name="Bank" class="rails.game.Bank">
> >
> >
> >    16               <Bank amount="8040"/>
> >    17               <Players number="3" cash="500" certLimit="10"/>
> >
> >
> >    18               <Players number="4" cash="430" certLimit="12"/>
> >
> >    19               <Players number="5" cash="400" certLimit="15"/>
> >
> >
> >    20               <Money format="$@"/>
> >    21       </Component>
> >
> >
> > Per the Rule Book (or at least per Blackwater Station,
> > http://18xx.net/1851/1851r008.txt) states:
> >
> > The number of Private Companies in the game is equal to the
> number of
> >
> > players.  If there are only three players use the L&O, M&C,
> and ET&WNC.
> > If there is a fourth player, add the O&W.  If there is a
> fifth player,
> >
> > add the CNO&TP.  Set aside any unused Private Companies.
> Distribute the
> >
> > starting capital to each player ($500 for 3 players, $430
> for 4 players,
> > or $400 for 5 players).
> > I have also been looking at the bug in 1830 about the B&O Private
> > Company being sold for $0 whether bought outright, or bought via
> > bidding. If bid on by two or more players,  then the last
> one to pass
> > actually gets B&O for $0, rather then the high bidder.
> Looking at the
> > code it appears that since B&O Public Company needs a share
> price set,
> > it sets that flag, and does not sell B&O Private for the right price
> > to the actual purchaser. And in the Bidding, the PASS
> action sets the
> > flag, rather than assigning it to the highest bidder for the proper
> > bid price. But changing it there to always assign to the proper
> > purchaser does not trigger the Price Setting of the Public Company.
> >
> > Logically to me it makes more sense to set the Share Price when the
> > President's share is transferred to the new President,
> unless there is
> > already a Par Share Price set. Let the Bidding Process upon the last
> > PASS, actually complete the Private Company Sale & Transfer. And if
> > straight purchase (no bid, or single bid) then Sale & Transfer be
> > performed. Only as a "after sale" Possible Action be to set
> President
> > Share transfer, and subsequent Par Value setting occur before the
> > first Stock Round.
> >
> > I am just not familiar enough with the code yet to get all
> the various
> > patches together yet for you. And what I suggest is some significant
> > code refactoring required.
> >
> > Mark Smith
> >
>
> Mark -
>
> The address to use is rai...@li.... I'll cc my
> reply to the list.
>
> The xml fix is fairly straightforward, and should relatively easy to
> generate a diff for that.
>
> As for the B&O issue. I think Erik can speak more knowledgeably about
> how that area of the code works. I suspect that with the current
> refactoring that I'm doing, once I am ready to work on the game code
> itself, I'll have some thoughts on how that stuff should work.
>
>
> ---Brett.
>
> Avoid the Gates of Hell.  Use Linux
>       -- unknown source
>
>

> --------------------------------------------------------------
> -----------
> This SF.Net email is sponsored by the Moblin Your Move
> Developer's challenge
> Build the coolest Linux based applications with Moblin SDK &
> win great prizes
> Grand prize is a trip for two to an Open Source event
> anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100
<http://moblin-contest.org/redirect.php?banner_id=100&url=/> &url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

Re: [Rails-devel] Haven't got the mail list stuff straight

[Erik V. <eri...@hc...>]

I have fixed the (my) stupid 1851 cash assigment error, and the B&O private
mishandling in 1830 (just two missing statements).

The B&O private was actually one of the hardest things to get right when we
completed 1830, 
and it apparently has since fallen victim to one of my many code
reorganizations.
(I'm not really in favour of changing that code yet another time...)

Thanks for the good catch.

Erik.

> -----Original Message-----
> From: Brett Lentz [mailto:wak...@gm...] 
> Sent: Thursday 09 October 2008 01:54
> To: Mark Smith
> Cc: rai...@li...
> Subject: Re: [Rails-devel] Haven't got the mail list stuff straight
> 
> On Wed, 2008-10-08 at 19:40 -0500, Mark Smith wrote:
> > Brett,
> > 
> > For the Rails Developement Mailing List, what is the proper e-mail
> > address to address messages to so that it gets added to the mailing
> > list?
> > 
> > I have looked over the Rails Bug List, and have at least one simple
> > fix to the following:
> > 
> > In the Bug Reports on Source Forge there is a report [1945662] 1851
> > starting cash wrong. The Solution is straight forward by correcting
> > the 1851 data file under 'data/1851/Game.xml' to update the Players
> > Tags as follows:
> > 
> > Currently reads as:
> > 15 	<Component name="Bank" class="rails.game.Bank">
> > 
> > 
> >    16 		<Bank amount="8040"/>
> >    17 		<Players number="3" cash="400" certLimit="10"/>
> > 
> > 
> >    18 		<Players number="4" cash="430" certLimit="12"/>
> > 
> >    19 		<Players number="5" cash="500" certLimit="15"/>
> > 
> > 
> >    20 		<Money format="$@"/>
> >    21 	</Component>
> > 
> > 
> > To Be:
> > 
> >    15 	<Component name="Bank" class="rails.game.Bank">
> > 
> > 
> >    16 		<Bank amount="8040"/>
> >    17 		<Players number="3" cash="500" certLimit="10"/>
> > 
> > 
> >    18 		<Players number="4" cash="430" certLimit="12"/>
> > 
> >    19 		<Players number="5" cash="400" certLimit="15"/>
> > 
> > 
> >    20 		<Money format="$@"/>
> >    21 	</Component>
> > 
> > 
> > Per the Rule Book (or at least per Blackwater Station,
> > http://18xx.net/1851/1851r008.txt) states:
> > 
> > The number of Private Companies in the game is equal to the 
> number of
> > 
> > players.  If there are only three players use the L&O, M&C, 
> and ET&WNC.
> > If there is a fourth player, add the O&W.  If there is a 
> fifth player,
> > 
> > add the CNO&TP.  Set aside any unused Private Companies.  
> Distribute the
> > 
> > starting capital to each player ($500 for 3 players, $430 
> for 4 players,
> > or $400 for 5 players).
> > I have also been looking at the bug in 1830 about the B&O Private
> > Company being sold for $0 whether bought outright, or bought via
> > bidding. If bid on by two or more players,  then the last 
> one to pass
> > actually gets B&O for $0, rather then the high bidder. 
> Looking at the
> > code it appears that since B&O Public Company needs a share 
> price set,
> > it sets that flag, and does not sell B&O Private for the right price
> > to the actual purchaser. And in the Bidding, the PASS 
> action sets the
> > flag, rather than assigning it to the highest bidder for the proper
> > bid price. But changing it there to always assign to the proper
> > purchaser does not trigger the Price Setting of the Public Company.
> > 
> > Logically to me it makes more sense to set the Share Price when the
> > President's share is transferred to the new President, 
> unless there is
> > already a Par Share Price set. Let the Bidding Process upon the last
> > PASS, actually complete the Private Company Sale & Transfer. And if
> > straight purchase (no bid, or single bid) then Sale & Transfer be
> > performed. Only as a "after sale" Possible Action be to set 
> President
> > Share transfer, and subsequent Par Value setting occur before the
> > first Stock Round.
> > 
> > I am just not familiar enough with the code yet to get all 
> the various
> > patches together yet for you. And what I suggest is some significant
> > code refactoring required. 
> > 
> > Mark Smith
> > 
> 
> Mark -
> 
> The address to use is rai...@li.... I'll cc my
> reply to the list. 
> 
> The xml fix is fairly straightforward, and should relatively easy to
> generate a diff for that.
> 
> As for the B&O issue. I think Erik can speak more knowledgeably about
> how that area of the code works. I suspect that with the current
> refactoring that I'm doing, once I am ready to work on the game code
> itself, I'll have some thoughts on how that stuff should work. 
> 
> 
> ---Brett.
> 
> Avoid the Gates of Hell.  Use Linux
> 	-- unknown source
> 
> 
> --------------------------------------------------------------
> -----------
> This SF.Net email is sponsored by the Moblin Your Move 
> Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & 
> win great prizes
> Grand prize is a trip for two to an Open Source event 
> anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
> 

Re: [Rails-devel] Haven't got the mail list stuff straight

[Brett L. <wak...@gm...>]

On Wed, 2008-10-08 at 19:40 -0500, Mark Smith wrote:
> Brett,
> 
> For the Rails Developement Mailing List, what is the proper e-mail
> address to address messages to so that it gets added to the mailing
> list?
> 
> I have looked over the Rails Bug List, and have at least one simple
> fix to the following:
> 
> In the Bug Reports on Source Forge there is a report [1945662] 1851
> starting cash wrong. The Solution is straight forward by correcting
> the 1851 data file under 'data/1851/Game.xml' to update the Players
> Tags as follows:
> 
> Currently reads as:
> 15 	<Component name="Bank" class="rails.game.Bank">
> 
> 
>    16 		<Bank amount="8040"/>
>    17 		<Players number="3" cash="400" certLimit="10"/>
> 
> 
>    18 		<Players number="4" cash="430" certLimit="12"/>
> 
>    19 		<Players number="5" cash="500" certLimit="15"/>
> 
> 
>    20 		<Money format="$@"/>
>    21 	</Component>
> 
> 
> To Be:
> 
>    15 	<Component name="Bank" class="rails.game.Bank">
> 
> 
>    16 		<Bank amount="8040"/>
>    17 		<Players number="3" cash="500" certLimit="10"/>
> 
> 
>    18 		<Players number="4" cash="430" certLimit="12"/>
> 
>    19 		<Players number="5" cash="400" certLimit="15"/>
> 
> 
>    20 		<Money format="$@"/>
>    21 	</Component>
> 
> 
> Per the Rule Book (or at least per Blackwater Station,
> http://18xx.net/1851/1851r008.txt) states:
> 
> The number of Private Companies in the game is equal to the number of
> 
> players.  If there are only three players use the L&O, M&C, and ET&WNC.
> If there is a fourth player, add the O&W.  If there is a fifth player,
> 
> add the CNO&TP.  Set aside any unused Private Companies.  Distribute the
> 
> starting capital to each player ($500 for 3 players, $430 for 4 players,
> or $400 for 5 players).
> I have also been looking at the bug in 1830 about the B&O Private
> Company being sold for $0 whether bought outright, or bought via
> bidding. If bid on by two or more players,  then the last one to pass
> actually gets B&O for $0, rather then the high bidder. Looking at the
> code it appears that since B&O Public Company needs a share price set,
> it sets that flag, and does not sell B&O Private for the right price
> to the actual purchaser. And in the Bidding, the PASS action sets the
> flag, rather than assigning it to the highest bidder for the proper
> bid price. But changing it there to always assign to the proper
> purchaser does not trigger the Price Setting of the Public Company.
> 
> Logically to me it makes more sense to set the Share Price when the
> President's share is transferred to the new President, unless there is
> already a Par Share Price set. Let the Bidding Process upon the last
> PASS, actually complete the Private Company Sale & Transfer. And if
> straight purchase (no bid, or single bid) then Sale & Transfer be
> performed. Only as a "after sale" Possible Action be to set President
> Share transfer, and subsequent Par Value setting occur before the
> first Stock Round.
> 
> I am just not familiar enough with the code yet to get all the various
> patches together yet for you. And what I suggest is some significant
> code refactoring required. 
> 
> Mark Smith
> 

Mark -

The address to use is rai...@li.... I'll cc my
reply to the list. 

The xml fix is fairly straightforward, and should relatively easy to
generate a diff for that.

As for the B&O issue. I think Erik can speak more knowledgeably about
how that area of the code works. I suspect that with the current
refactoring that I'm doing, once I am ready to work on the game code
itself, I'll have some thoughts on how that stuff should work. 


---Brett.

Avoid the Gates of Hell.  Use Linux
	-- unknown source

Re: [Rails-devel] Engine 18xx Board archive previously mentioned

[brett l. <wak...@gm...>]

On Mon, Sep 22, 2008 at 3:52 PM, Mark Smith <mar...@gm...> wrote:
> Brett,
>
> Attached is the 18xx Engine I had mentioned last Friday on the Rails
> Development Mailing List. I have written up a brief Read Me file that
> explains how to use it at this point in time.
>
> Note, as said before it cannot play any 18xx Game, and I provide it as a
> possible alternative towards the Rails Game Board your product currently
> provides.
>
> Feel free to come back with comments, or questions if you have difficulties
> running the product.
>
> Mark
>
>
>


I've posted your files at
http://rails.sourceforge.net/18XX_Engine.Sept_2008.zip for people to
review.

I am very impressed with your implementation.  There's a few usability
quibbles I have, but if we're only talking about your game board code,
I'm happy with how it operates.

If you're interested in bringing your code into our project, I'd be
pleased to have it. It's definitely a step up from what we currently
have.

For first steps, I'd recommend looking at the Options class, and see
what changes will be needed get started.

---Brett.

Re: [Rails-devel] track/route awareness

[Erik V. <eri...@hc...>]

I have reviewed what your Rails project does with the board, and the fact
that you make use of GIF images for the tiles, tokens, etc. I see at least
two distinct issues with your current implemenation, one is the tokens are
just a circle of a specific color. If you look at 1835, or 1837 there are
minor companies that share the SAME color, so that it is very hard to tell
from your map which minor company starts where.  

 
The latest version does draw company abbreviations on top of each token.
Not in vector graphics, though, so it probably won't scale.
 
Erik.

Re: [Rails-devel] track/route awareness

[Michał B. <ba...@tl...>]

I intend to try the brute-force approach and see why it fails :-) for 
single trains brute-force should be more than enough, and you don't see 
more than 2 big trains in one company at the same time, do you?

the other idea is that I know the lead programmer of 1830 computer game, 
I'll ask how he solved this problem (the PC game is extremely quick in 
calculating this)...

Mark Smith wrote:
> Now this last point brings up a philosphical question of game 
> implementation. Should the Player pick the routes, or should the 
> computer auto-calculate the routes to maximize income? The game rules 
> specifically state the player picks routes, and has to live with the 
> mistake. However, sometimes in some games (like 1853), you may want to 
> maximize your mail run, or maximize your revenue at different times.


the 1830pc game calculates it for you, and I would be more than willing 
to sacrifice the player not losing when he doesn't see a best route for 
the huge time gain this would bring. a player could still run the train 
for a shorther route if none of the shareholders objected...

best regards,
   mike bazynski

Re: [Rails-devel] track/route awareness

[brett l. <wak...@gm...>]

2008/9/18 Mark Smith <mar...@gm...>:
> I will get around to putting my JAR file together and determining the best
> place to put it so that it can be retrieved and reviewed by you and anyone
> else interested.
>

If it's not too big ( > 1 MB), e-mail it directly to me, and I can
throw it up on the rails webspace that sourceforge gives us.

> I had seen your SVG stuff, but I must admit I did not pay it a lot of
> attention. I probably should take another look at that aspect.
>
> As for keeping both options (User Select Route) and (Auto Generate Route)
> open is a viable alternative, but the first easier option is the User Select
> Route. The Auto Generate Route from a UI perspective is just a different
> method for selecting the actual points in the route, the graphical display
> method is the same regardless.
>

Agreed.

> As for your method for accepting new code contributors, your patches method
> is fine with me. I will get a fresh copy of your code base this weekend, and
> see what jumps out at me.
>
> With regards towards codebase merging, I will want to do a basic review of
> your current one and come up with a reasonable strategy to implement, and
> not disrupt your current intentions of transistioning to the network based
> plans you have in mind.
>

At this point, I don't think adding in your code will disrupt my
efforts for quite a while. I was going to rework the drawing code once
the lower level bits were ready. If you're going to be working in that
area, it just means there will be less for me to do because you've
already gotten to it. ;-)


> More thoughts to come over the weekend.
>
> Mark
>

---Brett


> On Thu, Sep 18, 2008 at 7:49 PM, Brett Lentz <wak...@gm...> wrote:
>>
>> On Thu, 2008-09-18 at 19:09 -0500, Mark Smith wrote:
>> > I have reviewed what your Rails project does with the board, and the
>> > fact that you make use of GIF images for the tiles, tokens, etc. I see
>> > at least two distinct issues with your current implemenation, one is
>> > the tokens are just a circle of a specific color. If you look at 1835,
>> > or 1837 there are minor companies that share the SAME color, so that
>> > it is very hard to tell from your map which minor company starts
>> > where. The other issue as I know from your forum has not yet been
>> > solved is scaling the map up and down in size.
>> >
>> > My approach was to use the track connections, station locations,
>> > types, and stuff to draw the board and tiles via Vector Graphics. This
>> > gives me not only a very clean drawing, but one that can scale up and
>> > down in size with little extra coding efforts.
>>
>> Looks like you missed the SVG tiles we have. We're using Batik to handle
>> the rendering of SVG tiles.
>>
>> Much of the drawing code is fairly basic and there are several known
>> issues that make me want to try rewriting large sections of it. However,
>> you're welcome to take on that task. It sounds like we've got a lot of
>> similar ideas on how it should be done. :-)
>>
>> > The tokens I draw are filled circles with the company
>> > Abbreviation/Number centered in the token. I also draw my default size
>> > larger, about the same size as the tiles if you hold them up to the
>> > screen. One other advantage is that since I draw the track segment by
>> > segment, I can draw routes for different trains in different colors,
>> > so that you can tell at a glance if they conflict, and the program
>> > will know if they would re-use track and say "NAY, no re-use buddy!".
>> > But I have not gotten it to that specific point yet. but I am close to
>> > actually implementing a way to have the user pick out the routes.
>> >
>>
>> Interesting.  Perhaps you'd like to work on merging codebases?
>>
>> > Now this last point brings up a philosphical question of game
>> > implementation. Should the Player pick the routes, or should the
>> > computer auto-calculate the routes to maximize income? The game rules
>> > specifically state the player picks routes, and has to live with the
>> > mistake. However, sometimes in some games (like 1853), you may want to
>> > maximize your mail run, or maximize your revenue at different times.
>> >
>>
>> I would prefer to support both options, and leave it up to the user to
>> select an option at runtime.
>>
>>
>> > If there is intrest, I could be enticed to provide a copy of a JAVA
>> > JAR File that could be run to see if my approach sparks interest. I
>> > feel that it might be worth considering a possible integration of my
>> > map/tile routines into Rails... but only the current product looks
>> > promising to those currently implementing.
>> >
>>
>> There's definitely interest. We can always use more people contributing
>> code. :-)
>>
>> In general, if you're contributing code, I (usually) leave the
>> implementation details up to you. Unless there's some compelling reason,
>> I don't tend to dictate how you implement things.
>>
>> So, if you've got a working graphics subsystem that's better than what
>> we've already got, I'm happy to let you try merging it or adapting the
>> ideas into our codebase using new code. How you want to do it, is fine
>> by me.
>>
>> My basic policy is that I like to see a couple of small-ish patches from
>> anyone that hasn't contributed before. It helps me get a feel for your
>> coding style, and assures that we're all moving in the same direction.
>>
>> After those initial steps, if you're still interested, I'm happy to
>> allow you to commit directly to CVS.
>>
>> How does that sound?
>>
>>
>> > Let me know what you all think.
>> > Mark Smith
>>
>>
>> ---Brett.
>>
>> > On Thu, Sep 18, 2008 at 5:05 PM, Jonathan Ferro
>> > <jon...@ya...> wrote:
>> >         Several years ago I posted an idea for an algorithm based on
>> >         computing half-routes (starting from stations an only going
>> >         one direction) but I've never proved to myself that the
>> >         filtering needed (to combine them into full routes and not
>> >         conflict with other trains in a corp) would be practical.  I
>> >         think that was the last time it was openly discussed, and I've
>> >         never looked into the code to see how that would fit with the
>> >         current game-state internals.
>> >
>> >         -- Jonathan
>> >
>> >
>> >
>> >
>> >         ----- Original Message ----
>> >         From: brett lentz <wak...@gm...>
>> >         To: Development list for Rails: an 18xx game
>> >         <rai...@li...>
>> >         Sent: Thursday, September 18, 2008 8:10:11 AM
>> >         Subject: Re: [Rails-devel] track/route awareness
>> >
>> >         On Thu, Sep 18, 2008 at 2:49 AM, Michał Bażyński
>> >         <ba...@tl...> wrote:
>> >         > hello,
>> >         >   is anybody working on things like recognizing legal track
>> >         builds or
>> >         > calculating revenue? not that I am ready to announce I
>> >         intend to do it,
>> >         > but if it's a work in progress already it won't make sense
>> >         for me to
>> >         > look into that problem at all...
>> >         >
>> >         > mike bazynski
>> >
>> >
>> >         Nobody has stepped forward to work on route awareness.  We
>> >         have a
>> >         basic implementation for matching legal plays by tracking
>> >         which sides
>> >         of each hex have track, and lining them up to the sides of
>> >         each other
>> >         hex that has track.
>> >
>> >         If you would like to work on route calculation, you're welcome
>> >         to it.
>> >
>> >         ---Brett.
>>
>>
>>
>> -------------------------------------------------------------------------
>> This SF.Net email is sponsored by the Moblin Your Move Developer's
>> challenge
>> Build the coolest Linux based applications with Moblin SDK & win great
>> prizes
>> Grand prize is a trip for two to an Open Source event anywhere in the
>> world
>> http://moblin-contest.org/redirect.php?banner_id=100&url=/
>> _______________________________________________
>> Rails-devel mailing list
>> Rai...@li...
>> https://lists.sourceforge.net/lists/listinfo/rails-devel
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>
>

Re: [Rails-devel] track/route awareness

[Mark S. <mar...@gm...>]

I will get around to putting my JAR file together and determining the best
place to put it so that it can be retrieved and reviewed by you and anyone
else interested.

I had seen your SVG stuff, but I must admit I did not pay it a lot of
attention. I probably should take another look at that aspect.

As for keeping both options (User Select Route) and (Auto Generate Route)
open is a viable alternative, but the first easier option is the User Select
Route. The Auto Generate Route from a UI perspective is just a different
method for selecting the actual points in the route, the graphical display
method is the same regardless.

As for your method for accepting new code contributors, your patches method
is fine with me. I will get a fresh copy of your code base this weekend, and
see what jumps out at me.

With regards towards codebase merging, I will want to do a basic review of
your current one and come up with a reasonable strategy to implement, and
not disrupt your current intentions of transistioning to the network based
plans you have in mind.

More thoughts to come over the weekend.

Mark

On Thu, Sep 18, 2008 at 7:49 PM, Brett Lentz <wak...@gm...> wrote:

> On Thu, 2008-09-18 at 19:09 -0500, Mark Smith wrote:
> > I have reviewed what your Rails project does with the board, and the
> > fact that you make use of GIF images for the tiles, tokens, etc. I see
> > at least two distinct issues with your current implemenation, one is
> > the tokens are just a circle of a specific color. If you look at 1835,
> > or 1837 there are minor companies that share the SAME color, so that
> > it is very hard to tell from your map which minor company starts
> > where. The other issue as I know from your forum has not yet been
> > solved is scaling the map up and down in size.
> >
> > My approach was to use the track connections, station locations,
> > types, and stuff to draw the board and tiles via Vector Graphics. This
> > gives me not only a very clean drawing, but one that can scale up and
> > down in size with little extra coding efforts.
>
> Looks like you missed the SVG tiles we have. We're using Batik to handle
> the rendering of SVG tiles.
>
> Much of the drawing code is fairly basic and there are several known
> issues that make me want to try rewriting large sections of it. However,
> you're welcome to take on that task. It sounds like we've got a lot of
> similar ideas on how it should be done. :-)
>
> > The tokens I draw are filled circles with the company
> > Abbreviation/Number centered in the token. I also draw my default size
> > larger, about the same size as the tiles if you hold them up to the
> > screen. One other advantage is that since I draw the track segment by
> > segment, I can draw routes for different trains in different colors,
> > so that you can tell at a glance if they conflict, and the program
> > will know if they would re-use track and say "NAY, no re-use buddy!".
> > But I have not gotten it to that specific point yet. but I am close to
> > actually implementing a way to have the user pick out the routes.
> >
>
> Interesting.  Perhaps you'd like to work on merging codebases?
>
> > Now this last point brings up a philosphical question of game
> > implementation. Should the Player pick the routes, or should the
> > computer auto-calculate the routes to maximize income? The game rules
> > specifically state the player picks routes, and has to live with the
> > mistake. However, sometimes in some games (like 1853), you may want to
> > maximize your mail run, or maximize your revenue at different times.
> >
>
> I would prefer to support both options, and leave it up to the user to
> select an option at runtime.
>
>
> > If there is intrest, I could be enticed to provide a copy of a JAVA
> > JAR File that could be run to see if my approach sparks interest. I
> > feel that it might be worth considering a possible integration of my
> > map/tile routines into Rails... but only the current product looks
> > promising to those currently implementing.
> >
>
> There's definitely interest. We can always use more people contributing
> code. :-)
>
> In general, if you're contributing code, I (usually) leave the
> implementation details up to you. Unless there's some compelling reason,
> I don't tend to dictate how you implement things.
>
> So, if you've got a working graphics subsystem that's better than what
> we've already got, I'm happy to let you try merging it or adapting the
> ideas into our codebase using new code. How you want to do it, is fine
> by me.
>
> My basic policy is that I like to see a couple of small-ish patches from
> anyone that hasn't contributed before. It helps me get a feel for your
> coding style, and assures that we're all moving in the same direction.
>
> After those initial steps, if you're still interested, I'm happy to
> allow you to commit directly to CVS.
>
> How does that sound?
>
>
> > Let me know what you all think.
> > Mark Smith
>
>
> ---Brett.
>
> > On Thu, Sep 18, 2008 at 5:05 PM, Jonathan Ferro
> > <jon...@ya...> wrote:
> >         Several years ago I posted an idea for an algorithm based on
> >         computing half-routes (starting from stations an only going
> >         one direction) but I've never proved to myself that the
> >         filtering needed (to combine them into full routes and not
> >         conflict with other trains in a corp) would be practical.  I
> >         think that was the last time it was openly discussed, and I've
> >         never looked into the code to see how that would fit with the
> >         current game-state internals.
> >
> >         -- Jonathan
> >
> >
> >
> >
> >         ----- Original Message ----
> >         From: brett lentz <wak...@gm...>
> >         To: Development list for Rails: an 18xx game
> >         <rai...@li...>
> >         Sent: Thursday, September 18, 2008 8:10:11 AM
> >         Subject: Re: [Rails-devel] track/route awareness
> >
> >         On Thu, Sep 18, 2008 at 2:49 AM, Michał Bażyński
> >         <ba...@tl...> wrote:
> >         > hello,
> >         >   is anybody working on things like recognizing legal track
> >         builds or
> >         > calculating revenue? not that I am ready to announce I
> >         intend to do it,
> >         > but if it's a work in progress already it won't make sense
> >         for me to
> >         > look into that problem at all...
> >         >
> >         > mike bazynski
> >
> >
> >         Nobody has stepped forward to work on route awareness.  We
> >         have a
> >         basic implementation for matching legal plays by tracking
> >         which sides
> >         of each hex have track, and lining them up to the sides of
> >         each other
> >         hex that has track.
> >
> >         If you would like to work on route calculation, you're welcome
> >         to it.
> >
> >         ---Brett.
>
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

Re: [Rails-devel] track/route awareness

[Brett L. <wak...@gm...>]

On Thu, 2008-09-18 at 19:09 -0500, Mark Smith wrote:
> I have reviewed what your Rails project does with the board, and the
> fact that you make use of GIF images for the tiles, tokens, etc. I see
> at least two distinct issues with your current implemenation, one is
> the tokens are just a circle of a specific color. If you look at 1835,
> or 1837 there are minor companies that share the SAME color, so that
> it is very hard to tell from your map which minor company starts
> where. The other issue as I know from your forum has not yet been
> solved is scaling the map up and down in size.
> 
> My approach was to use the track connections, station locations,
> types, and stuff to draw the board and tiles via Vector Graphics. This
> gives me not only a very clean drawing, but one that can scale up and
> down in size with little extra coding efforts. 

Looks like you missed the SVG tiles we have. We're using Batik to handle
the rendering of SVG tiles.

Much of the drawing code is fairly basic and there are several known
issues that make me want to try rewriting large sections of it. However,
you're welcome to take on that task. It sounds like we've got a lot of
similar ideas on how it should be done. :-)

> The tokens I draw are filled circles with the company
> Abbreviation/Number centered in the token. I also draw my default size
> larger, about the same size as the tiles if you hold them up to the
> screen. One other advantage is that since I draw the track segment by
> segment, I can draw routes for different trains in different colors,
> so that you can tell at a glance if they conflict, and the program
> will know if they would re-use track and say "NAY, no re-use buddy!".
> But I have not gotten it to that specific point yet. but I am close to
> actually implementing a way to have the user pick out the routes.
> 

Interesting.  Perhaps you'd like to work on merging codebases?

> Now this last point brings up a philosphical question of game
> implementation. Should the Player pick the routes, or should the
> computer auto-calculate the routes to maximize income? The game rules
> specifically state the player picks routes, and has to live with the
> mistake. However, sometimes in some games (like 1853), you may want to
> maximize your mail run, or maximize your revenue at different times.
> 

I would prefer to support both options, and leave it up to the user to
select an option at runtime.


> If there is intrest, I could be enticed to provide a copy of a JAVA
> JAR File that could be run to see if my approach sparks interest. I
> feel that it might be worth considering a possible integration of my
> map/tile routines into Rails... but only the current product looks
> promising to those currently implementing.
> 

There's definitely interest. We can always use more people contributing
code. :-)

In general, if you're contributing code, I (usually) leave the
implementation details up to you. Unless there's some compelling reason,
I don't tend to dictate how you implement things.

So, if you've got a working graphics subsystem that's better than what
we've already got, I'm happy to let you try merging it or adapting the
ideas into our codebase using new code. How you want to do it, is fine
by me.

My basic policy is that I like to see a couple of small-ish patches from
anyone that hasn't contributed before. It helps me get a feel for your
coding style, and assures that we're all moving in the same direction.

After those initial steps, if you're still interested, I'm happy to
allow you to commit directly to CVS.

How does that sound?


> Let me know what you all think.
> Mark Smith


---Brett.

> On Thu, Sep 18, 2008 at 5:05 PM, Jonathan Ferro
> <jon...@ya...> wrote:
>         Several years ago I posted an idea for an algorithm based on
>         computing half-routes (starting from stations an only going
>         one direction) but I've never proved to myself that the
>         filtering needed (to combine them into full routes and not
>         conflict with other trains in a corp) would be practical.  I
>         think that was the last time it was openly discussed, and I've
>         never looked into the code to see how that would fit with the
>         current game-state internals.
>         
>         -- Jonathan
>         
>         
>         
>         
>         ----- Original Message ----
>         From: brett lentz <wak...@gm...>
>         To: Development list for Rails: an 18xx game
>         <rai...@li...>
>         Sent: Thursday, September 18, 2008 8:10:11 AM
>         Subject: Re: [Rails-devel] track/route awareness
>         
>         On Thu, Sep 18, 2008 at 2:49 AM, Michał Bażyński
>         <ba...@tl...> wrote:
>         > hello,
>         >   is anybody working on things like recognizing legal track
>         builds or
>         > calculating revenue? not that I am ready to announce I
>         intend to do it,
>         > but if it's a work in progress already it won't make sense
>         for me to
>         > look into that problem at all...
>         >
>         > mike bazynski
>         
>         
>         Nobody has stepped forward to work on route awareness.  We
>         have a
>         basic implementation for matching legal plays by tracking
>         which sides
>         of each hex have track, and lining them up to the sides of
>         each other
>         hex that has track.
>         
>         If you would like to work on route calculation, you're welcome
>         to it.
>         
>         ---Brett.

Re: [Rails-devel] track/route awareness

[Mark S. <mar...@gm...>]

Back many years ago early 90's I started a personal project to implement the
18xx game system on my computer, using C++. When JAVA came out, I translated
what I had done to JAVA and then put it away for a decade. When I found your
SourceForge project back late 2006 I brought it back out and made progress.
My initial focus, and still to be fully complete to my satisfaction is
drawing and manipulating the board, tiles, track, stations, etc. My feeling
is that the board is central to the entire game system, and if not done
properly, you can't really play the game properly.

I have reviewed what your Rails project does with the board, and the fact
that you make use of GIF images for the tiles, tokens, etc. I see at least
two distinct issues with your current implemenation, one is the tokens are
just a circle of a specific color. If you look at 1835, or 1837 there are
minor companies that share the SAME color, so that it is very hard to tell
from your map which minor company starts where. The other issue as I know
from your forum has not yet been solved is scaling the map up and down in
size.

My approach was to use the track connections, station locations, types, and
stuff to draw the board and tiles via Vector Graphics. This gives me not
only a very clean drawing, but one that can scale up and down in size with
little extra coding efforts. The tokens I draw are filled circles with the
company Abbreviation/Number centered in the token. I also draw my default
size larger, about the same size as the tiles if you hold them up to the
screen. One other advantage is that since I draw the track segment by
segment, I can draw routes for different trains in different colors, so that
you can tell at a glance if they conflict, and the program will know if they
would re-use track and say "NAY, no re-use buddy!". But I have not gotten it
to that specific point yet. but I am close to actually implementing a way to
have the user pick out the routes.

Now this last point brings up a philosphical question of game
implementation. Should the Player pick the routes, or should the computer
auto-calculate the routes to maximize income? The game rules specifically
state the player picks routes, and has to live with the mistake. However,
sometimes in some games (like 1853), you may want to maximize your mail run,
or maximize your revenue at different times.

If there is intrest, I could be enticed to provide a copy of a JAVA JAR File
that could be run to see if my approach sparks interest. I feel that it
might be worth considering a possible integration of my map/tile routines
into Rails... but only the current product looks promising to those
currently implementing.

Let me know what you all think.
Mark Smith

On Thu, Sep 18, 2008 at 5:05 PM, Jonathan Ferro <jon...@ya...>wrote:

> Several years ago I posted an idea for an algorithm based on computing
> half-routes (starting from stations an only going one direction) but I've
> never proved to myself that the filtering needed (to combine them into full
> routes and not conflict with other trains in a corp) would be practical.  I
> think that was the last time it was openly discussed, and I've never looked
> into the code to see how that would fit with the current game-state
> internals.
>
> -- Jonathan
>
>
>
> ----- Original Message ----
> From: brett lentz <wak...@gm...>
> To: Development list for Rails: an 18xx game <
> rai...@li...>
> Sent: Thursday, September 18, 2008 8:10:11 AM
> Subject: Re: [Rails-devel] track/route awareness
>
> On Thu, Sep 18, 2008 at 2:49 AM, Michał Bażyński <ba...@tl...> wrote:
> > hello,
> >   is anybody working on things like recognizing legal track builds or
> > calculating revenue? not that I am ready to announce I intend to do it,
> > but if it's a work in progress already it won't make sense for me to
> > look into that problem at all...
> >
> > mike bazynski
>
>
> Nobody has stepped forward to work on route awareness.  We have a
> basic implementation for matching legal plays by tracking which sides
> of each hex have track, and lining them up to the sides of each other
> hex that has track.
>
> If you would like to work on route calculation, you're welcome to it.
>
> ---Brett.
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

Re: [Rails-devel] track/route awareness

[Jonathan F. <jon...@ya...>]

Several years ago I posted an idea for an algorithm based on computing half-routes (starting from stations an only going one direction) but I've never proved to myself that the filtering needed (to combine them into full routes and not conflict with other trains in a corp) would be practical.  I think that was the last time it was openly discussed, and I've never looked into the code to see how that would fit with the current game-state internals.

-- Jonathan



----- Original Message ----
From: brett lentz <wak...@gm...>
To: Development list for Rails: an 18xx game <rai...@li...>
Sent: Thursday, September 18, 2008 8:10:11 AM
Subject: Re: [Rails-devel] track/route awareness

On Thu, Sep 18, 2008 at 2:49 AM, Michał Bażyński <ba...@tl...> wrote:
> hello,
>   is anybody working on things like recognizing legal track builds or
> calculating revenue? not that I am ready to announce I intend to do it,
> but if it's a work in progress already it won't make sense for me to
> look into that problem at all...
>
> mike bazynski


Nobody has stepped forward to work on route awareness.  We have a
basic implementation for matching legal plays by tracking which sides
of each hex have track, and lining them up to the sides of each other
hex that has track.

If you would like to work on route calculation, you're welcome to it.

---Brett.
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Rails-devel mailing list
Rai...@li...
https://lists.sourceforge.net/lists/listinfo/rails-devel

Re: [Rails-devel] track/route awareness

[brett l. <wak...@gm...>]

On Thu, Sep 18, 2008 at 2:49 AM, Michał Bażyński <ba...@tl...> wrote:
> hello,
>   is anybody working on things like recognizing legal track builds or
> calculating revenue? not that I am ready to announce I intend to do it,
> but if it's a work in progress already it won't make sense for me to
> look into that problem at all...
>
> mike bazynski


Nobody has stepped forward to work on route awareness.  We have a
basic implementation for matching legal plays by tracking which sides
of each hex have track, and lining them up to the sides of each other
hex that has track.

If you would like to work on route calculation, you're welcome to it.

---Brett.

[Rails-devel] track/route awareness

[Michał B. <ba...@tl...>]

hello,
   is anybody working on things like recognizing legal track builds or 
calculating revenue? not that I am ready to announce I intend to do it, 
but if it's a work in progress already it won't make sense for me to 
look into that problem at all...

mike bazynski

Re: [Rails-devel] Starting on a client/server

[Brett L. <wak...@gm...>]

> -----Original Message-----
> From: rai...@li... [mailto:rails-devel-
> bo...@li...] On Behalf Of Erik Vos
> Sent: Thursday, September 04, 2008 12:51 PM
> To: 'Development list for Rails: an 18xx game'
> Subject: Re: [Rails-devel] Starting on a client/server
> 
> > > Sockets are byte channels, and you can send whatever you want
> across
> > > them,
> > > such as serialized objects.
> >
> > Great!  So then tossing our objects over the SocketChannel ought to
> be
> > relatively easy.
> 
> You might need to think about some little protocol around that, as a
> byte
> channel has no "sentence integrity",
> i.e. it does not know where one message ends and another one starts.
> Not
> sure if that is really needed in this case.
> 

We'll need to test this a bit. I suspect it shouldn't be an issue, as we'll
be attaching a ByteStreamReader (or is it ByteReader, I forget) to the
SocketChannel, and read until we get a whole serialized object.

> > It seems that java.nio.SocketChannel is probably a better choice than
> > java.net.Socket. It seems like it's easier to manage multiple
> > connections to
> > the same port more easily with a SocketChannel.  I'm new to
> > this part of the
> > Java API, but that's my first impression.
> 
> A while ago I built a kind of package around java.nio.SocketChannel for
> a
> specific case where one client controls multiple servers. I tried to
> make it
> as generic as possible. With some changes it might be turned into
> something
> that we could use. I'll check tomorrow.
> 
> Erik.
> 

That would definitely speed things up, if you've already got the basic
SocketChannel implementation.

---Brett.

Re: [Rails-devel] Starting on a client/server

[Erik V. <eri...@hc...>]

> > Sockets are byte channels, and you can send whatever you want across
> > them,
> > such as serialized objects.
> 
> Great!  So then tossing our objects over the SocketChannel ought to be
> relatively easy.

You might need to think about some little protocol around that, as a byte
channel has no "sentence integrity",
i.e. it does not know where one message ends and another one starts. Not
sure if that is really needed in this case.
 
> It seems that java.nio.SocketChannel is probably a better choice than
> java.net.Socket. It seems like it's easier to manage multiple 
> connections to
> the same port more easily with a SocketChannel.  I'm new to 
> this part of the
> Java API, but that's my first impression.

A while ago I built a kind of package around java.nio.SocketChannel for a
specific case where one client controls multiple servers. I tried to make it
as generic as possible. With some changes it might be turned into something
that we could use. I'll check tomorrow.

Erik.

Re: [Rails-devel] Starting on a client/server

[John A. T. <ja...@ja...>]

brett lentz wrote:
> Any suggestions on which API that might be?
>   
I would decide what needs to be communicated (ie, is it going to be 
RPC-style [probably easier to do for web-based clients since everything 
has to be async there], keeping object graphs synchronized, etc) and 
build the API at that level, then build multiple implementations across 
various transports.  Obviously, some of the transports are going to be 
better at certain styles of interaction than others, so you have to keep 
in mind the types of transports you want to support.

--
John A. Tamplin					ja...@ja...
770/436-5387 HOME				4116 Manson Ave
                		 		Smyrna, GA  30082-3723

Re: [Rails-devel] Starting on a client/server

[brett l. <wak...@gm...>]

On Wed, Sep 3, 2008 at 8:27 PM, John A. Tamplin <ja...@ja...> wrote:
> Brett Lentz wrote:
>> It seems that java.nio.SocketChannel is probably a better choice than
>> java.net.Socket. It seems like it's easier to manage multiple connections to
>> the same port more easily with a SocketChannel.  I'm new to this part of the
>> Java API, but that's my first impression.
>>
> I think the only real difference is the buffering is more efficient in
> nio, but in either case I think you want a higher level API that can
> encapsulate same-machine via synchronized queues, between machines with
> sockets, between machines with some multicast protocol, etc.
>
> --
> John A. Tamplin                                 ja...@ja...
> 770/436-5387 HOME                               4116 Manson Ave
>

Any suggestions on which API that might be?

---Brett.