rails-devel — List for discussion of Rails development

Re: [Rails-devel] Removal of static class variables (git: 8b9df8)

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

Stefan -

I'm definitely identifying a different point of separation, but not
strictly the "dumb" client that you describe.

The thing that needs communicating is two things: the action and the data.

Consider the possibility that the server-side is a Web server with a
REST interface. (I select this example, because it could support a
variety of clients, from the current Swing interface, to a mobile app,
to a browser-based interface.)

To buy a share, clients could connect to the server, authenticate, and
then issue a GET to "/stockmarket/company/3/buy" and receive a
response.

The response can be a variety of things, depending on how "thick" or
"thin" the client is.

1. It could be a simple "success/failure" response.
2. It could be success/failure, with an updated client Wallet value.
3. It could be success/failure, with the current state of every
player's holdings, the whole stock market, or even whole game.

We can even always provide a "thin" client response that "thicker"
clients take and only use pieces of and discard the rest. It's
inefficient, but a conceivable implementation.

However, during client startup there is going to be a decision point
of "how do I get the current state of the game?"

We need to be able to initialize a GameManager and then tell it "you
are at this phase of the game", and provide it with enough information
that it can initialize the BankManager and other bits of the game.

With client/server play, the presumption that all clients are starting
at the beginning of the game does not need to be true. Servers can
(and should) be long-running services that maintain the state of all
games they are moderating. When returning to an in-progress game,
(this is the important part) there is no need to reread most of the
XML files. The server's already running and should know the state of
the game. The clients simply need to retrieve the game state and take
the next turn, if applicable.

Now... to be clear... I'm not saying that clients won't read *any*
XML, I'm saying that decoupling game engine from how we read
configuration data will allow more flexibility in how we read that
config data. That flexibility is a pre-requisite for client/server
implementation because while we may still load the tile manifest from
XML, we'll likely be transmitting the state of the map over the wire
because we will only care about the current state, not the beginning
state of the map.

---Brett.


On Sat, Jul 23, 2011 at 10:42 AM, Stefan Frey <ste...@we...> wrote:
> Erik & Brett,
> it seems to me it depends on what you the suggested client/server split looks
> like.
>
> >From what I understood from previous discussions is that Erik prefers
> the fat client approach, where "clients" all have a full game back-end running
> and only pass around actions between them. The server merely acts as a message
> broker.
>
> Another possibility is a client/server split between the game backend as the
> server and the clients only consisting of the GUI as defined by the swing
> classes of Rails. However it requires passing along the updates between the
> server and all connected clients.
>
> It seems to me that Brett thinks of the latter separation, however I still do
> not understand why the configureFromXML method matters here as all Managers
> using configureFromXML belong to the server part. So I might be wrong and
> Brett has other ideas.
>
> >From my point of view I think that Erik's approach is the easiest to implement
> and has some advantages in UI responsiveness. It might also be better suited
> for a migration to mobile devices like Android tablets, which I think should
> be the longer-term goal, which are restricted in bandwidth and latency, but
> have enough computing power to run the game backend.
>
> Stefan
>
>
> On Saturday, July 23, 2011 05:56:20 pm Erik Vos wrote:
>> > -----Original Message-----
>> > From: brett lentz [mailto:bre...@gm...]
>> > Sent: Saturday, July 23, 2011 5:36 PM
>> > To: Development list for Rails: an 18xx game
>> > Subject: Re: [Rails-devel] Removal of static class variables (git:
>> > 8b9df8) To be clear, the thing I was trying to recommend against is
>> > *just* the configureFromXML() method that we typically use to initialize
>> > the *Manager classes.
>> >
>> > That's the biggest point where a client/server model will differ.
>>
>> Why?
>> The client won't have the same managers, but why can't whatever will
>> replace the managers call the same configureFromXML() methods of the
>> objects that they need to know about?
>>
>> > We'll need to instantiate the various *Manager classes, but those classes
>> > will need to be instantiated using normal constructors that can then be
>> > configured, probably by getters and setters to allow the method of
>> > obtaining the information to be either local or remote, and the "game
>> > engine" won't care which.
>>
>> Always local, in my concept.  Static info does not need to squeezed through
>> the client/server interface. Unless you want to make the client really
>> dumb, of course.  And slower.
>>
>> Erik.
>>
>> > So, my main point is that I'd like us to be conscious of which classes
>> > have awareness of lower-level information (like XML parsing) and ensure
>> > that knowledge is appropriately abstracted away from the game logic in
>> > the same way that we try to separate the game logic from the
>> > display/view logic.
>> >
>> > ---Brett.
>>
>> ---------------------------------------------------------------------------
>> --- Storage Efficiency Calculator
>> This modeling tool is based on patent-pending intellectual property that
>> has been used successfully in hundreds of IBM storage optimization engage-
>> ments, worldwide.  Store less, Store more with what you own, Move data to
>> the right place. Try It Now!
>> http://www.accelacomm.com/jaw/sfnl/114/51427378/
>> _______________________________________________
>> Rails-devel mailing list
>> Rai...@li...
>> https://lists.sourceforge.net/lists/listinfo/rails-devel
>
> ------------------------------------------------------------------------------
> Storage Efficiency Calculator
> This modeling tool is based on patent-pending intellectual property that
> has been used successfully in hundreds of IBM storage optimization engage-
> ments, worldwide.  Store less, Store more with what you own, Move data to
> the right place. Try It Now! http://www.accelacomm.com/jaw/sfnl/114/51427378/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

Re: [Rails-devel] Removal of static class variables (git: 8b9df8)

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

On Sat, Jul 23, 2011 at 8:56 AM, Erik Vos <eri...@xs...> wrote:
>
>
>> -----Original Message-----
>> From: brett lentz [mailto:bre...@gm...]
>> Sent: Saturday, July 23, 2011 5:36 PM
>> To: Development list for Rails: an 18xx game
>> Subject: Re: [Rails-devel] Removal of static class variables (git: 8b9df8)
>> To be clear, the thing I was trying to recommend against is *just* the
>> configureFromXML() method that we typically use to initialize the *Manager
>> classes.
>>
>> That's the biggest point where a client/server model will differ.
>
> Why?
> The client won't have the same managers, but why can't whatever will replace the managers call the same configureFromXML() methods of the objects that they need to know about?
>

Because the clients should take the "authoritative" values for things
given to it by the server, not whatever arbitrary values somebody
decides to put into their own local XML files. Trusting the client to
do the right thing on its own makes exploitation far too easy.

>> We'll need to instantiate the various *Manager classes, but those classes will
>> need to be instantiated using normal constructors that can then be
>> configured, probably by getters and setters to allow the method of obtaining
>> the information to be either local or remote, and the "game engine" won't
>> care which.
>
> Always local, in my concept.  Static info does not need to squeezed through the client/server interface.
> Unless you want to make the client really dumb, of course.  And slower.
>

Over-use of static info also makes supporting multiple threads more
problematic; it increases the likelihood of race conditions and other
difficult-to-debug issues. In general, most of the stuff that's
currently static doesn't really _need_ to be static.

> Erik.
>
>> So, my main point is that I'd like us to be conscious of which classes have
>> awareness of lower-level information (like XML parsing) and ensure that
>> knowledge is appropriately abstracted away from the game logic in the same
>> way that we try to separate the game logic from the display/view logic.
>>
>> ---Brett.
>

---Brett.

Re: [Rails-devel] 1880 Patch

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

I'll add that for any patches you send to the list, please send them
in "git diff" format.

I can apply them and am familiar with how git deals with them. Also,
they're simply more readable. It makes it easier to see what is
changing.

---Brett.

On Sat, Jul 23, 2011 at 10:15 AM, Stefan Frey <ste...@we...> wrote:
> Erik & Martin,
> just by chance I came upon this looking for other things:
>
> Inside this tutorial:
> http://unicase.blogspot.com/2011/01/egit-tutorial-for-beginners.html
>
> Take a look at the paragraph Creating Patch:
>
> It states that egit cannot import the git patch format.
> One has to use either a command line tool (do not ask which one)
> or should stay with the normal default diff-patch.
>
> The diff-patch format can be selected in the create patch dialog by NOT
> selecting the git patch format.
>
> However a patch can only contain exactly one commit, thus for several commits
> the commits have to be bundled using rebase before creating the patch.
>
> Stefan
>
>
> On Saturday, July 23, 2011 06:37:52 pm Erik Vos wrote:
>> This is a “git” patch, and for some reason this patch does not behave
>> correctly when I’m trying to apply it via Eclipse/Egit:
>>
>>
>>
>> diff --git a/data/1880/CompanyManager.xml b/data/1880/CompanyManager.xml
>>
>> index c02e2d2..cd8dc10 100644
>>
>> --- a/data/1880/CompanyManager.xml
>>
>> +++ b/data/1880/CompanyManager.xml
>>
>> … etc.
>>
>>
>>
>> It creates a subdirectory ‘a’ under the project root, and that’s not
>> correct.
>>
>>
>>
>> Yesterday I tried the same thing to apply some earlier work of myself to
>> the new repo master, and it failed similarly.
>>
>>
>>
>> Does anyone know how to deal with patches in Git/Egit?
>>
>> Perhaps we should use “normal” patches (that don’t have these a/ and b/
>> prefixes)?
>>
>> What are “git” patches useful for at all?
>>
>>
>>
>> Erik.
>>
>>
>>
>>
>>
>>
>>
>> From: Dr....@t-... [mailto:Dr....@t-...]
>> Sent: Saturday, July 23, 2011 5:09 PM
>> To: Rails Development
>> Subject: [Rails-devel] 1880 Patch
>>
>>
>>
>>  hi all,
>>
>>
>>
>> after having a heck of problems with eclipse and [e,j]git i managed to
>> untangle the knot with the external git implementation...
>>
>>
>>
>>
>>
>>
>>     Added changes to the map, introducing the tiles -4008 and -4009 for
>>     the water connections (ferries).
>>
>>     Altered the stockmarket display according to other examples with
>>     colored names for the additional revenue hexes.
>>
>>     Revised the Trains and Phase mechanismns according to the recent
>>     changes done to 1835.
>>
>>
>>
>> <<<<<
>>
>>
>>
>> And yes i'll keep tying to find out why i cant update the local clone from
>> the remote master and make a simple patch file in eclipse from my working
>> branch against an updated local clone...
>>
>>
>>
>> Workflow as intended:
>>
>>
>>
>> Cloned the remote origin into remote/master (local working clone)
>>
>> branched my 1880-devel-branch of the remote/master  (local working clone)
>>
>> edit files
>>
>> commit against 1880 devel branch (so far everything was okay...)
>>
>> updating local working clone via pull/fetch from remote origin (shouldnt
>> overwrite 1880 branch) (didnt work, couldnt get an updated history view of
>> remote/master that was consistent with the history of remote/origin)
>>
>> check for conflicts between 1880 devel branch (.....)
>>
>> draw patch of 1880 devel branch against remote master
>>
>>
>>
>> from here on i choose to merge the 1880 branch with my remote master (that
>> of course in the external git was somehow aligned with the
>> remote/origin...) in the external git and draw a patch against the last
>> state of the remote origin/master...
>>
>> branched a new development branch.(current state of confusion :))
>>
>>
>>
>> I wasnt able to create a small an simple patch from the egit/jgit inside
>> eclipse it created a monster patch over all files even though i had only
>> changed those 6. Probably something to do with the windows/unix CRLF/LF
>> problematic...
>>
>>
>>
>> And yes i if anybody has some advice dont hesitate even if it makes me look
>> like stupid :)
>>
>>
>>
>> Regards Martin
>
> ------------------------------------------------------------------------------
> Storage Efficiency Calculator
> This modeling tool is based on patent-pending intellectual property that
> has been used successfully in hundreds of IBM storage optimization engage-
> ments, worldwide.  Store less, Store more with what you own, Move data to
> the right place. Try It Now! http://www.accelacomm.com/jaw/sfnl/114/51427378/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

Re: [Rails-devel] Removal of static class variables (git: 8b9df8)

[Stefan F. <ste...@we...>]

Erik & Brett,
it seems to me it depends on what you the suggested client/server split looks 
like.

From what I understood from previous discussions is that Erik prefers
the fat client approach, where "clients" all have a full game back-end running 
and only pass around actions between them. The server merely acts as a message 
broker.

Another possibility is a client/server split between the game backend as the 
server and the clients only consisting of the GUI as defined by the swing 
classes of Rails. However it requires passing along the updates between the 
server and all connected clients.

It seems to me that Brett thinks of the latter separation, however I still do 
not understand why the configureFromXML method matters here as all Managers 
using configureFromXML belong to the server part. So I might be wrong and 
Brett has other ideas.

From my point of view I think that Erik's approach is the easiest to implement 
and has some advantages in UI responsiveness. It might also be better suited 
for a migration to mobile devices like Android tablets, which I think should 
be the longer-term goal, which are restricted in bandwidth and latency, but 
have enough computing power to run the game backend.

Stefan


On Saturday, July 23, 2011 05:56:20 pm Erik Vos wrote:
> > -----Original Message-----
> > From: brett lentz [mailto:bre...@gm...]
> > Sent: Saturday, July 23, 2011 5:36 PM
> > To: Development list for Rails: an 18xx game
> > Subject: Re: [Rails-devel] Removal of static class variables (git:
> > 8b9df8) To be clear, the thing I was trying to recommend against is
> > *just* the configureFromXML() method that we typically use to initialize
> > the *Manager classes.
> > 
> > That's the biggest point where a client/server model will differ.
> 
> Why?
> The client won't have the same managers, but why can't whatever will
> replace the managers call the same configureFromXML() methods of the
> objects that they need to know about?
> 
> > We'll need to instantiate the various *Manager classes, but those classes
> > will need to be instantiated using normal constructors that can then be
> > configured, probably by getters and setters to allow the method of
> > obtaining the information to be either local or remote, and the "game
> > engine" won't care which.
> 
> Always local, in my concept.  Static info does not need to squeezed through
> the client/server interface. Unless you want to make the client really
> dumb, of course.  And slower.
> 
> Erik.
> 
> > So, my main point is that I'd like us to be conscious of which classes
> > have awareness of lower-level information (like XML parsing) and ensure
> > that knowledge is appropriately abstracted away from the game logic in
> > the same way that we try to separate the game logic from the
> > display/view logic.
> > 
> > ---Brett.
> 
> ---------------------------------------------------------------------------
> --- Storage Efficiency Calculator
> This modeling tool is based on patent-pending intellectual property that
> has been used successfully in hundreds of IBM storage optimization engage-
> ments, worldwide.  Store less, Store more with what you own, Move data to
> the right place. Try It Now!
> http://www.accelacomm.com/jaw/sfnl/114/51427378/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

Re: [Rails-devel] 1880 Patch

[Stefan F. <ste...@we...>]

Erik & Martin,
just by chance I came upon this looking for other things:

Inside this tutorial:
http://unicase.blogspot.com/2011/01/egit-tutorial-for-beginners.html

Take a look at the paragraph Creating Patch:

It states that egit cannot import the git patch format. 
One has to use either a command line tool (do not ask which one)
or should stay with the normal default diff-patch. 

The diff-patch format can be selected in the create patch dialog by NOT 
selecting the git patch format. 

However a patch can only contain exactly one commit, thus for several commits 
the commits have to be bundled using rebase before creating the patch.

Stefan


On Saturday, July 23, 2011 06:37:52 pm Erik Vos wrote:
> This is a “git” patch, and for some reason this patch does not behave
> correctly when I’m trying to apply it via Eclipse/Egit:
> 
> 
> 
> diff --git a/data/1880/CompanyManager.xml b/data/1880/CompanyManager.xml
> 
> index c02e2d2..cd8dc10 100644
> 
> --- a/data/1880/CompanyManager.xml
> 
> +++ b/data/1880/CompanyManager.xml
> 
> … etc.
> 
> 
> 
> It creates a subdirectory ‘a’ under the project root, and that’s not 
> correct.
> 
> 
> 
> Yesterday I tried the same thing to apply some earlier work of myself to
> the new repo master, and it failed similarly.
> 
> 
> 
> Does anyone know how to deal with patches in Git/Egit?
> 
> Perhaps we should use “normal” patches (that don’t have these a/ and b/
> prefixes)?
> 
> What are “git” patches useful for at all?
> 
> 
> 
> Erik.
> 
> 
> 
> 
> 
> 
> 
> From: Dr....@t-... [mailto:Dr....@t-...]
> Sent: Saturday, July 23, 2011 5:09 PM
> To: Rails Development
> Subject: [Rails-devel] 1880 Patch
> 
> 
> 
>  hi all,
> 
> 
> 
> after having a heck of problems with eclipse and [e,j]git i managed to
> untangle the knot with the external git implementation...
> 
> 
> 
> 
> 
> 
>     Added changes to the map, introducing the tiles -4008 and -4009 for
>     the water connections (ferries).
> 
>     Altered the stockmarket display according to other examples with
>     colored names for the additional revenue hexes.
> 
>     Revised the Trains and Phase mechanismns according to the recent
>     changes done to 1835.
> 
> 
> 
> <<<<<
> 
> 
> 
> And yes i'll keep tying to find out why i cant update the local clone from
> the remote master and make a simple patch file in eclipse from my working
> branch against an updated local clone...
> 
> 
> 
> Workflow as intended:
> 
> 
> 
> Cloned the remote origin into remote/master (local working clone)
> 
> branched my 1880-devel-branch of the remote/master  (local working clone)
> 
> edit files
> 
> commit against 1880 devel branch (so far everything was okay...)
> 
> updating local working clone via pull/fetch from remote origin (shouldnt
> overwrite 1880 branch) (didnt work, couldnt get an updated history view of
> remote/master that was consistent with the history of remote/origin)
> 
> check for conflicts between 1880 devel branch (.....)
> 
> draw patch of 1880 devel branch against remote master
> 
> 
> 
> from here on i choose to merge the 1880 branch with my remote master (that
> of course in the external git was somehow aligned with the
> remote/origin...) in the external git and draw a patch against the last
> state of the remote origin/master...
> 
> branched a new development branch.(current state of confusion :))
> 
> 
> 
> I wasnt able to create a small an simple patch from the egit/jgit inside
> eclipse it created a monster patch over all files even though i had only
> changed those 6. Probably something to do with the windows/unix CRLF/LF
> problematic...
> 
> 
> 
> And yes i if anybody has some advice dont hesitate even if it makes me look
> like stupid :)
> 
> 
> 
> Regards Martin

Re: [Rails-devel] 1880 Patch

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

This is a “git” patch, and for some reason this patch does not behave correctly when I’m trying to apply it via Eclipse/Egit:

 

diff --git a/data/1880/CompanyManager.xml b/data/1880/CompanyManager.xml

index c02e2d2..cd8dc10 100644

--- a/data/1880/CompanyManager.xml

+++ b/data/1880/CompanyManager.xml

… etc.

 

It creates a subdirectory ‘a’ under the project root, and that’s not  correct. 

 

Yesterday I tried the same thing to apply some earlier work of myself to the new repo master, and it failed similarly.

 

Does anyone know how to deal with patches in Git/Egit?

Perhaps we should use “normal” patches (that don’t have these a/ and b/ prefixes)? 

What are “git” patches useful for at all?  

 

Erik.

 

 

 

From: Dr....@t-... [mailto:Dr....@t-...] 
Sent: Saturday, July 23, 2011 5:09 PM
To: Rails Development
Subject: [Rails-devel] 1880 Patch

 

 hi all,

 

after having a heck of problems with eclipse and [e,j]git i managed to untangle the knot with the external git implementation...

 

>>>> 


    Added changes to the map, introducing the tiles -4008 and -4009 for
    the water connections (ferries).
    
    Altered the stockmarket display according to other examples with
    colored names for the additional revenue hexes.
    
    Revised the Trains and Phase mechanismns according to the recent
    changes done to 1835.

 

<<<<< 

 

And yes i'll keep tying to find out why i cant update the local clone from the remote master and make a simple patch file in eclipse from my working branch against an updated local clone...

 

Workflow as intended:

 

Cloned the remote origin into remote/master (local working clone)

branched my 1880-devel-branch of the remote/master  (local working clone)

edit files

commit against 1880 devel branch (so far everything was okay...)

updating local working clone via pull/fetch from remote origin (shouldnt overwrite 1880 branch) (didnt work, couldnt get an updated history view of remote/master that was consistent with the history of remote/origin)

check for conflicts between 1880 devel branch (.....)

draw patch of 1880 devel branch against remote master

 

from here on i choose to merge the 1880 branch with my remote master (that of course in the external git was somehow aligned with the remote/origin...) in the external git and draw a patch against the last state of the remote origin/master...

branched a new development branch.(current state of confusion :))

 

I wasnt able to create a small an simple patch from the egit/jgit inside eclipse it created a monster patch over all files even though i had only changed those 6. Probably something to do with the windows/unix CRLF/LF problematic...

 

And yes i if anybody has some advice dont hesitate even if it makes me look like stupid :)

 

Regards Martin

 

 

 

Re: [Rails-devel] Removal of static class variables (git: 8b9df8)

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

> -----Original Message-----
> From: brett lentz [mailto:bre...@gm...]
> Sent: Saturday, July 23, 2011 5:36 PM
> To: Development list for Rails: an 18xx game
> Subject: Re: [Rails-devel] Removal of static class variables (git: 8b9df8)
> To be clear, the thing I was trying to recommend against is *just* the
> configureFromXML() method that we typically use to initialize the *Manager
> classes.
> 
> That's the biggest point where a client/server model will differ.

Why? 
The client won't have the same managers, but why can't whatever will replace the managers call the same configureFromXML() methods of the objects that they need to know about?

> We'll need to instantiate the various *Manager classes, but those classes will
> need to be instantiated using normal constructors that can then be
> configured, probably by getters and setters to allow the method of obtaining
> the information to be either local or remote, and the "game engine" won't
> care which.

Always local, in my concept.  Static info does not need to squeezed through the client/server interface.
Unless you want to make the client really dumb, of course.  And slower.

Erik.
 
> So, my main point is that I'd like us to be conscious of which classes have
> awareness of lower-level information (like XML parsing) and ensure that
> knowledge is appropriately abstracted away from the game logic in the same
> way that we try to separate the game logic from the display/view logic.
> 
> ---Brett.

Re: [Rails-devel] Hex trains support

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

> -----Original Message-----
> From: Chris Shaffer [mailto:chr...@gm...]
> Sent: Saturday, July 23, 2011 4:56 PM
> To: Development list for Rails: an 18xx game
> Subject: Re: [Rails-devel] Hex trains support

> Note that in some games, dual type trains retain their type when discarded
> to the pool.

That's new to me.  Which games are you aware of?

Erik.

Re: [Rails-devel] Removal of static class variables (git: 8b9df8)

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

On Sat, Jul 23, 2011 at 7:47 AM, Erik Vos <eri...@xs...> wrote:
>> -----Original Message-----
>> From: Stefan Frey [mailto:ste...@we...]
>
>> I have to admit that do not fully understand what Brett's concerns are: In
> my
>> understanding it will either be a fat client/lean server (means that each
>> instance has a full game model behind and the server component only
>> broker
>> messages) or a lean client/fat sever (means that the client only shows a
> view
>> of game model on the server) and in that case the client would have none
> of
>> the current Manager which are configured by xml.
>
> The current exchange of action objects is based on (what I think amounts to)
> yet another concept, that I mentioned a few days ago in the Game Notes
> thread:
>

To be clear, the thing I was trying to recommend against is *just* the
configureFromXML() method that we typically use to initialize the
*Manager classes.

That's the biggest point where a client/server model will differ.
We'll need to instantiate the various *Manager classes, but those
classes will need to be instantiated using normal constructors that
can then be configured, probably by getters and setters to allow the
method of obtaining the information to be either local or remote, and
the "game engine" won't care which.

So, my main point is that I'd like us to be conscious of which classes
have awareness of lower-level information (like XML parsing) and
ensure that knowledge is appropriately abstracted away from the game
logic in the same way that we try to separate the game logic from the
display/view logic.

---Brett.


>> -----Original Message-----
>> From: Erik Vos [mailto:eri...@xs...]
>> For what it's worth:
>>
>> My idea about client/server was that both sides would parse the XML and
>> thus would have access to all static (as opposed to dynamic) objects.
>> Only the dynamic (state) info would be in the server only.  Parsing at the
>> client would been initiated by much simpler (and probably less) managers,
>> probably not really worth that name.
>>
>> But I haven't really thought this through yet.
>
> The current communication between GUI and engine mainly consists of the
> action objects.  After a client/server split, the action objects would be
> exchanged in some serialized form,  and only contain the unique IDs of the
> relevant objects, not these objects themselves (which are declared
> transient).
> So the unique IDs have to be identical at both sides. I consider that a main
> design consideration for the client/server split.  It may put constraints on
> the parsing sequence at both sides.
>
> Erik.
>
>
>
> ------------------------------------------------------------------------------
> Storage Efficiency Calculator
> This modeling tool is based on patent-pending intellectual property that
> has been used successfully in hundreds of IBM storage optimization engage-
> ments, worldwide.  Store less, Store more with what you own, Move data to
> the right place. Try It Now! http://www.accelacomm.com/jaw/sfnl/114/51427378/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

Re: [Rails-devel] Hex trains support

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

BTW We should also keep this in mind:

> From: Chris Shaffer [mailto:chr...@gm...] 
> Sent: Wednesday, May 04, 2011 4:39 PM
> To: Development list for Rails: an 18xx game
> Subject: Re: [Rails-devel] Obsolete Trains

 > In 1844, when a new type of train is purchased, the previous rank are
converted into H trains.

So a train can be converted into a quite different type.  No big deal, I
think.

Erik.

> -----Original Message-----
> From: Stefan Frey [mailto:ste...@we...]
> Sent: Thursday, July 21, 2011 7:18 AM
> To: Development list for Rails: an 18xx game
> Subject: Re: [Rails-devel] Hex trains support
> 
> Erik,
> I have only checked the methods that train and TrainType support instead
of
> having a look at the xml files ;-)
> 
> I intend to keep it as closely to your current specification. So I agree
with your
> proposal.  Considering your specification I only wonder why major and
minor
> are an attribute of the TrainType tag instead of Reach?
> Maybe to keep the xml files more compact?
> 
> The only change will be that there is an additional hexes attribute inside
of
> the TrainType tag, which should be used for hex trains.
> 
> This attribute allows the separation of name and range (thus the hex train
in
> 1844 will be defined with a hexes attribute one less than their name).
> 
> Some implementation questions:
> 
> The main task is to add methods to support the variables and I will to
that.
> Just to be sure I understood your recent changes:
> 
> TrainCerticate: One for each certificate
> 
> TrainType: One for each type
> 
> Train: One per train certificate (standard case), two per train
certificate (dual
> case)
> 
> Train objects have a pointer to the TrainType object, which itself
contains all
> relevant attributes required for the revenue calculation.
> 
> For the case of operating I can safely ignore the dual case as the
operating
> company has the correct trains in its portfolio.
> 
> So I retrieve all trains for a company from the portfolio and for each
train get
> its TrainType and then query the TrainType for the attributes.
> 
> To make this more transparent I suggest to remove all methods related to
> range etc. from the Train class (and TrainI interface) and only keep those
in
> TrainType. I currently sometimes query the Train and sometimes the
> TrainType and this seemed somehow inconsistent.
> 
> Stefan
> 
> 
> On Wednesday, July 20, 2011 10:52:02 am Erik Vos wrote:
> > Stefan,
> >
> > My original intention was to set this attribute in the <Reach> tag.
> > See this fragment from 1830/Game.xml:
> >
> > 	<Reach base="stops" countTowns="yes"/>
> > 	<!-- Alternative values:
> > 		base="hexes" for H-trains as in 1826, 1849 etc.
> > 		countTowns="no" for all trains in 1841, 18EU, etc.,
> > 		  where towns score but do not count against the train
> length.
> > 		  Otherwise, towns are counted as minor or major stops,
> > 		  depending on the presence or absence of a "minorStops"
> > value.
> > 	-->
> >
> > This all dates back to the early days when everything was provisional.
> > It's basically up to you how to work this out for hex trains, but I'll
> > express my thoughts below.
> >
> > I admit that 'base' is not very descriptive, these days I would
> > probably have looked for a better word, something like 'reachBase' or
> > 'countingBase'. But I doubt if it's worthwhile to change it.
> >
> > For the hex count you could reuse 'majorStops', but it's nicer to call
> > it 'hexes'.  Internally these could point to the same integer variable
> > (or not). Keep in mind that in one game (1844) the hex count includes
> > the starting hex; in all other H-train games it is excluded. Either
> > another attribute must indicate this special case, or the official
> > 'hexes' counts must be decreased by one in this game.
> >
> > You could also drop 'base' and detect from the presence of
> > 'majorStops' or 'hexes' what type of train it is, but that's an
> > approach that may not be to everybody's liking.  In any case, these
> > two attributes should exclude each other.
> >
> > If you ask for my personal preferences, the answer would be:
> > - keep 'base'
> > - use 'hexes' as an alias for 'majorStops'
> > - don't add a special attribute for 1844.
> >
> > But, as said, it's up to you.
> >
> > Erik.
> >
> > > -----Original Message-----
> > > From: Stefan Frey [mailto:ste...@we...]
> > > Sent: Wednesday, July 20, 2011 8:18 AM
> > > To: Development list for Rails: an 18xx game
> > > Subject: [Rails-devel] Hex trains support
> > >
> > > Erik,
> > > I have implemented revenue calculation for hex trains (for 1826).
> > > Instead
> >
> > of a
> >
> > > dynamic modifier I kept the (older) plan to add that to the revenue
> >
> > calculator
> >
> > > directly. So I treat hex trains as "first-class" citizens of Rails
> > > and
> >
> > they should
> >
> > > be supported directly via attributes in XML.
> > >
> > > What is your preference for the naming and type of such an attribute?
> > > I would currently prefer a simple boolean isHexTrain = "yes" and
> > > (mis-)
> >
> > use
> >
> > > the number of majors as the number of hexes the train can travel.
> > >
> > > But maybe something like distanceType={stops|hexes} and then an
> > > integer attribute for hexDistance is preferable?
> > > Stefan
> >
> > ----------------------------------------------------------------------
> > -----
> > - --
> >
> > > 10 Tips for Better Web Security
> > > Learn 10 ways to better secure your business today. Topics covered
> >
> > include:
> > > Web security, SSL, hacker attacks & Denial of Service (DoS), private
> > > keys, security Microsoft Exchange, secure Instant Messaging, and
> > > much more. http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > > _______________________________________________
> > > Rails-devel mailing list
> > > Rai...@li...
> > > https://lists.sourceforge.net/lists/listinfo/rails-devel
> >
> > ----------------------------------------------------------------------
> > -----
> > --- 10 Tips for Better Web Security
> > Learn 10 ways to better secure your business today. Topics covered
> include:
> > Web security, SSL, hacker attacks & Denial of Service (DoS), private
> > keys, security Microsoft Exchange, secure Instant Messaging, and much
> more.
> > http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > _______________________________________________
> > Rails-devel mailing list
> > Rai...@li...
> > https://lists.sourceforge.net/lists/listinfo/rails-devel
> 
>
----------------------------------------------------------------------------
--
> 5 Ways to Improve & Secure Unified Communications Unified
> Communications promises greater efficiencies for business. UC can improve
> internal communications as well as offer faster, more efficient ways to
> interact with customers and streamline customer service. Learn more!
> http://www.accelacomm.com/jaw/sfnl/114/51426253/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

Re: [Rails-devel] Hex trains support

[Chris S. <chr...@gm...>]

>>> Train: One per train certificate (standard case), two per train
>> certificate (dual
>>> case)
>> 
>> No, one train per certificate.  Its initial type is null, will be set if the
>> train is bought, and unset if it is discarded to the Pool.

Note that in some games, dual type trains retain their type when discarded to the pool. 

--
Chris

Re: [Rails-devel] Removal of static class variables (git: 8b9df8)

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

> -----Original Message-----
> From: Stefan Frey [mailto:ste...@we...]

> I have to admit that do not fully understand what Brett's concerns are: In
my
> understanding it will either be a fat client/lean server (means that each
> instance has a full game model behind and the server component only
> broker
> messages) or a lean client/fat sever (means that the client only shows a
view
> of game model on the server) and in that case the client would have none
of
> the current Manager which are configured by xml.

The current exchange of action objects is based on (what I think amounts to)
yet another concept, that I mentioned a few days ago in the Game Notes
thread:

> -----Original Message-----
> From: Erik Vos [mailto:eri...@xs...]
> For what it's worth:
> 
> My idea about client/server was that both sides would parse the XML and
> thus would have access to all static (as opposed to dynamic) objects.
> Only the dynamic (state) info would be in the server only.  Parsing at the
> client would been initiated by much simpler (and probably less) managers,
> probably not really worth that name.
> 
> But I haven't really thought this through yet.

The current communication between GUI and engine mainly consists of the
action objects.  After a client/server split, the action objects would be
exchanged in some serialized form,  and only contain the unique IDs of the
relevant objects, not these objects themselves (which are declared
transient). 
So the unique IDs have to be identical at both sides. I consider that a main
design consideration for the client/server split.  It may put constraints on
the parsing sequence at both sides.

Erik.

Re: [Rails-devel] Git troubleshooting

[Stefan F. <ste...@we...>]

Even after the few days I love the easy switch between branches. It is much 
easier to simply test if things worked before your change, how the code looked 
like at the latest tag etc. Overall I can already see an improvement.

And I agree that it seems a bit strange that egit pretends to not knowing that 
the remote repo you pushed to is not at the same commit like the repo you 
pushed from and you still have to pull from there to get that information.

One observation for egit however is: If there is uncommitted code in a branch 
and you switch to another branch where the local changes are compatible (thus 
the files you edited have the same version in the two branches) the local 
changes simply carry over. 
Only after a commit before the switch the files revert to the state of the 
other branch as expected.

It makes sense from a git point of view, but it still can surprise you first.
If the files you have edited in the other branch are not compatible egit 
raises a warning, that the switch will delete your local uncommitted changes.

Stefan


On Friday, July 22, 2011 09:49:41 pm brett lentz wrote:
> That makes a *lot* more sense.
> 
> I suspect there's probably an easier way to achieve the same result,
> but I'll save that for a later discussion. ;-)
> 
> I'm happy to see the issues with Git are starting to be sorted out. It
> sounds like the Windows implementation is lagging behind the other
> platforms, but is still usable.
> 
> ---Brett.
> 
> On Fri, Jul 22, 2011 at 12:25 PM, Erik Vos <eri...@xs...> wrote:
> > Good, I have recloned the repository and done a few commits.  Seems to
> > work so far.
> > One thing is a bit curious: after the push I need to do a pull to get my
> > local status fully updated (in Egit as well as in the GitBash CLI).
> > I suppose that's the way it is...
> > 
> > So far I have only worked in and committed to the master branch. I
> > suppose that's part of why I got into problems earlier.  I probably need
> > to educate myself on this aspect....
> > 
> > 
> > The commits revert Brett's Tag change that have I protested against, and
> > I have added some hopefully helpful comments to clarify the original
> > code.
> > 
> > I have also found why that code exists in Tag.  The game option
> > "UnlimitedTopTrains" is parametrized with a value "D" to allow it to be
> > displayed as "Unlimited D-trains".
> > This way we don't need a separate name and corresponding entry in
> > LocalizedText.properties for each different train type for which that
> > option is used.
> > 
> > To make that parameter persistent, this option is saved with the name
> > "UnlimitedTopTrains_D". To find its value, the code must therefore find a
> > saved option name that starts with the unparametrized name, and that's
> > what happens in that code.  So it's not obsolete at all.
> > 
> > And if this code does find a value, the 'if' condition value has changed
> > between its first and second check, so these ifs cannot be merged.
> > 
> > Erik.
> > 
> >> -----Original Message-----
> >> From: brett lentz [mailto:bre...@gm...]
> >> Sent: Thursday, July 21, 2011 8:00 PM
> >> To: Development list for Rails: an 18xx game
> >> Subject: Re: [Rails-devel] Git troubleshooting
> >> 
> >> Ok... I've made a few small commits with EGit, and I think I'm getting
> >> the hang of the interface. It's a bit clumsy, but it seems to do
> >> everything I
> > 
> > expect.
> > 
> >> The EGit user guide is a fairly lengthy document, but as Stefan noted
> > 
> > earlier,
> > 
> >> it's got a lot of important details that will help you learn the
> > 
> > interface:
> >> http://wiki.eclipse.org/EGit/User_Guide
> >> 
> >> I think everyone should probably take a step back. I have a feeling we
> > 
> > tried
> > 
> >> to run before we learned to walk. ;-)
> >> 
> >> Set aside the work you've done, do a clean clone of the repo in a new
> >> location following the EGit User Guide. Then, find some small one-line
> >> change to make. Commit it, push it, as directed by the User Guide.
> >> Let's try to get everyone to commit and successfully push a few small
> >> changes before we try to bite off anything bigger. Once we've got a few
> >> pushes and pulls done, we can come back to the bigger pieces of work
> >> that are being done.
> >> 
> >> Next...
> >> 
> >> Also, in your git configs (usually a .gitconfig file in your home
> > 
> > directory), it's
> > 
> >> very handy to default to always rebasing incoming changes. Merge commits
> >> are supposed to be meaningful, but the default behavior is to always
> >> merge incoming changes (and create a merge commit for them). The
> >> tradeoff here is that rebasing changes the commit history, which is
> >> potentially
> > 
> > dangerous
> > 
> >> and less obvious than adding a bunch of extraneous merge commits every
> >> time you do a pull.
> >> 
> >> You can configure certain branches to always do this without the
> >> --rebase flag:
> >> 
> >> # make `git pull` on master always use rebase $ git config
> >> branch.master.rebase true You can also set up a global option to set the
> > 
> > last
> > 
> >> property for every new tracked branch:
> >> 
> >> # setup rebase for every tracking branch $ git config --global
> >> branch.autosetuprebase always
> >> 
> >> The .gitconfig is an ini-style file, so the above is also represented as
> > 
> > this if you
> > 
> >> want to update the file directly:
> >> 
> >> [branch]
> >>    autosetuprebase = always
> >> 
> >> 
> >> Finally...
> >> 
> >> I'm happy to continue lending support as y'all run into troubles. It
> >> took
> > 
> > me a
> > 
> >> while to fully acclimate to Git's workflow, too. It's different, but
> >> very powerful once we get past the first few hurdles.
> >> 
> >> :-)
> >> 
> >> ---Brett.
> > 
> > -------------------------------------------------------------------------
> > --- --
> > 
> >> 5 Ways to Improve & Secure Unified Communications Unified
> >> Communications promises greater efficiencies for business. UC can
> >> improve internal communications as well as offer faster, more efficient
> >> ways to interact with customers and streamline customer service. Learn
> >> more! http://www.accelacomm.com/jaw/sfnl/114/51426253/
> >> _______________________________________________
> >> Rails-devel mailing list
> >> Rai...@li...
> >> https://lists.sourceforge.net/lists/listinfo/rails-devel
> > 
> > -------------------------------------------------------------------------
> > ----- 10 Tips for Better Web Security
> > Learn 10 ways to better secure your business today. Topics covered
> > include: Web security, SSL, hacker attacks & Denial of Service (DoS),
> > private keys, security Microsoft Exchange, secure Instant Messaging, and
> > much more. http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > _______________________________________________
> > Rails-devel mailing list
> > Rai...@li...
> > https://lists.sourceforge.net/lists/listinfo/rails-devel
> 
> ---------------------------------------------------------------------------
> --- 10 Tips for Better Web Security
> Learn 10 ways to better secure your business today. Topics covered include:
> Web security, SSL, hacker attacks & Denial of Service (DoS), private keys,
> security Microsoft Exchange, secure Instant Messaging, and much more.
> http://www.accelacomm.com/jaw/sfnl/114/51426210/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

[Rails-devel] Removal of static class variables (git: 8b9df8)

[Stefan F. <ste...@we...>]

With respect to Brett wishes below I decided to not add further Managers, but 
instead added a storage possibility to the GameManager.

However I tried to make this a general storage for all kind of objects instead 
of adding two specific storages.
Due to the fact that Token stores ids as Strings ("Token_#") with indices 
starting at zero and SpecialProperties as Integer starting at one, there is 
some legacy support code which potentially could be remove when
we decide to break save file compatibility.

I have to admit that do not fully understand what Brett's concerns are: In my 
understanding it will either be a fat client/lean server (means that each 
instance has a full game model behind and the server component only broker 
messages) or a lean client/fat sever (means that the client only shows a view 
of game model on the server) and in that case the client would have none of 
the current Manager which are configured by xml.

Remark: To work with my IMAP based e-mail client the search using git ids only 
works if they are added to the subject efficiently.


On Thursday, July 21, 2011 08:09:28 am brett lentz wrote:
> Yup, that sounds like it might be necessary if one of the existing
> *Managers isn't a good fit.
> 
> I'm trying to look ahead to our dream of client/server separation
> where I expect the server will handle all the XML reading and pass
> it's authoritative information to the client. In this case, the client
> side will need to be able to instantiate objects without calling the
> ConfigureFromXML() methods in each of the *Managers.
> 
> So, if your implementation of new *Managers involves new
> ConfigureFromXML() methods, I'd like to suggest this may be the time
> to work on an alternative initialization methodology. It seems that
> GameInfoParser might be part of the solution, but I'm not certain.
> 
> Unravelling the current game init code is taking me more time than I like.
> 
> ---Brett.
> 

[Rails-devel] Phase management: train limits

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

(Although this is not a small commit, I couldn't resist the urge to include
the phase management updates that I have been working on earlier this week.
This time I created a branch, and merged it into my master - and only then
realised that I had to push it immediately to avoid getting into out-of-sync
problems again. Or not?)

Phase management upgrade step 3: train limits.
    
<Phase><Trains> now has a new attribute 'limitStep', which starts at 1.
<Company[Type]><Trains> now lists the number of trains per step, not per
phase.
These changes have been applied to 1835 ONLY. Temporary code ensures that
the old way still works for other games.
    
<Phase> has another new optional attribute 'realName', intended to have the
'official' name of a phase, as in the physical game. This real name is used
in displaying the current phase, see below.
    
1835 new-style phase example:
  <Phase name="4" realName="2">
    <Trains limitStep="2" rusted="2"/>
  </Phase>
In the game status window this phase is now displayed as "4 [2]".
The major CompanyType now has
  <Trains limit="4,3,2"/>
    
This change also fixes the 1835 bug that the train limit was reduced to 2 at
buying the 4+4-train instead of at buying the 5-train, as reported by John
David Galt in his recent playtest report.

Erik.

Re: [Rails-devel] Git troubleshooting

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

That makes a *lot* more sense.

I suspect there's probably an easier way to achieve the same result,
but I'll save that for a later discussion. ;-)

I'm happy to see the issues with Git are starting to be sorted out. It
sounds like the Windows implementation is lagging behind the other
platforms, but is still usable.

---Brett.



On Fri, Jul 22, 2011 at 12:25 PM, Erik Vos <eri...@xs...> wrote:
> Good, I have recloned the repository and done a few commits.  Seems to work
> so far.
> One thing is a bit curious: after the push I need to do a pull to get my
> local status fully updated (in Egit as well as in the GitBash CLI).
> I suppose that's the way it is...
>
> So far I have only worked in and committed to the master branch. I suppose
> that's part of why I got into problems earlier.  I probably need to educate
> myself on this aspect....
>
>
> The commits revert Brett's Tag change that have I protested against, and I
> have added some hopefully helpful comments to clarify the original code.
>
> I have also found why that code exists in Tag.  The game option
> "UnlimitedTopTrains" is parametrized with a value "D" to allow it to be
> displayed as "Unlimited D-trains".
> This way we don't need a separate name and corresponding entry in
> LocalizedText.properties for each different train type for which that option
> is used.
>
> To make that parameter persistent, this option is saved with the name
> "UnlimitedTopTrains_D". To find its value, the code must therefore find a
> saved option name that starts with the unparametrized name, and that's what
> happens in that code.  So it's not obsolete at all.
>
> And if this code does find a value, the 'if' condition value has changed
> between its first and second check, so these ifs cannot be merged.
>
> Erik.
>
>> -----Original Message-----
>> From: brett lentz [mailto:bre...@gm...]
>> Sent: Thursday, July 21, 2011 8:00 PM
>> To: Development list for Rails: an 18xx game
>> Subject: Re: [Rails-devel] Git troubleshooting
>>
>> Ok... I've made a few small commits with EGit, and I think I'm getting the
>> hang of the interface. It's a bit clumsy, but it seems to do everything I
> expect.
>> The EGit user guide is a fairly lengthy document, but as Stefan noted
> earlier,
>> it's got a lot of important details that will help you learn the
> interface:
>> http://wiki.eclipse.org/EGit/User_Guide
>>
>> I think everyone should probably take a step back. I have a feeling we
> tried
>> to run before we learned to walk. ;-)
>>
>> Set aside the work you've done, do a clean clone of the repo in a new
>> location following the EGit User Guide. Then, find some small one-line
>> change to make. Commit it, push it, as directed by the User Guide.
>> Let's try to get everyone to commit and successfully push a few small
>> changes before we try to bite off anything bigger. Once we've got a few
>> pushes and pulls done, we can come back to the bigger pieces of work that
>> are being done.
>>
>> Next...
>>
>> Also, in your git configs (usually a .gitconfig file in your home
> directory), it's
>> very handy to default to always rebasing incoming changes. Merge commits
>> are supposed to be meaningful, but the default behavior is to always merge
>> incoming changes (and create a merge commit for them). The tradeoff here
>> is that rebasing changes the commit history, which is potentially
> dangerous
>> and less obvious than adding a bunch of extraneous merge commits every
>> time you do a pull.
>>
>> You can configure certain branches to always do this without the --rebase
>> flag:
>>
>> # make `git pull` on master always use rebase $ git config
>> branch.master.rebase true You can also set up a global option to set the
> last
>> property for every new tracked branch:
>>
>> # setup rebase for every tracking branch $ git config --global
>> branch.autosetuprebase always
>>
>> The .gitconfig is an ini-style file, so the above is also represented as
> this if you
>> want to update the file directly:
>>
>> [branch]
>>    autosetuprebase = always
>>
>>
>> Finally...
>>
>> I'm happy to continue lending support as y'all run into troubles. It took
> me a
>> while to fully acclimate to Git's workflow, too. It's different, but very
>> powerful once we get past the first few hurdles.
>> :-)
>>
>> ---Brett.
>>
>>
> ----------------------------------------------------------------------------
> --
>> 5 Ways to Improve & Secure Unified Communications Unified
>> Communications promises greater efficiencies for business. UC can improve
>> internal communications as well as offer faster, more efficient ways to
>> interact with customers and streamline customer service. Learn more!
>> http://www.accelacomm.com/jaw/sfnl/114/51426253/
>> _______________________________________________
>> Rails-devel mailing list
>> Rai...@li...
>> https://lists.sourceforge.net/lists/listinfo/rails-devel
>
>
> ------------------------------------------------------------------------------
> 10 Tips for Better Web Security
> Learn 10 ways to better secure your business today. Topics covered include:
> Web security, SSL, hacker attacks & Denial of Service (DoS), private keys,
> security Microsoft Exchange, secure Instant Messaging, and much more.
> http://www.accelacomm.com/jaw/sfnl/114/51426210/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
>

Re: [Rails-devel] Git troubleshooting

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

Good, I have recloned the repository and done a few commits.  Seems to work
so far.
One thing is a bit curious: after the push I need to do a pull to get my
local status fully updated (in Egit as well as in the GitBash CLI).
I suppose that's the way it is...

So far I have only worked in and committed to the master branch. I suppose
that's part of why I got into problems earlier.  I probably need to educate
myself on this aspect....


The commits revert Brett's Tag change that have I protested against, and I
have added some hopefully helpful comments to clarify the original code.

I have also found why that code exists in Tag.  The game option
"UnlimitedTopTrains" is parametrized with a value "D" to allow it to be
displayed as "Unlimited D-trains".
This way we don't need a separate name and corresponding entry in
LocalizedText.properties for each different train type for which that option
is used.

To make that parameter persistent, this option is saved with the name
"UnlimitedTopTrains_D". To find its value, the code must therefore find a
saved option name that starts with the unparametrized name, and that's what
happens in that code.  So it's not obsolete at all.

And if this code does find a value, the 'if' condition value has changed
between its first and second check, so these ifs cannot be merged.

Erik.

> -----Original Message-----
> From: brett lentz [mailto:bre...@gm...]
> Sent: Thursday, July 21, 2011 8:00 PM
> To: Development list for Rails: an 18xx game
> Subject: Re: [Rails-devel] Git troubleshooting
> 
> Ok... I've made a few small commits with EGit, and I think I'm getting the
> hang of the interface. It's a bit clumsy, but it seems to do everything I
expect.
> The EGit user guide is a fairly lengthy document, but as Stefan noted
earlier,
> it's got a lot of important details that will help you learn the
interface:
> http://wiki.eclipse.org/EGit/User_Guide
> 
> I think everyone should probably take a step back. I have a feeling we
tried
> to run before we learned to walk. ;-)
> 
> Set aside the work you've done, do a clean clone of the repo in a new
> location following the EGit User Guide. Then, find some small one-line
> change to make. Commit it, push it, as directed by the User Guide.
> Let's try to get everyone to commit and successfully push a few small
> changes before we try to bite off anything bigger. Once we've got a few
> pushes and pulls done, we can come back to the bigger pieces of work that
> are being done.
> 
> Next...
> 
> Also, in your git configs (usually a .gitconfig file in your home
directory), it's
> very handy to default to always rebasing incoming changes. Merge commits
> are supposed to be meaningful, but the default behavior is to always merge
> incoming changes (and create a merge commit for them). The tradeoff here
> is that rebasing changes the commit history, which is potentially
dangerous
> and less obvious than adding a bunch of extraneous merge commits every
> time you do a pull.
> 
> You can configure certain branches to always do this without the --rebase
> flag:
> 
> # make `git pull` on master always use rebase $ git config
> branch.master.rebase true You can also set up a global option to set the
last
> property for every new tracked branch:
> 
> # setup rebase for every tracking branch $ git config --global
> branch.autosetuprebase always
> 
> The .gitconfig is an ini-style file, so the above is also represented as
this if you
> want to update the file directly:
> 
> [branch]
>    autosetuprebase = always
> 
> 
> Finally...
> 
> I'm happy to continue lending support as y'all run into troubles. It took
me a
> while to fully acclimate to Git's workflow, too. It's different, but very
> powerful once we get past the first few hurdles.
> :-)
> 
> ---Brett.
> 
>
----------------------------------------------------------------------------
--
> 5 Ways to Improve & Secure Unified Communications Unified
> Communications promises greater efficiencies for business. UC can improve
> internal communications as well as offer faster, more efficient ways to
> interact with customers and streamline customer service. Learn more!
> http://www.accelacomm.com/jaw/sfnl/114/51426253/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

Re: [Rails-devel] Git troubleshooting

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

On Fri, Jul 22, 2011 at 7:02 AM, Phil Davies <de...@gm...> wrote:
> On 21 July 2011 20:27, brett lentz <bre...@gm...> wrote:
>> Yes. Unfortunately, this is one of the down sides to EGit. For better
>> or worse, they decided to use a complete re-implementation of Git in
>> Java (JGit) instead of just hooking into the pre-existing tools and
>> libraries. It's not the choice I'd have made, but that's what they
>> went with. I'm certain it'll get better over time. Judging by the
>> mailing list, the project is under active development, so we should
>> see fixes relatively soon. The project has already come a long way in
>> the last year (when we opted for CVS->SVN instead of CVS->Git) and I'm
>> confident these issues will be sorted relatively soon.
>
> If you are using IDEA instead of Eclipse, that does use the standard
> Git client.  I've managed to pull the source locally but still finding
> time to go over the git documents and understand exactly what I need
> to understand.  Thankfully I don't have commit rights, this is
> probably for the best :)
>
> Phil

Yep, your workflow is much simpler:

1. Do work in a topic branch, committing as usual.
2. "git format-patch" when you're ready.
3. Post patches to this list. ;-)


---Brett.

Re: [Rails-devel] Git troubleshooting

[Phil D. <de...@gm...>]

On 21 July 2011 20:27, brett lentz <bre...@gm...> wrote:
> Yes. Unfortunately, this is one of the down sides to EGit. For better
> or worse, they decided to use a complete re-implementation of Git in
> Java (JGit) instead of just hooking into the pre-existing tools and
> libraries. It's not the choice I'd have made, but that's what they
> went with. I'm certain it'll get better over time. Judging by the
> mailing list, the project is under active development, so we should
> see fixes relatively soon. The project has already come a long way in
> the last year (when we opted for CVS->SVN instead of CVS->Git) and I'm
> confident these issues will be sorted relatively soon.

If you are using IDEA instead of Eclipse, that does use the standard
Git client.  I've managed to pull the source locally but still finding
time to go over the git documents and understand exactly what I need
to understand.  Thankfully I don't have commit rights, this is
probably for the best :)

Phil

Re: [Rails-devel] Hex trains support

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

> -----Original Message-----
> From: Stefan Frey [mailto:ste...@we...]
> I have only checked the methods that train and TrainType support instead
of
> having a look at the xml files ;-)
> 
> I intend to keep it as closely to your current specification. So I agree
with your
> proposal.  Considering your specification I only wonder why major and
minor
> are an attribute of the TrainType tag instead of Reach?
> Maybe to keep the xml files more compact?

That might be part of the answer, but I think the real reason was very
practical: the number of stops cannot be defaulted, whereas the <Reach> and
<Score> attributes usually are identical for all trains and so only a
game-wide default value is needed. 

Surely it is not very consistent this way: I could as well have added 'base'
and 'countTowns' etc. as attributes to a <DefaultType> tag.  I think I
created <Reach> and <Score> as separate tags because I felt the attributes
of these tags could better be split than live together as attributes of just
one tag.

I'm not against aligning these things either way, but I don't consider it
very important.


> The only change will be that there is an additional hexes attribute inside
of
> the TrainType tag, which should be used for hex trains.
> 
> This attribute allows the separation of name and range (thus the hex train
in
> 1844 will be defined with a hexes attribute one less than their name).
> 
> Some implementation questions:
> 
> The main task is to add methods to support the variables and I will to
that.
> Just to be sure I understood your recent changes:
> 
> TrainCerticate: One for each certificate
> 
> TrainType: One for each type

Correct so far.

> Train: One per train certificate (standard case), two per train
certificate (dual
> case)

No, one train per certificate.  Its initial type is null, will be set if the
train is bought, and unset if it is discarded to the Pool.

> Train objects have a pointer to the TrainType object, which itself
contains all
> relevant attributes required for the revenue calculation.

... if not null.

> For the case of operating I can safely ignore the dual case as the
operating
> company has the correct trains in its portfolio.
> 
> So I retrieve all trains for a company from the portfolio and for each
train get
> its TrainType and then query the TrainType for the attributes.
> 
> To make this more transparent I suggest to remove all methods related to
> range etc. from the Train class (and TrainI interface) and only keep those
in
> TrainType. I currently sometimes query the Train and sometimes the
> TrainType and this seemed somehow inconsistent.

Agreed.

Erik.

Re: [Rails-devel] [Rails-commits] rails/common

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

> -----Original Message-----
> From: brett lentz [mailto:bre...@gm...]
> My thinking is that two back-to-back if-statements that are checking the
> same condition is redundant and unnecessary. You've checked the condition
> once and have a code block for it. There's no need to do the exact same
> check again immediately afterward. 

Unless the condition may have changed in the meantime, which can be the case here if the first check is still needed at all.  

I'll check what the effect of removing it will be, and if it's harmless to my many old save files, I'll remove that code.

Erik.

Re: [Rails-devel] 1856 bridge private token

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

Stefan,

Great analysis.  

Historically, type A (low-level) and type B (high-level) moves have
developed in parallel, and initially type B moves predominated indeed.  
Over time we discovered more and more simple variables that had to be turned
into state variables, and that is probably when things got mixed up.

One main reason that I created these "high-level" type B moves with all
associated addObject and removeObject  and similar methods was to avoid the
need to create stateful collections.
For those cases where it was unavoidable, I created AddToList,
DoubleMapChange and similar low-level (type A) moves.
Another reason was, I think, to keep the debug log readable: it's often more
helpful to read that a token has moved, than that multiple lists have been
updated.  But I must admit that this way we lose a lot of detail that we
need in debugging undo/redo problems like this one.  For debug logging, I
think we need both levels of detail.

Now that you have created a number of stateful collections anyway,
disentangling the mess by doing away with type B moves comes in sight.  
I'm not against that, but it'll need a lot of care.  And I still would like
to see messages logged on the type B level (on initial execution; we'll lose
such messages on undo and redo).

Just a thought aside: wouldn't it be a good idea to include the currently
executed player action into each MoveSet? So that we can log *which* player
action is being undone/redone?

Erik.


> -----Original Message-----
> From: Stefan Frey [mailto:ste...@we...]
> Sent: Friday, July 22, 2011 8:43 AM
> To: Development list for Rails: an 18xx game
> Subject: Re: [Rails-devel] 1856 bridge private token
> 
> I took the opportunity to go through old mails and this bug proved to be
an
> interesting problem. It might even be the root of some strange, but
reported
> behavior with the undo/redo mechanisms. As Rails does not save undo/redo
> actions directly those are usually difficult to recreate.
> 
> Warning: The text below most likely is only of interest for those with
good
> knowledge of Rails internal working.
> 
> This undo issue of the save file (buying and scoring bonuses twice) was
> difficult to track down, as the usual suspects of the undo problem all
proved
> to be innocent.
> 
> In this case the undo/redo mechanisms fires twice. There are two
> approaches currently in Rails to implement state/moves in parallel. I will
call
> them type A States/Moves and type B States/Moves.
> 
> (From a chronological point of view I assume that B came before A, but I
am
> not sure).
> 
> Type A States/Moves
> 
> Create Move objects that change simple java data types/objects (like
> boolean, integer, ArrayList). The Move objects required are encapsulated
> inside the State classes (BooleanState, IntegerState, ArrayListState).
Thus to
> add undo/redo one (simply) replaces the usual java type with the state
type
> and everything works.
> 
> Type B States/Moves
> 
> Define a set of actions that are related to a specific Rails activity
(e.g.
> moving a Token from the bank to a player). For those are Move objects (in
> this example ObjectMove) defined that take care of all the actions
required
> by calling the appropriate methods from the effected objects. Those
objects
> are not State classes themselves (for example all the lists inside the
class
> Portfolio).
> 
> The issue what happened in the case of the 1856 bridge token now is that
> the token itself is moved via an ObjectMove, which calls the appropriate
> addition/removal code of the bonus token. However most of the relevant
> variables there are already changed to behave as state variables
themselves.
> 
> Thus in effect all Moves to add and remove tokens are created twice (once
> via the ObjectMove and once via the according Moves on the java state
> types).
> On the first undo Rails now tries to undo things twice (which usually
fails as
> they were executed only once) and on a redo it redos things twice.
> 
> Thus it seems that mixing types A and B is dangerous.
> 
> At least the recommendation inside the comments of type B Moves to take
> care of side effects is wrong: One only has to take care of side effects
which
> effect variables that are not themselves type A state variables.
> 
> Given that I am planning and preparing a refactoring of the State/Move
> approach (to include the trigger mechanism, but also to remove the usage
of
> Move creations in the main code to change ArrayListStates etc.) I suggest
to
> remove the type B moves altogether and replace them by changing the plain
> variables effected to type A  states.
> 
> To be sure: Before I will do this I would like to hear you others pros and
cons.
> And I would like to extend the automated game testing to allow testing of
> undos/redos as well (by allowing Rails to save and store undos/redos if a
> specific flag is set).
> 
> However one of the main advantages will be that it allows address at least
> some of those problems which arose for heavy gamers, where no one was
> able to recreate the exact sequence of undos/redos.
> 
> But for now it will be better to have a release first with all those
finished and
> nearly finished games out in the public, after we got git running ;-)
Stefan
> 
> 
> On Monday, May 16, 2011 11:12:46 pm Erik Vos wrote:
> > Phil,
> >
> > If I replay from the point just before buying the Tunnel, after a Save
and
> > Load at that point, it is OK. Buying a second Bonus token is not
allowed.
> > I don't see how you ever could have managed to do that.  Has any Undoing
> > happened around that time?  Clearly Undo still has its weaknesses.
> >
> > Erik.
> >
> > > -----Oorspronkelijk bericht-----
> > > Van: Phil Davies [mailto:de...@gm...]
> > > Verzonden: maandag 16 mei 2011 16:50
> > > Aan: Development list for Rails: an 18xx game
> > > Onderwerp: Re: [Rails-devel] 1856 bridge private token
> > >
> > > Actually, don't bother investigating this, I've backtracked and worked
> > > out what's wrong.  Somehow I managed to get LPS to buy a bonus bridge
> > > token from itself!  Rails should prevent this but I hardly count this
as
> >
> > critical rules
> >
> > > enforcement and is clearly a player instigated error,
> > >
> > > Phil
> > >
> > > On 16 May 2011 15:40, Phil Davies <de...@gm...> wrote:
> > > > Something odd is happening with the bridge private
> > > >
> > > > Load up the attached game, for a start, LPS has the bridge token
> > > > showing twice in it's tokens list.  Secondly, if you try and rewind
> > > > the game back to an earlier state (try OR 1, when LPS first runs)
and
> > > > without even playing forward, click again on a more up to date entry
> > > > in the game report log, and you will see another token get added to
> > > > the list that LPS has.
> > > >
> > > > The latter problem isn't really that bad as it doesn't persist
across
> > > > saves, but it does imply an issue with the undo process.  The former
> > > > issue (LPS having two tokens on reload) is more of a problem -
> > > > although fixable if the players are aware of it
> > > >
> > > > Phil
> >
> >
---------------------------------------------------------------------------
> > - --
> >
> > > Achieve unprecedented app performance and reliability What every
> C/C++
> > > and Fortran developer should know.
> > > Learn how Intel has extended the reach of its next-generation tools to
> >
> > help
> >
> > > boost performance applications - inlcuding clusters.
> > > http://p.sf.net/sfu/intel-dev2devmay
> > > _______________________________________________
> > > Rails-devel mailing list
> > > Rai...@li...
> > > https://lists.sourceforge.net/lists/listinfo/rails-devel
> >
> >
---------------------------------------------------------------------------
> > --- Achieve unprecedented app performance and reliability
> > What every C/C++ and Fortran developer should know.
> > Learn how Intel has extended the reach of its next-generation tools
> > to help boost performance applications - inlcuding clusters.
> > http://p.sf.net/sfu/intel-dev2devmay
> > _______________________________________________
> > Rails-devel mailing list
> > Rai...@li...
> > https://lists.sourceforge.net/lists/listinfo/rails-devel
> 
>
----------------------------------------------------------------------------
--
> 10 Tips for Better Web Security
> Learn 10 ways to better secure your business today. Topics covered
include:
> Web security, SSL, hacker attacks & Denial of Service (DoS), private keys,
> security Microsoft Exchange, secure Instant Messaging, and much more.
> http://www.accelacomm.com/jaw/sfnl/114/51426210/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

Re: [Rails-devel] 1856 bridge private token

[Stefan F. <ste...@we...>]

I took the opportunity to go through old mails and this bug proved to be 
an interesting problem. It might even be the root of some strange, but 
reported behavior with the undo/redo mechanisms. As Rails does not save 
undo/redo actions directly those are usually difficult to recreate.

Warning: The text below most likely is only of interest for those with good 
knowledge of Rails internal working.

This undo issue of the save file (buying and scoring bonuses twice) was 
difficult to track down, as the usual suspects of the undo problem all proved 
to be innocent.

In this case the undo/redo mechanisms fires twice. There are two approaches 
currently in Rails to implement state/moves in parallel. I will call them 
type A States/Moves and type B States/Moves.

(From a chronological point of view I assume that B came before A, but I am 
not sure).

Type A States/Moves

Create Move objects that change simple java data types/objects (like boolean, 
integer, ArrayList). The Move objects required are encapsulated inside the 
State classes (BooleanState, IntegerState, ArrayListState). Thus to add 
undo/redo one (simply) replaces the usual java type with the state type and 
everything works.

Type B States/Moves

Define a set of actions that are related to a specific Rails activity (e.g. 
moving a Token from the bank to a player). For those are Move objects (in this 
example ObjectMove) defined that take care of all the actions required by 
calling the appropriate methods from the effected objects. Those objects are
not State classes themselves (for example all the lists inside the class 
Portfolio).

The issue what happened in the case of the 1856 bridge token now is that the 
token itself is moved via an ObjectMove, which calls the appropriate 
addition/removal code of the bonus token. However most of the relevant 
variables there are already changed to behave as state variables themselves.

Thus in effect all Moves to add and remove tokens are created twice (once via 
the ObjectMove and once via the according Moves on the java state types).
On the first undo Rails now tries to undo things twice (which usually fails as 
they were executed only once) and on a redo it redos things twice.

Thus it seems that mixing types A and B is dangerous.

At least the recommendation inside the comments of type B Moves to take care 
of side effects is wrong: One only has to take care of side effects which 
effect variables that are not themselves type A state variables.

Given that I am planning and preparing a refactoring of the State/Move 
approach (to include the trigger mechanism, but also to remove the usage of 
Move creations in the main code to change ArrayListStates etc.) I suggest to 
remove the type B moves altogether and replace them by changing the plain 
variables effected to type A  states.

To be sure: Before I will do this I would like to hear you others pros and 
cons. And I would like to extend the automated game testing to allow testing 
of undos/redos as well (by allowing Rails to save and store undos/redos if a 
specific flag is set).

However one of the main advantages will be that it allows address at least 
some of those problems which arose for heavy gamers, where no one was able to 
recreate the exact sequence of undos/redos.

But for now it will be better to have a release first with all those finished 
and nearly finished games out in the public, after we got git running ;-)
Stefan


On Monday, May 16, 2011 11:12:46 pm Erik Vos wrote:
> Phil,
> 
> If I replay from the point just before buying the Tunnel, after a Save and
> Load at that point, it is OK. Buying a second Bonus token is not allowed.
> I don't see how you ever could have managed to do that.  Has any Undoing
> happened around that time?  Clearly Undo still has its weaknesses.
> 
> Erik.
> 
> > -----Oorspronkelijk bericht-----
> > Van: Phil Davies [mailto:de...@gm...]
> > Verzonden: maandag 16 mei 2011 16:50
> > Aan: Development list for Rails: an 18xx game
> > Onderwerp: Re: [Rails-devel] 1856 bridge private token
> > 
> > Actually, don't bother investigating this, I've backtracked and worked
> > out what's wrong.  Somehow I managed to get LPS to buy a bonus bridge
> > token from itself!  Rails should prevent this but I hardly count this as
> 
> critical rules
> 
> > enforcement and is clearly a player instigated error,
> > 
> > Phil
> > 
> > On 16 May 2011 15:40, Phil Davies <de...@gm...> wrote:
> > > Something odd is happening with the bridge private
> > > 
> > > Load up the attached game, for a start, LPS has the bridge token
> > > showing twice in it's tokens list.  Secondly, if you try and rewind
> > > the game back to an earlier state (try OR 1, when LPS first runs) and
> > > without even playing forward, click again on a more up to date entry
> > > in the game report log, and you will see another token get added to
> > > the list that LPS has.
> > > 
> > > The latter problem isn't really that bad as it doesn't persist across
> > > saves, but it does imply an issue with the undo process.  The former
> > > issue (LPS having two tokens on reload) is more of a problem -
> > > although fixable if the players are aware of it
> > > 
> > > Phil
> 
> ---------------------------------------------------------------------------
> - --
> 
> > Achieve unprecedented app performance and reliability What every C/C++
> > and Fortran developer should know.
> > Learn how Intel has extended the reach of its next-generation tools to
> 
> help
> 
> > boost performance applications - inlcuding clusters.
> > http://p.sf.net/sfu/intel-dev2devmay
> > _______________________________________________
> > Rails-devel mailing list
> > Rai...@li...
> > https://lists.sourceforge.net/lists/listinfo/rails-devel
> 
> ---------------------------------------------------------------------------
> --- Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel

Re: [Rails-devel] Git troubleshooting [was: Git repository isnowavailable.]

[<Dr....@t-...>]

Hi,

Helios release windows based with jgit and egit from the helios download site.

-----Original-Nachricht-----
> Subject: Re: [Rails-devel] Git troubleshooting [was: Git repository
> isnowavailable.]
> Date: Thu, 21 Jul 2011 21:56:13 +0200
> From: Stefan Frey <ste...@we...>
> To: "Development list for Rails: an 18xx game"
> <rai...@li...>

> Brett raised the issue of different problems depending on
> platform/eclipse version and he started reporting the platform he
> uses.
> 
> I run Eclipse Helios (Build id: 20100617-1415)
> on an OpenSuse 11.3 Linux distribution.
> 
> Egit is from the Helios update link.
> Stefan
> 
> 
> On Thursday, July 21, 2011 06:25:40 pm brett lentz wrote:
> 
> > I also noticed that there's a new version of Eclipse out. This
> > version is named "Indigo". I wonder if there are differences that
> > Egit isn't handling very well.
> > 
> > ---Brett.
> > 
> > 
> > 
> > On Thu, Jul 21, 2011 at 3:24 AM, Dr....@t-...
> > 
> > <Dr....@t-...> wrote:
> > 
> > > Hi all,
> > > I also installed the jgit and egit modules and repeatedly ran into
> > > problems trying to fetch/get the commits and have them displayed
> > > properly in the eclipse history. Even though the fetch or pull
> > > finished and stated there was no new ref on the remote origin the
> > > history showed that one or more commits didnt reach my working
> > > copy. I.e the history was showng a different hash asa the actual
> > > git repository on rails.sf.net.
> > > 
> > > i currently (state of yesterday) rebuild the local repository with
> > > a new clone and was able to see the latest changes. I suspect that
> > > either something is still buggy inside the eclipse modules or
> > > sourceforges git implementation is not what it should be.
> > > 
> > > Or that the setup itself is needing some more foolproofing :) 
> > > Regards Martin
> > > 
> > > 
> > > -----Original-Nachricht-----
> > > 
> > > 
> > >> Subject: Re: [Rails-devel] Git troubleshooting [was: Git
> > repository >> is    now     available.]
> > >> Date: Thu, 21 Jul 2011 11:46:42 +0200
> > >> From: "Erik Vos" <eri...@xs...>
> > >> To: "'Development list for Rails: an 18xx game'"
> > >> <rai...@li...>
> > >>
> > >> OK, I have installed Git for Windows and am running Git Bash.
> > >>
> > >> 'git status' says:
> > >>
> > >> ----------------------------------------
> > >> # On branch master
> > >> # Your branch and 'origin/master' have diverged,
> > >> # and have 1 and 3 different commit(s) each, respectively.
> > >> #
> > >> # Changes not staged for commit:
> > >> #   (use "git add <file>..." to update what will be committed) >>
> > #   (use "git checkout -- <file>..." to discard changes in working
> > >> directory)
> > >> #
> > >> #     modified:   data/1826/Game.xml
> > >> #     modified:   data/1870/Game.xml
> > >> #     modified:   rails/algorithms/NetworkTrain.java
> > >> #     modified:   rails/algorithms/RevenueAdapter.java
> > >> #     modified:   rails/algorithms/RevenueCalculator.java
> > >> #     modified:   rails/algorithms/RevenueCalculatorMulti.java >>
> > #     modified:   rails/algorithms/RevenueCalculatorSimple.java >> #
> >     modified:   rails/game/Train.java
> > >> #     modified:   rails/game/TrainI.java
> > >> #
> > >> # Untracked files:
> > >> #   (use "git add <file>..." to include in what will be
> > committed) # >> #    
> > rails/algorithms/RevenueCalculatorMultiHex.java
> > >> no changes added to commit (use "git add" and/or "git commit -a")
> > >> -------------------------------------------------
> > >>
> > >> 1. The "modified" files are Stefan's updates.  Why are these
> > marked as >> "Changes not staged for commit"?  Because these are in
> > a branch? What >> am I supposed to do?
> > >>
> > >> 2. What is the real status of the "Untracked" file
> > >> RevenueCalculatorMultiHex.java?  This is the file for which Egit
> > >> showed a merge conflict.
> > >> Indeed the file icon is marked "?" so it's currently untracked.
> > What >> should I do?
> > >>
> > >> 'git push' says:
> > >>
> > >> -----------------------------------------------
> > >> ! [rejected]        master -> master (non-fast-forward)
> > >> error: failed to push some refs to
> > >>
> > 'ssh://evos:<password>@rails.git.sourceforge.net/gitroot/rails/rails'
> > >> To prevent you from losing history, non-fast-forward updates were
> > >> rejected Merge the remote changes (e.g. 'git pull') before
> > pushing >> again.  See the 'Note about fast-forwards' section of
> > 'git push >> --help' for details.
> > >> Aborting
> > >> ----------------------------------------------
> > >>
> > >> So I tried 'git pull', which says:
> > >>
> > >> -----------------------------------------------
> > >> error: Your local changes to the following files would be
> > overwritten >> by merge:
> > >> data/1826/Game.xml
> > >> data/1870/Game.xml
> > >> rails/algorithms/NetworkTrain.java
> > >> rails/algorithms/RevenueAdapter.java
> > >> rails/algorithms/RevenueCalculator.java
> > >> rails/algorithms/RevenueCalculatorMulti.java
> > >> rails/algorithms/RevenueCalculatorSimple.java
> > >> rails/game/Train.java
> > >> rails/game/TrainI.java
> > >> Please, commit your changes or stash them before you can merge.
> > >> error: The following untracked working tree files would be
> > overwritten >> by merge:
> > >> rails/algorithms/RevenueCalculatorMultiHex.java
> > >> Please move or remove them before you can merge.
> > >> --------------------------------------------------
> > >>
> > >> So it seems that Stefan's changes are already in my repository,
> > but >> are nevertheless in the way of pulling the very same changes
> > into my >> repository (again??).
> > >>
> > >> Before I start trying lots of other things: is there an easier
> > way out >> than to do it all over again, as Brett suggested?
> > >>
> > >> Erik.
> > >>
> > >> > -----Original Message-----
> > >> > From: Erik Vos [mailto:eri...@xs...]
> > >> > Sent: Wednesday, July 20, 2011 11:47 PM
> > >> > To: 'Development list for Rails: an 18xx game'
> > >> > Subject: Re: [Rails-devel] Git troubleshooting [was: Git
> > repository >> > is now available.]
> > >> >
> > >> > My HEAD is at dab610 (parent fd4a38) and includes my commit,
> > but not >> > Stefan's.
> > >> > I'll try again tomorrow.
> > >> >
> > >> > Erik.
> > >> >
> > >> > > -----Original Message-----
> > >> > > From: brett lentz [mailto:bre...@gm...]
> > >> > > Sent: Wednesday, July 20, 2011 11:30 PM
> > >> > > To: Development list for Rails: an 18xx game
> > >> > > Subject: Re: [Rails-devel] Git troubleshooting [was: Git >> >
> > > repository is now available.]
> > >> > >
> > >> > >
> > >> > > One thing you can try is clone a new repository. Then, export
> > your >> > > commits as a diff that can be applied to the newly
> > cloned tree.
> > >> > > Once done, you can try pushing from a fresh tree.
> > >> > >
> > >> > >
> > >> > > While Git tries to allow your to be ambigious and tries to do
> > the >> > > right thing, you can also be explicit with your pushes:
> > >> > >
> > >> > >
> > >> > > $ git push origin master:refs/heads/master
> > >> > >
> > >> > > This says "Push local master branch to refs/heads/master on
> > >> > > origin."
> > >> > >
> > >> > > According to the GitWeb, HEAD in the repo should be at >> > >
> > d123b894c.
> > >> > >
> > >> > >
> > >> > > ---Brett.
> > >> > >
> > >> > >
> > >> > > On Wed, Jul 20, 2011 at 2:27 PM, Erik Vos
> > <eri...@xs...> >> > >
> > >> > > wrote:
> > >> > > > Something has worked: Team|Merge on FETCH_HEAD has updated
> > my >> > >
> > >> > > working
> > >> > >
> > >> > > > copies with Stefan's changes.
> > >> > > > All these changes have disappeared from the Synchronize
> > view, >> > > > except RevenueCalculatorMultiHex.java, which now
> > shows a merge >> > > > conflict, even though the local and remote
> > sources are >> > > > identical.  "Mark as
> > >> >
> > >> > merged"
> > >> >
> > >> > > > does nothing.
> > >> > > >
> > >> > > > And I still can't push, with the same error message.
> > >> > > >
> > >> > > > Erik.
> > >> > > >
> > >> > > >> -----Original Message-----
> > >> > > >> From: Erik Vos [mailto:eri...@xs...]
> > >> > > >> Sent: Wednesday, July 20, 2011 11:11 PM
> > >> > > >> To: 'Development list for Rails: an 18xx game'
> > >> > > >> Subject: Re: [Rails-devel] Git troubleshooting [was: Git
> > >> > >
> > >> > > repository >> is now available.]
> > >> > >
> > >> > > >> Hmm, now *I* seem to be royally stuck.
> > >> > > >>
> > >> > > >> I have prepared and committed some changes (in master) and
> > >> > >
> > >> > > tried to >> push this commit, but it keeps being rejected
> > with a >> > > message >> "non-fast
> > >> > >
> > >> > > > forward".
> > >> > > >
> > >> > > >> Google tells me that this means that there is stuff
> > waiting to >> > >
> > >> > > be >> pulled
> > >> > >
> > >> > > > first,
> > >> > > >
> > >> > > >> so I tried that (not sure if I started with fetch or with
> > pull, >> > > >> tried
> > >> > > >
> > >> > > > both), but I
> > >> > > >
> > >> > > >> keep getting "No ref to fetch from origin - everything up
> > to >> > >
> > >> > > date".
> > >> > >
> > >> > > >> Merge attempts fail as well.  My working copy of
> > 1826/Game.xml >> > > >> (just one file changed by Stefan) is not
> > updated.
> > >> > > >>
> > >> > > >> I suppose that Stefan's branching is somehow related to
> > this >> > >
> > >> > > problem.
> > >> > >
> > >> > > >> Why has this H_train branch been pushed at all?  For now
> > I'm >> > >
> > >> > > only >> interested
> > >> > >
> > >> > > > in
> > >> > > >
> > >> > > >> what's in the master branch.
> > >> > > >> I would think we can better refrain from branching until
> > the >> > >
> > >> > > normal
> > >> > >
> > >> > > > (master)
> > >> > > >
> > >> > > >> pushes and pulls have proved to work at all sides.
> > >> > > >>
> > >> > > >> I'll keep trying, and perhaps I'll have to try to use git
> > on >> > >
> > >> > > the >> command
> > >> > >
> > >> > > > line to
> > >> > > >
> > >> > > >> get this fixed.
> > >> > > >>
> > >> > > >> Anyhow, it's clear that Git is boggling my mind... 
> > Perhaps a >> > >
> > >> > > fresh >> look tomorrow will help.  But PLEASE - keep it
> > simple >> > > initially! >>
> > >> > >
> > >> > > >> Erik.
> > >> > > >>
> > >> > > >> > -----Original Message-----
> > >> > > >> > From: Stefan Frey [mailto:ste...@we...]
> > >> > > >> > Sent: Wednesday, July 20, 2011 5:03 PM
> > >> > > >> > To: Development list for Rails: an 18xx game
> > >> > > >> > Subject: Re: [Rails-devel] Git troubleshooting [was: Git
> > >> > >> > >
> > >> > > repository is now available.]
> > >> > >
> > >> > > >> > Brett:
> > >> > > >> > sorry I was using the wrong wording below (still kept
> > somehow >> > >
> > >> > > in >> > my subversion-world, where a branch is a copy of the
> > >> > > trunk).
> > >> > >
> > >> > > >> > I did not clone the local repo, instead I created a
> > branch >> > >
> > >> > > inside >> > the same repo. So my actual workflow was the one
> > you >> > > suggested below.
> > >> > >
> > >> > > >> > So I have two branches shown in the git explorer of
> > egit: >> > > master >> > and
> > >> > >
> > >> > > >> h_train.
> > >> > > >>
> > >> > > >> > Anyhow I tested things again in the last half hour and
> > it did >> > >
> > >> > > not >> > work
> > >> > >
> > >> > > >> first, but
> > >> > > >>
> > >> > > >> > now it worked out of the blue.
> > >> > > >> >
> > >> > > >> > Have you changed anything on the settings or did I
> > torture >> > >
> > >> > > egit >> > for so
> > >> > >
> > >> > > >> long,
> > >> > > >>
> > >> > > >> > that it had pity with me.
> > >> > > >> >
> > >> > > >> > Stefan
> > >> > > >> >
> > >> > > >> >
> > >> > > >> >
> > >> > > >> >
> > >> > > >> > On Wednesday, July 20, 2011 04:45:02 pm brett lentz
> > wrote: >> >> > > > >
> > >> > > > > Comments inline...
> > >> > > > >
> > >> > > >> > > On Wed, Jul 20, 2011 at 4:39 AM, Stefan Frey
> > >> > > >> > > <ste...@we...>
> > >> > > >> >
> > >> > > >> > wrote:
> > >> > > >> > > > Actually I had to use the push uri:
> > >> > > ssh://{username}@rails.git.sourceforge.net/gitroot/rails/rail
> > >> > >> > >
> > >> > > > > s
> > >> > > > >
> > >> > > >> > > > Using the git protocol was not possible, as egit >>
> > > >
> > >> > > complained >> > > > that it does not support a username.
> > >> > >
> > >> > > >> > > Correct. To be able to push changes, your remote uri
> > needs >> > >
> > >> > > to >> > > be the ssh uri.
> > >> > >
> > >> > > >> > > You can modify this setting on the fly, and don't need
> > to >> > > >> > > re-clone the repository. The "git config" CLI
> > commands can >> > > >> > > update any repository setting.
> > >> > > >> > >
> > >> > > >> > > > I prefer not to use the synchronize perspective with
> > git, >> > >
> > >> > > as >> > > > I wonder what it actually shows, as there are now
> > two >> > > states >> > > > of changes (changes not committed,
> > changes >> > > committed but not >> > > > pushed
> > >> > >
> > >> > > >> > yet).
> > >> > > >> >
> > >> > > >> > > I'm not yet familiar with EGit, so most of the
> > following >> >> > > > >
> > >> > > > > information will be based on my knowledge of the CLI
> > tools.
> > >> > > > >
> > >> > > >> > > > So my intended workflow to implement a new feature
> > was: >> > > >> > > > A) Create a branch for that feature locally by
> > cloning my >> > > >> > > > local master which is itself a clone of
> > the remote >> > >
> > >> > > master.
> > >> > >
> > >> > > >> > > > B) Commit the changes in the local branch.
> > >> > > >> > > > C) Push/merge those changes into my local master.
> > >> > > >> > > > D) Push the changes from the local master into the
> > remote >>
> > >> master.
> > >>
> > >> > > >> > > Creating two local repositories is unnecessary and
> > might be >> > > >> > > adding extra complications to your workflow.
> > >> > > >> > >
> > >> > > >> > > Let's see if we can get the simple use case working,
> > then >> > >
> > >> > > you >> > > can decide if you need more.
> > >> > >
> > >> > > >> > > Try doing this:
> > >> > > >> > >
> > >> > > >> > > 1. Create a topic branch.
> > >> > > >> > > 2. Work as needed in the topic branch, making local
> > commits >> > > >> > > when it makes sense. 3. When you've got an
> > amount of work >> > > >> > > that's ready to push, update your local
> > master and then >> > >
> > >> > > merge >> > > your topic branch to your
> > >> > >
> > >> > > >> > local master.
> > >> > > >> >
> > >> > > >> > > 4. Push your changes to the remote repository.
> > >> > > >> > >
> > >> > > >> > > Let's even save rebasing for later.
> > >> > > >> > >
> > >> > > >> > > > Unfortunately I was not able to push the changes to
> > the >> > > >> > > > master branch, I got a rejected error message
> > for this >> > >
> > >> > > (both >> > > > on dry run and real). But my local master is
> > >> > > up-to-date with >> > > > the repo
> > >> > > master.
> > >> > >
> > >> > > >> > > If you post the error message, it would help.
> > >> > > >> > >
> > >> > > >> > > The most common reasons for a failed push are:
> > >> > > >> > >
> > >> > > >> > > 1. Authentication and permissions problems (including
> > using >> > >
> > >> > > the >> > > wrong remote.origin.url.
> > >> > >
> > >> > > >> > > 2. Local branch being out of sync with remote branch.
> > (try >> > >
> > >> > > a >> > > git pull, then retry your push.)
> > >> > >
> > >> > > >> > > Unline SVN, Git won't let you push changes if your
> > local >> > >
> > >> > > tree >> > > is out of sync with the remote tree. You need to
> > pull >> > > any >> > > remote changes into your local tree, resolve
> > any merge >> > >
> > >> > > >> > > conflicts (if
> > >> > > >> > > any) locally, then push your changes.
> > >> > > >> > >
> > >> > > >> > > > However I was able to push my branch H_train into
> > the >> > >
> > >> > > repo >> > > > and was able to push my committed changes
> > there.
> > >> > >
> > >> > > >> > > > My ref spec for the push to the master is:
> > >> > > >> > > > refs/heads/master:refs/heads/master
> > >> > > >> > >
> > >> > > >> > > This looks correct.
> > >> > > >> > >
> > >> > > >> > > It would be more useful to diagnose the issue to do
> > this: >> > > >> > >
> > >> > > >> > > $ git fetch
> > >> > > >> > > $ git status
> > >> > > >> > > $ git branch -vv
> > >> > > >> > > $ git diff origin/master master --stat
> > >> > > >> > >
> > >> > > >> > > The fetch will update your remote refs without
> > changing >> > >
> > >> > > your >> > > working copy. Git pull simply is a shortcut to
> > running >> > > fetch >> > > then either merge or rebase (depending
> > on chosen >> > > options).
> > >> > >
> > >> > > >> > > The output of the status, branch, and diff commands
> > will >> > >
> > >> > > tell >> > > you how out of sync you are with the remote side,
> > >> > > often telling >> > > you how many commits ahead or behind you
> > are.
> > >> > >
> > >> > > >> > > > What have I done wrong?
> > >> > > >> > > > Maybe it is better to wait for Brett to help here, I
> > do >> > >
> > >> > > not >> > > > want to break anything and he has daily working
> > >> > > knowledge with
> > >> >
> > >> > git.
> > >> >
> > >> > > >> > > > Stefan
> > >> > > >> > >
> > >> > > >> > > ---Brett.
> > >> > > >> > >
> > >> > > >> > > > On Wednesday, July 20, 2011 11:12:44 am Erik Vos
> > wrote: >> > > >> > > >> See below for my findings
> > >> > > >> > > >>
> > >> > > >> > > >> > -----Original Message-----
> > >> > > >> > > >> > From: Stefan Frey [mailto:ste...@we...] >>
> > > > >> > > >> >
> > >> > > >> > > >> > Brett or Erik:
> > >> > > >> > > >> > I believe I am doing something wrong or something
> > is >> > >
> > >> > > not >> > > >> > setup correctly,
> > >> > >
> > >> > > >> > > >> but
> > >> > > >> > > >>
> > >> > > >> > > >> > I do not get the push working after the commit.
> > >> > > >> > > >> >
> > >> > > >> > > >> > I did setup a complete new workspace for git,
> > cloned >> > > >> > > >> > remote, changed , committed, then pushed
> > to remote.
> > >> > > >> > > >> > All steps worked, except the final push.
> > >> > > >> > > >>
> > >> > > >> > > >> I did the same thing, and it all appears to have
> > worked.
> > >> > > >> > > >> I'm not happy with the way Egit reports a
> > successful >> > >
> > >> > > push: >> > > >> the synchronization view is not updated
> > (unlike >> > > Subversive, >> > > >> which cleaned up that view
> > after a commit).
> > >> > > Basically I have >> > > >> to conclude success from the
> > absence of >> > > error messages; I >> > > >> would have preferred
> > >> > >
> > >> > > >> > some positive message.
> > >> > > >> >
> > >> > > >> > > >> > The error message I get is that the push is not
> > >> > >
> > >> > > permitted.
> > >> > >
> > >> > > >> > > >> > My thoughts are:
> > >> > > >> > > >> >
> > >> > > >> > > >> > I have not setup a specific push-URI, only used
> > the >> >> > > > > >> >
> > >> > > > > >> > sourceforge link
> > >> > > > > >> >
> > >> > > >> > > >> >
> > git://rails.git.sourceforge.net/gitroot/rails/rails >> >> > > >> > >
> > >>
> > >> > > >> > > >> Same as I did, except that I had inserted my
> > username, >> > >
> > >> > > as >> > > >> per the Egit help info
> > >> > >
> > >> > > >>
> > (git://ev...@ra.../gitroot/rails/rails).
> > >> > > >>
> > >> > > >> > > >>  I wonder: could and should I also insert my
> > password? >> > >
> > >> > > Now >> > > >> I'm asked to type my Sourceforge password
> > *every* >> > > time I >> > > >> access remote (Subversive only
> > required it once >> > > per Eclipse
> > >> >
> > >> > session).
> > >> >
> > >> > > >> > > >> I could not find a place to configure that either.
> > >> > > >> > > >>
> > >> > > >> > > >> > During the clone the fields for authentication
> > were >> > >
> > >> > > grayed
> > >>
> > >> out.
> > >>
> > >> > > >> > > >> Can't remember.
> > >> > > >> > > >>
> > >> > > >> > > >> > During the push I never got asked for my
> > user-account >> > >
> > >> > > or >> > > >> > password on sourceforge.
> > >> > >
> > >> > > >> > > >> I think I was asked for my password only, as I had
> > >> > >
> > >> > > included >> > > >> mu username in the URL.
> > >> > >
> > >> > > >> > > >> > I was not able to locate anywhere in all those
> > Eclipse >> > > >> > > >> > dialogs where I
> > >> > > >> > > >>
> > >> > > >> > > >> might be
> > >> > > >> > > >>
> > >> > > >> > > >> > able to setup my account.
> > >> > > >> > > >>
> > >> > > >> > > >> The only place I found was in the Push Wizard >> >
> > > >> > > >> (Team|Remote|Push), but I haven't used the Custom part
> > >> > >
> > >> > > of >> > > >> that
> > >> > > either.
> > >> > >
> > >> > > >> > > >> > Could you please provide help?
> > >> > > >> > > >>
> > >> > > >> > > >> Unfortunately I can't speak the final word on this
> > >> > >
> > >> > > matter.
> > >> > >
> > >> > > >> > > >> Let's together try to find what works best.
> > >> > > >> > > >>
> > >> > > >> > > >> Erik.
> > >> > >
> > >> > > ------------------------------------------------------------
> > >> > >> > >
> > >> > > > >> --
> > >> > > > >>
> > >> > > >> > > >> ---
> > >> > > >> > > >> --
> > >> > > >> > > >> -----
> > >> > > >> > > >> --- --- 10 Tips for Better Web Security Learn 10
> > ways to >> > > >> > > >> better secure your business today. Topics
> > covered >> > > >> > > >>
> > >> > > >> include: Web security, SSL, hacker attacks & Denial of >>
> > > > >> > > >> Service (DoS), private keys, security Microsoft
> > Exchange, >> > >> > > >>
> > >> > > > >> secure Instant Messaging, and much more.
> > >> > > > >>
> > >> > > >> > > >> http://www.accelacomm.com/jaw/sfnl/114/51426210/ >>
> > > > >> > > >> _______________________________________________ >> > >
> > >> > > >> Rails-devel mailing list
> > >> > > >> > > >> Rai...@li...
> > >> > > >> > > >>
> > https://lists.sourceforge.net/lists/listinfo/rails-devel >> > >
> > >> > > -------------------------------------------------------------
> > >> > >> > >
> > >> > > > > --
> > >> > > > >
> > >> > > >> > > > ---
> > >> > > >> > > > --
> > >> > > >> > > > -----
> > >> > > >> > > > ----- 10 Tips for Better Web Security Learn 10 ways
> > to >> > >
> > >> > > better >> > > > secure your business today. Topics covered >>
> > > >
> > >> > > >> > > > include: Web security, SSL, hacker attacks & Denial
> > of >> >> > > > > >
> > >> > > > > > Service (DoS), private keys, security Microsoft
> > Exchange, >> >> > > > > > secure Instant Messaging, and much more.
> > >> > > > > >
> > >> > > >> > > > http://www.accelacomm.com/jaw/sfnl/114/51426210/ >>
> > > > >> > > > _______________________________________________ >> > >
> > >> > > > Rails-devel mailing list
> > >> > > >> > > > Rai...@li...
> > >> > > >> > > >
> > https://lists.sourceforge.net/lists/listinfo/rails-devel >> > >
> > >> > >
> > --------------------------------------------------------------- >>
> > >> > >
> > >> > > > > --
> > >> > > > >
> > >> > > >> > > ---
> > >> > > >> > > --
> > >> > > >> > > -----
> > >> > > >> > > --- 10 Tips for Better Web Security Learn 10 ways to
> > better >> > > >> > > secure your business today. Topics covered
> > >> > > >> >
> > >> > > >> > include:
> > >> > > >> > > Web security, SSL, hacker attacks & Denial of Service
> > >> > >
> > >> > > (DoS), >> > > private keys, security Microsoft Exchange,
> > secure >> > > Instant >> > > Messaging, and much
> > >> > >
> > >> > > >> > more.
> > >> > > >> >
> > >> > > >> > > http://www.accelacomm.com/jaw/sfnl/114/51426210/ >> >
> > > >> > > _______________________________________________ >> > > >> >
> > > Rails-devel mailing list
> > >> > > >> > > Rai...@li...
> > >> > > >> > >
> > https://lists.sourceforge.net/lists/listinfo/rails-devel >> >> > > >
> > >> > > >
> > ------------------------------------------------------------------
> > >> > > > -- --
> > >> > > > ------
> > >> > > >
> > >> > > >> --
> > >> > > >>
> > >> > > >> > 10 Tips for Better Web Security
> > >> > > >> > Learn 10 ways to better secure your business today.
> > Topics >> >> > > >
> > >> > > > covered
> > >> > > >
> > >> > > >> include:
> > >> > > >> > Web security, SSL, hacker attacks & Denial of Service
> > (DoS), >> > > >> > private keys, security Microsoft Exchange, secure
> > Instant >> >> > > >
> > >> > > > Messaging, and much
> > >> > > >
> > >> > > >> more.
> > >> > > >>
> > >> > > >> > http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > >> > > >> > _______________________________________________
> > >> > > >> > Rails-devel mailing list
> > >> > > >> > Rai...@li...
> > >> > > >> > https://lists.sourceforge.net/lists/listinfo/rails-devel
> > >> >> > > >
> > >> > > >
> > ------------------------------------------------------------------
> > >> > > > -- --
> > >> > > > ------
> > >> > > > --
> > >> > > >
> > >> > > >> 10 Tips for Better Web Security
> > >> > > >> Learn 10 ways to better secure your business today. Topics
> > >> > >
> > >> > > covered
> > >> > >
> > >> > > > include:
> > >> > > >> Web security, SSL, hacker attacks & Denial of Service
> > (DoS), >> >> > >
> > >> > > private keys, security Microsoft Exchange, secure Instant >>
> > >> > > Messaging, and much
> > >> > > more.
> > >> > >
> > >> > > >> http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > >> > > >> _______________________________________________
> > >> > > >> Rails-devel mailing list
> > >> > > >> Rai...@li...
> > >> > > >> https://lists.sourceforge.net/lists/listinfo/rails-devel >
> > >> > > >
> > >> > > >
> > ------------------------------------------------------------------
> > >> > > > -- --
> > >> > > > --------
> > >> > > > 10 Tips for Better Web Security
> > >> > > > Learn 10 ways to better secure your business today. Topics
> > >> > > > covered
> > >> > >
> > >> > > include:
> > >> > > > Web security, SSL, hacker attacks & Denial of Service
> > (DoS), >> > > > private keys, security Microsoft Exchange, secure
> > Instant >> > > > Messaging, and much
> > >> > >
> > >> > > more.
> > >> > >
> > >> > > > http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > >> > > > _______________________________________________
> > >> > > > Rails-devel mailing list
> > >> > > > Rai...@li...
> > >> > > > https://lists.sourceforge.net/lists/listinfo/rails-devel >>
> > > >
> > >> > >
> > --------------------------------------------------------------------
> > >> > > -- --------
> > >> > > 10 Tips for Better Web Security
> > >> > > Learn 10 ways to better secure your business today. Topics
> > covered >> >
> > >> > include:
> > >> > > Web security, SSL, hacker attacks & Denial of Service (DoS),
> > >> > > private keys, security Microsoft Exchange, secure Instant >>
> > > > Messaging, and much
> > >> >
> > >> > more.
> > >> >
> > >> > > http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > >> > > _______________________________________________
> > >> > > Rails-devel mailing list
> > >> > > Rai...@li...
> > >> > > https://lists.sourceforge.net/lists/listinfo/rails-devel >>
> > >>
> > ------------------------------------------------------------------------
> > >> ---- --
> > >>
> > >> > 10 Tips for Better Web Security
> > >> > Learn 10 ways to better secure your business today. Topics
> > covered >>
> > >> include:
> > >> > Web security, SSL, hacker attacks & Denial of Service (DoS),
> > private >> > keys, security Microsoft Exchange, secure Instant
> > Messaging, and >> > much more.
> > >> > http://www.accelacomm.com/jaw/sfnl/114/51426210/
> > >> > _______________________________________________
> > >> > Rails-devel mailing list
> > >> > Rai...@li...
> > >> > https://lists.sourceforge.net/lists/listinfo/rails-devel >>
> > >>
> > ------------------------------------------------------------------------
> > >> ------ 5 Ways to Improve & Secure Unified Communications
> > >> Unified Communications promises greater efficiencies for
> > business. UC >> can improve internal communications as well as offer
> > faster, more >> efficient ways to interact with customers and
> > streamline customer >> service. Learn more!
> > http://www.accelacomm.com/jaw/sfnl/114/51426253/ >>
> > _______________________________________________
> > >> Rails-devel mailing list
> > >> Rai...@li...
> > >> https://lists.sourceforge.net/lists/listinfo/rails-devel
> > 
> > > 
> > > -------------------------------------------------------------------------
> > > ----- 5 Ways to Improve & Secure Unified Communications
> > > Unified Communications promises greater efficiencies for business.
> > > UC can improve internal communications as well as offer faster,
> > > more efficient ways to interact with customers and streamline
> > > customer service. Learn more!
> > > http://www.accelacomm.com/jaw/sfnl/114/51426253/
> > > _______________________________________________
> > > Rails-devel mailing list
> > > Rai...@li...
> > > https://lists.sourceforge.net/lists/listinfo/rails-devel
> > > 
> > 
> > ---------------------------------------------------------------------------
> > --- 5 Ways to Improve & Secure Unified Communications
> > Unified Communications promises greater efficiencies for business.
> > UC can improve internal communications as well as offer faster, more
> > efficient ways to interact with customers and streamline customer
> > service. Learn more!
> > http://www.accelacomm.com/jaw/sfnl/114/51426253/
> > _______________________________________________
> > Rails-devel mailing list
> > Rai...@li...
> > https://lists.sourceforge.net/lists/listinfo/rails-devel
> > 
> 
> ------------------------------------------------------------------------------
> 5 Ways to Improve & Secure Unified Communications
> Unified Communications promises greater efficiencies for business. UC
> can improve internal communications as well as offer faster, more
> efficient ways to interact with customers and streamline customer
> service. Learn more! http://www.accelacomm.com/jaw/sfnl/114/51426253/
> _______________________________________________
> Rails-devel mailing list
> Rai...@li...
> https://lists.sourceforge.net/lists/listinfo/rails-devel
> 
> 

Re: [Rails-devel] Git troubleshooting

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

On Thu, Jul 21, 2011 at 3:38 PM, Erik Vos <eri...@xs...> wrote:
> It seems to me that Egit is not yet mature in all respects.
> For instance, the errors I have got are much better explained on the command
> line than they are in Egit.
> But we'll have to live with it.
>
> Do you have an example of how to configure --rebase in .gitconfig?
>
> Erik.
>


(Trimming the other thread bits for brevity's sake).

There was another post in this thread that included the syntax. You
want the [branch] portion below...

My full .gitconfig looks like this:

[blentz@blentzmac ~] (master %)$ cat ~/.gitconfig
[user]
        name = Brett Lentz
        email = bre...@gm...

[color]
        diff = auto
        status = auto
        branch = auto
        interactive = auto
        ui = true
        pager = true

[color "branch"]
        current = yellow reverse
        local = yellow
        remote = green

[color "diff"]
        meta = yellow bold
        frag = magenta bold
        old = red bold
        new = green bold

[color "status"]
        added = yellow
        changed = green
        untracked = cyan

[core]
        pager = less -FRSX
        whitespace = fix,-indent-with-non-tab,trailing-space,cr-at-eol

[alias]
	ci = commit
	ch = cherry-pick -x
	co = checkout
	br = branch -v
	st = status
	tag = tag -a

[branch]
	autosetuprebase = always

[merge]
	tool = vimdiff

[receive]
	denyDeleteCurrent = warn