Thread: [Rails-devel] Bug: 1835, 1.7.0

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

In the operating round just finished, PR formed and all its privates and
minors merged.  However, the M5 head in Berlin has not gone away as it should.

Hi John,
On loading the save file everything is ok.

Please send an earlier save file if possible.

Regards,
Martin

-----Ursprüngliche Nachricht-----
Von: John David Galt [mailto:jd...@di...] 
Gesendet: Sonntag, 1. April 2012 20:29
An: rai...@li...
Betreff: [Rails-devel] Bug: 1835, 1.7.0

In the operating round just finished, PR formed and all its privates and
minors merged.  However, the M5 head in Berlin has not gone away as it
should.

It should be possible to reconstruct earlier stages by repeated undoing.
That works to some extent here, but it also reveals another bug: the PR
tokens do not disappear when PR formation is undone, and the minor tokens do
not return.
I'm pretty sure that this is due to the need for an explicit update() in
HexMap.layBaseToken() (line ; that must have the consequence that update()
is not called on an undo.
If I then save the game and reload it, the PR tokens are gone and the minor
tokens reappear, which more or less proves the above analysis.

>From that earlier saved file I can indeed confirm the reported bug.  On
first sight, I cannot find a clear reason why this is so, but my guess is
that it's another missing update().

I'll look for an easy fix, but I'm actually afraid that it's a real design
flaw.
All this could even mean that having HexMap extend ModelObject was a bad
idea (although it seems to work for tile lays). 
Anyhow,  I'll reconsider the whole subject of Observing token lays.

Erik

> -----Original Message-----
> From: Dr. Martin Brumm [mailto:dr....@t-...]
> Sent: Sunday, April 01, 2012 9:03 PM
> To: 'Development list for Rails: an 18xx game'
> Cc: jd...@di...
> Subject: Re: [Rails-devel] Bug: 1835, 1.7.0
> 
> Hi John,
> On loading the save file everything is ok.
> 
> Please send an earlier save file if possible.
> 
> Regards,
> Martin
> 
> -----Ursprüngliche Nachricht-----
> Von: John David Galt [mailto:jd...@di...]
> Gesendet: Sonntag, 1. April 2012 20:29
> An: rai...@li...
> Betreff: [Rails-devel] Bug: 1835, 1.7.0
> 
> In the operating round just finished, PR formed and all its privates and
minors
> merged.  However, the M5 head in Berlin has not gone away as it should.
> 
> 
> 
>
----------------------------------------------------------------------------
--
> This SF email is sponsosred by:
> Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-
> msazure _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

Erik:
those kind of problems were the motivation for starting coding on 
Rails2.0. Actually HexMap is the one the classes that I am currently 
refactoring to add those new elements I have created for this reason.
But it is one of the harder cases.

So at this point you have to consider if you really want to fix that 
inside the Rails1.x branch or you want jump ship and get involved in 
Rails2.0 at some point in time.

Stefan

On 04/01/2012 11:46 PM, Erik Vos wrote:
> It should be possible to reconstruct earlier stages by repeated undoing.
> That works to some extent here, but it also reveals another bug: the PR
> tokens do not disappear when PR formation is undone, and the minor tokens do
> not return.
> I'm pretty sure that this is due to the need for an explicit update() in
> HexMap.layBaseToken() (line ; that must have the consequence that update()
> is not called on an undo.
> If I then save the game and reload it, the PR tokens are gone and the minor
> tokens reappear, which more or less proves the above analysis.
>
>> From that earlier saved file I can indeed confirm the reported bug.  On
> first sight, I cannot find a clear reason why this is so, but my guess is
> that it's another missing update().
>
> I'll look for an easy fix, but I'm actually afraid that it's a real design
> flaw.
> All this could even mean that having HexMap extend ModelObject was a bad
> idea (although it seems to work for tile lays).
> Anyhow,  I'll reconsider the whole subject of Observing token lays.
>
> Erik
>
>> -----Original Message-----
>> From: Dr. Martin Brumm [mailto:dr....@t-...]
>> Sent: Sunday, April 01, 2012 9:03 PM
>> To: 'Development list for Rails: an 18xx game'
>> Cc: jd...@di...
>> Subject: Re: [Rails-devel] Bug: 1835, 1.7.0
>>
>> Hi John,
>> On loading the save file everything is ok.
>>
>> Please send an earlier save file if possible.
>>
>> Regards,
>> Martin
>>
>> -----Ursprüngliche Nachricht-----
>> Von: John David Galt [mailto:jd...@di...]
>> Gesendet: Sonntag, 1. April 2012 20:29
>> An: rai...@li...
>> Betreff: [Rails-devel] Bug: 1835, 1.7.0
>>
>> In the operating round just finished, PR formed and all its privates and
> minors
>> merged.  However, the M5 head in Berlin has not gone away as it should.
>>
>>
>>
>>
> ----------------------------------------------------------------------------
> --
>> This SF email is sponsosred by:
>> Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-
>> msazure _______________________________________________
>> Rails-devel mailing list
>> Rai...@li...
>> https://lists.sourceforge.net/lists/listinfo/rails-devel
>
>
>
> ------------------------------------------------------------------------------
> This SF email is sponsosred by:
> Try Windows Azure free for 90 days Click Here
> http://p.sf.net/sfu/sfd2d-msazure
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

Stefan,

Perhaps a neat 1.x-style solution would be to create (perhaps two separate)
inner State subclasses to maintain the MapHex tile and token states.
This would be similar to what I'm working on for the player order (but it
only partly works yet).

Whether or not this will really materialize before Rails 2.0 becomes master
remains to be seen.

Erik. 

> -----Original Message-----
> From: Stefan Frey [mailto:ste...@we...]
> Sent: Monday, April 02, 2012 9:31 AM
> To: rai...@li...
> Subject: Re: [Rails-devel] Bug: 1835, 1.7.0
> 
> Erik:
> those kind of problems were the motivation for starting coding on
Rails2.0.
> Actually HexMap is the one the classes that I am currently refactoring to
add
> those new elements I have created for this reason.
> But it is one of the harder cases.
> 
> So at this point you have to consider if you really want to fix that
inside the
> Rails1.x branch or you want jump ship and get involved in
> Rails2.0 at some point in time.
> 
> Stefan
> 
> 
> On 04/01/2012 11:46 PM, Erik Vos wrote:
> > It should be possible to reconstruct earlier stages by repeated undoing.
> > That works to some extent here, but it also reveals another bug: the
> > PR tokens do not disappear when PR formation is undone, and the minor
> > tokens do not return.
> > I'm pretty sure that this is due to the need for an explicit update()
> > in
> > HexMap.layBaseToken() (line ; that must have the consequence that
> > update() is not called on an undo.
> > If I then save the game and reload it, the PR tokens are gone and the
> > minor tokens reappear, which more or less proves the above analysis.
> >
> >> From that earlier saved file I can indeed confirm the reported bug.
> >> On
> > first sight, I cannot find a clear reason why this is so, but my guess
> > is that it's another missing update().
> >
> > I'll look for an easy fix, but I'm actually afraid that it's a real
> > design flaw.
> > All this could even mean that having HexMap extend ModelObject was a
> > bad idea (although it seems to work for tile lays).
> > Anyhow,  I'll reconsider the whole subject of Observing token lays.
> >
> > Erik
> >
> >> -----Original Message-----
> >> From: Dr. Martin Brumm [mailto:dr....@t-...]
> >> Sent: Sunday, April 01, 2012 9:03 PM
> >> To: 'Development list for Rails: an 18xx game'
> >> Cc: jd...@di...
> >> Subject: Re: [Rails-devel] Bug: 1835, 1.7.0
> >>
> >> Hi John,
> >> On loading the save file everything is ok.
> >>
> >> Please send an earlier save file if possible.
> >>
> >> Regards,
> >> Martin
> >>
> >> -----Ursprüngliche Nachricht-----
> >> Von: John David Galt [mailto:jd...@di...]
> >> Gesendet: Sonntag, 1. April 2012 20:29
> >> An: rai...@li...
> >> Betreff: [Rails-devel] Bug: 1835, 1.7.0
> >>
> >> In the operating round just finished, PR formed and all its privates
> >> and
> > minors
> >> merged.  However, the M5 head in Berlin has not gone away as it should.
> >>
> >>
> >>
> >>
> > ----------------------------------------------------------------------
> > ------
> > --
> >> This SF email is sponsosred by:
> >> Try Windows Azure free for 90 days Click Here
> >> http://p.sf.net/sfu/sfd2d- msazure
> >> _______________________________________________
> >> Rails-devel mailing list
> >> Rai...@li...
> >> https://lists.sourceforge.net/lists/listinfo/rails-devel
> >
> >
> >
> > ----------------------------------------------------------------------
> > --------
> > This SF email is sponsosred by:
> > Try Windows Azure free for 90 days Click Here
> > http://p.sf.net/sfu/sfd2d-msazure
> > _______________________________________________
> > Rails-devel mailing list
> > Rai...@li...
> > https://lists.sourceforge.net/lists/listinfo/rails-devel
> 
> 
>
----------------------------------------------------------------------------
--
> This SF email is sponsosred by:
> Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-
> msazure _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

On 2012-04-01 12:02, Dr. Martin Brumm wrote:
> Hi John,
> On loading the save file everything is ok.

That's true for me also.  Probably Erik is right, it just needs to do a refresh.

Later in the same game, I run into this: the route calculator allows a run into
the Alsace-Lorraine for zero (which is against the rules).

John David:
this error is confirmed:
The value for Alsace is zero for the non-valid periods.

I was tempted to make off-board hexes blocked for zero value, however
there is no reason why not a designer choose to add an off-board area
with a zero value but still being a valid destination.

So I will provide a StaticModifier similar to the one used for 
Birmingham in 1851 (or was it 18TN - cannot remember now)?

Stefan

On 04/01/2012 11:59 PM, John David Galt wrote:
> On 2012-04-01 12:02, Dr. Martin Brumm wrote:
>> Hi John,
>> On loading the save file everything is ok.
>
> That's true for me also.  Probably Erik is right, it just needs to do a refresh.
>
> Later in the same game, I run into this: the route calculator allows a run into
> the Alsace-Lorraine for zero (which is against the rules).
>
>
>
> ------------------------------------------------------------------------------
> This SF email is sponsosred by:
> Try Windows Azure free for 90 days Click Here
> http://p.sf.net/sfu/sfd2d-msazure
>
>
>
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

> So I will provide a StaticModifier similar to the one used for Birmingham
in
> 1851 (or was it 18TN - cannot remember now)?

It was 1851 indeed.  

Erik