Hey, guys. I''ve made a chess game and right now, users have to click on the piece, then click on the square they want it to go to. I''d like to allow the user to just drag and drop the piece. What I need, however, is when the user drags the piece to a square, my normal ajax function fires off to determine whether or not the move is legal. If it is, let it stay there (centered on the square). If it''s not legal, make it snap back where it was before. Obviously if, when I create the draggable (the piece), I set revert to true, the piece will snap back and false otherwise. What I need it to do, however, is kinda wait at the square while my code determines whether or not the move is legal. I can''t figure out an intelligent way to do this. If I set revert to true, it will snap back to it''s original position when I want it to stay at the square while checking to see if the move is legit. If I set revert to false and the move is illegal, it''s not going to know where to snap back to. Anybody have any ideas how to go about this? Thanks in advance. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
On 7/5/07, Red <redlocyellow-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> I can''t figure out an intelligent way to do this. If I set revert to true, > it will snap back to it''s original position when I want it to stay at the > square while checking to see if the move is legit. If I set revert to false > and the move is illegal, it''s not going to know where to snap back to. > > Anybody have any ideas how to go about this? Thanks in advance. >How about pre-fetching your legal moves for the piece using the XHR call? Then you can validate the move against a local set of data. For example: In a grid where the rows are numbered and the columns are lettered, we''ll say a pawn is in square 4A The pre-fetch for that piece says that it can only legally move to square 5A (if there were an opponent''s piece in 5B, that square would also be returned) The user moves the piece forward 1 square (legally) to square 5A (If they tried 6A, the event would return false and snap the piece back.) If the ondrop event says it''s "OK" for that piece, you fire off an XHR to get the next list of legal moves for that piece I''m sure it will get more complicated with regards to taking another person''s piece, etc - but using Ajax before the person actually has to make the move would be more sensible rather than making the person wait for the response for each move. You could then maintain an object with 32 (16 black, 16 white) elements with arrays of "legal" moves. My two cents :) Jon --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
Whether I do that or do it the way I''m doing it, it would still require that I set revert to true or false depending on whether or not the move is legal. So I''d have to set revert programatically. I know it can be done, but I don''t know how to do it. It shouldn''t matter whether I use ajax to verify the move or store the legal moves locally. The ajax verification happens pretty instantaneously. The issue becomes making it either stay at revert=true or revert=false depending on whether or not the move is legal. Because no matter what, I need the piece to stay there for a sec or so while the verification function checks to see if the move is legal regardless of whether or not it''s checking locally or using ajax. So I need it to basically go to revert=false (and center itself in the square, which is a div) if the move is legal and revert if the move is false. I''m not even sure if this is possible, unless I come up with something custom. Thanks for the response. On 7/5/07, Jon Trelfa <jtrelfa-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > > On 7/5/07, Red <redlocyellow-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > I can''t figure out an intelligent way to do this. If I set revert to > true, > > it will snap back to it''s original position when I want it to stay at > the > > square while checking to see if the move is legit. If I set revert to > false > > and the move is illegal, it''s not going to know where to snap back to. > > > > Anybody have any ideas how to go about this? Thanks in advance. > > > > How about pre-fetching your legal moves for the piece using the XHR call? > Then you can validate the move against a local set of data. For example: > > In a grid where the rows are numbered and the columns are lettered, > we''ll say a pawn is in square 4A > The pre-fetch for that piece says that it can only legally move to > square 5A (if there were an opponent''s piece in 5B, that square would > also be returned) > The user moves the piece forward 1 square (legally) to square 5A (If > they tried 6A, the event would return false and snap the piece back.) > If the ondrop event says it''s "OK" for that piece, you fire off an XHR > to get the next list of legal moves for that piece > > I''m sure it will get more complicated with regards to taking another > person''s piece, etc - but using Ajax before the person actually has to > make the move would be more sensible rather than making the person > wait for the response for each move. > > You could then maintain an object with 32 (16 black, 16 white) > elements with arrays of "legal" moves. > > My two cents :) > > Jon > > > >--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
I believe revert=true has to be set before the dragging occurs. Why not just move the chess piece back to the square it came from using a function if the move is illegal? On Jul 5, 3:44 pm, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> Whether I do that or do it the way I''m doing it, it would still require that > I set revert to true or false depending on whether or not the move is legal. > So I''d have to set revert programatically. I know it can be done, but I > don''t know how to do it. > > It shouldn''t matter whether I use ajax to verify the move or store the legal > moves locally. The ajax verification happens pretty instantaneously. The > issue becomes making it either stay at revert=true or revert=false depending > on whether or not the move is legal. Because no matter what, I need the > piece to stay there for a sec or so while the verification function checks > to see if the move is legal regardless of whether or not it''s checking > locally or using ajax. > > So I need it to basically go to revert=false (and center itself in the > square, which is a div) if the move is legal and revert if the move is > false. I''m not even sure if this is possible, unless I come up with > something custom. > > Thanks for the response. > > On 7/5/07, Jon Trelfa <jtre...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > > > On 7/5/07, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > I can''t figure out an intelligent way to do this. If I set revert to > > true, > > > it will snap back to it''s original position when I want it to stay at > > the > > > square while checking to see if the move is legit. If I set revert to > > false > > > and the move is illegal, it''s not going to know where to snap back to. > > > > Anybody have any ideas how to go about this? Thanks in advance. > > > How about pre-fetching your legal moves for the piece using the XHR call? > > Then you can validate the move against a local set of data. For example: > > > In a grid where the rows are numbered and the columns are lettered, > > we''ll say a pawn is in square 4A > > The pre-fetch for that piece says that it can only legally move to > > square 5A (if there were an opponent''s piece in 5B, that square would > > also be returned) > > The user moves the piece forward 1 square (legally) to square 5A (If > > they tried 6A, the event would return false and snap the piece back.) > > If the ondrop event says it''s "OK" for that piece, you fire off an XHR > > to get the next list of legal moves for that piece > > > I''m sure it will get more complicated with regards to taking another > > person''s piece, etc - but using Ajax before the person actually has to > > make the move would be more sensible rather than making the person > > wait for the response for each move. > > > You could then maintain an object with 32 (16 black, 16 white) > > elements with arrays of "legal" moves. > > > My two cents :) > > > Jon--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
You could also change the list of droppables, so that the piece could simply not be dropped on an illegal square. You would have to fetch (or determine) the list of legal target squares before the move begins. On Jul 5, 3:44 pm, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> Whether I do that or do it the way I''m doing it, it would still require that > I set revert to true or false depending on whether or not the move is legal. > So I''d have to set revert programatically. I know it can be done, but I > don''t know how to do it. > > It shouldn''t matter whether I use ajax to verify the move or store the legal > moves locally. The ajax verification happens pretty instantaneously. The > issue becomes making it either stay at revert=true or revert=false depending > on whether or not the move is legal. Because no matter what, I need the > piece to stay there for a sec or so while the verification function checks > to see if the move is legal regardless of whether or not it''s checking > locally or using ajax. > > So I need it to basically go to revert=false (and center itself in the > square, which is a div) if the move is legal and revert if the move is > false. I''m not even sure if this is possible, unless I come up with > something custom. > > Thanks for the response. > > On 7/5/07, Jon Trelfa <jtre...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > > > On 7/5/07, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > I can''t figure out an intelligent way to do this. If I set revert to > > true, > > > it will snap back to it''s original position when I want it to stay at > > the > > > square while checking to see if the move is legit. If I set revert to > > false > > > and the move is illegal, it''s not going to know where to snap back to. > > > > Anybody have any ideas how to go about this? Thanks in advance. > > > How about pre-fetching your legal moves for the piece using the XHR call? > > Then you can validate the move against a local set of data. For example: > > > In a grid where the rows are numbered and the columns are lettered, > > we''ll say a pawn is in square 4A > > The pre-fetch for that piece says that it can only legally move to > > square 5A (if there were an opponent''s piece in 5B, that square would > > also be returned) > > The user moves the piece forward 1 square (legally) to square 5A (If > > they tried 6A, the event would return false and snap the piece back.) > > If the ondrop event says it''s "OK" for that piece, you fire off an XHR > > to get the next list of legal moves for that piece > > > I''m sure it will get more complicated with regards to taking another > > person''s piece, etc - but using Ajax before the person actually has to > > make the move would be more sensible rather than making the person > > wait for the response for each move. > > > You could then maintain an object with 32 (16 black, 16 white) > > elements with arrays of "legal" moves. > > > My two cents :) > > > Jon--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
On 7/5/07, Diodeus <diodeus-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > You could also change the list of droppables, so that the piece could > simply not be dropped on an illegal square. You would have to fetch > (or determine) the list of legal target squares before the move > begins. >Which is pretty much where I was going with it. The "revert=true" doesn''t have to be a boolean. With newer versions of scriptaculous, revert can be a generic function reference that does a check after the movement stops. http://wiki.script.aculo.us/scriptaculous/show/Draggable You could also have your Ajax call return an array of "legal" moves and create a droppable for each one? For an even more fun approach, you could make your grid a bunch of droppables - then change the ''accept'' array in each box to only allow certain pieces - sort of a grid-based vs a piece-based movement around the board.> > It shouldn''t matter whether I use ajax to verify the move or store the legal > moves locally. The ajax verification happens pretty instantaneously >What if your chess game gets linked from Ajaxian.com, Digg or some other high-traffic site? - are you sure you have the bandwidth to have "instant" response from the call? Jon --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
Yeah, I got about 6000 hits on the first day from my recent mention on Ajaxian. Chess is a lot more fun/newsworthy so Red''s hits would probably be a lot higher. On Jul 5, 4:02 pm, "Jon Trelfa" <jtre...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> What if your chess game gets linked from Ajaxian.com, Digg or some > other high-traffic site? - are you sure you have the bandwidth to have > "instant" response from the call? > > Jon--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
Okay, that''s doable. I''d have to keep an array of legal droppables for each piece before each move. After each move, that pieces legal droppables would have to be updated. Doable. I might go that route. On 7/5/07, Diodeus <diodeus-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > > You could also change the list of droppables, so that the piece could > simply not be dropped on an illegal square. You would have to fetch > (or determine) the list of legal target squares before the move > begins. > > On Jul 5, 3:44 pm, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > Whether I do that or do it the way I''m doing it, it would still require > that > > I set revert to true or false depending on whether or not the move is > legal. > > So I''d have to set revert programatically. I know it can be done, but I > > don''t know how to do it. > > > > It shouldn''t matter whether I use ajax to verify the move or store the > legal > > moves locally. The ajax verification happens pretty instantaneously. The > > issue becomes making it either stay at revert=true or revert=false > depending > > on whether or not the move is legal. Because no matter what, I need the > > piece to stay there for a sec or so while the verification function > checks > > to see if the move is legal regardless of whether or not it''s checking > > locally or using ajax. > > > > So I need it to basically go to revert=false (and center itself in the > > square, which is a div) if the move is legal and revert if the move is > > false. I''m not even sure if this is possible, unless I come up with > > something custom. > > > > Thanks for the response. > > > > On 7/5/07, Jon Trelfa <jtre...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > > > > > > > On 7/5/07, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > > I can''t figure out an intelligent way to do this. If I set revert to > > > true, > > > > it will snap back to it''s original position when I want it to stay > at > > > the > > > > square while checking to see if the move is legit. If I set revert > to > > > false > > > > and the move is illegal, it''s not going to know where to snap back > to. > > > > > > Anybody have any ideas how to go about this? Thanks in advance. > > > > > How about pre-fetching your legal moves for the piece using the XHR > call? > > > Then you can validate the move against a local set of data. For > example: > > > > > In a grid where the rows are numbered and the columns are lettered, > > > we''ll say a pawn is in square 4A > > > The pre-fetch for that piece says that it can only legally move to > > > square 5A (if there were an opponent''s piece in 5B, that square would > > > also be returned) > > > The user moves the piece forward 1 square (legally) to square 5A (If > > > they tried 6A, the event would return false and snap the piece back.) > > > If the ondrop event says it''s "OK" for that piece, you fire off an XHR > > > to get the next list of legal moves for that piece > > > > > I''m sure it will get more complicated with regards to taking another > > > person''s piece, etc - but using Ajax before the person actually has to > > > make the move would be more sensible rather than making the person > > > wait for the response for each move. > > > > > You could then maintain an object with 32 (16 black, 16 white) > > > elements with arrays of "legal" moves. > > > > > My two cents :) > > > > > Jon > > > > >--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
Well, I''d want the same kind of snapping action that scriptaculous drag and drop gives. I guess I could write a function that keeps calling itself to slowly move the piece back. Not sure if this is the way to go. And now that I think about it, here''s the issue with creating a list of droppables for each piece. If I do that, the list for each piece has to be updated with each and every move. I guess it''s just a little bit of text, so bandwidth wouldn''t be that much of an issue. But, for example, when black moves a piece, I''d have to fire off a script to get white''s latest draggables and vice versa. I guess it''s not that bad, but I don''t know. I guess if nothing else, I can go this route. On 7/5/07, Diodeus <diodeus-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> > > I believe revert=true has to be set before the dragging occurs. > > Why not just move the chess piece back to the square it came from > using a function if the move is illegal? > > On Jul 5, 3:44 pm, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > Whether I do that or do it the way I''m doing it, it would still require > that > > I set revert to true or false depending on whether or not the move is > legal. > > So I''d have to set revert programatically. I know it can be done, but I > > don''t know how to do it. > > > > It shouldn''t matter whether I use ajax to verify the move or store the > legal > > moves locally. The ajax verification happens pretty instantaneously. The > > issue becomes making it either stay at revert=true or revert=false > depending > > on whether or not the move is legal. Because no matter what, I need the > > piece to stay there for a sec or so while the verification function > checks > > to see if the move is legal regardless of whether or not it''s checking > > locally or using ajax. > > > > So I need it to basically go to revert=false (and center itself in the > > square, which is a div) if the move is legal and revert if the move is > > false. I''m not even sure if this is possible, unless I come up with > > something custom. > > > > Thanks for the response. > > > > On 7/5/07, Jon Trelfa <jtre...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > > > > > > > On 7/5/07, Red <redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote: > > > > I can''t figure out an intelligent way to do this. If I set revert to > > > true, > > > > it will snap back to it''s original position when I want it to stay > at > > > the > > > > square while checking to see if the move is legit. If I set revert > to > > > false > > > > and the move is illegal, it''s not going to know where to snap back > to. > > > > > > Anybody have any ideas how to go about this? Thanks in advance. > > > > > How about pre-fetching your legal moves for the piece using the XHR > call? > > > Then you can validate the move against a local set of data. For > example: > > > > > In a grid where the rows are numbered and the columns are lettered, > > > we''ll say a pawn is in square 4A > > > The pre-fetch for that piece says that it can only legally move to > > > square 5A (if there were an opponent''s piece in 5B, that square would > > > also be returned) > > > The user moves the piece forward 1 square (legally) to square 5A (If > > > they tried 6A, the event would return false and snap the piece back.) > > > If the ondrop event says it''s "OK" for that piece, you fire off an XHR > > > to get the next list of legal moves for that piece > > > > > I''m sure it will get more complicated with regards to taking another > > > person''s piece, etc - but using Ajax before the person actually has to > > > make the move would be more sensible rather than making the person > > > wait for the response for each move. > > > > > You could then maintain an object with 32 (16 black, 16 white) > > > elements with arrays of "legal" moves. > > > > > My two cents :) > > > > > Jon > > > > >--~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
On 7/5/07, Red <redlocyellow-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> Well, I''d want the same kind of snapping action that scriptaculous drag and > drop gives. I guess I could write a function that keeps calling itself to > slowly move the piece back. Not sure if this is the way to go. >I think you''re too focused on the revert in the "boolean" sense. The draggable effect takes into account the "delta" of the movement. If "revert" is set to true, it moves the element back to the original position based on the difference between where it started and where it is when you finish dragging. Take a look at the dragdrop.js file around line 423 - there''s a method used called "currentDelta". You can see in that code how the revert uses that to move stuff back - that should help you to handle the revert if you need to without worrying about hacking your code to simulate the way scriptaculous already does it.> And now that I think about it, here''s the issue with creating a list of > droppables for each piece. If I do that, the list for each piece has to be > updated with each and every move.Why for each piece? If I move a piece onto a square and 3 other pieces have that square in their list of "legal" moves, only those pieces affected would need to be updated rather than the whole set. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Ruby on Rails: Spinoffs" group. To post to this group, send email to rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org To unsubscribe from this group, send email to rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org For more options, visit this group at http://groups.google.com/group/rubyonrails-spinoffs?hl=en -~----------~----~----~----~------~----~------~--~---
Hi Red
I did this with a drag and drop calendar i''ve been writing.
Basically, you leave revert as false, and do your checks (in my case im
making sure the item is not being dropped on another) and if it is, i use
code adapted from the sortable code to do a manual revert to the ''last
known
position''.
In that, every time a peice is moved, I update an X/Y on the item to say
where it is now, a valid position... then ondrop, I can move it back to that
position if the drop/move is invalid.
var dropAllowed = true;
//Perform checks to set dropAllowed to be false.
if (dropAllowed)
{
//Valid drop
var ContainerOffset =
Position.cumulativeOffset($(''Board''));
//hourEl is the ''square'' the droppable got dropped on, Im
not sure why I
need the +30s but it fixes a location bug I''m having
var newTop = parsePx(hourEl.style.top) + 30;
var newLeft = parsePx(hourEl.style.left) + 30;
//We create a 2 element array to store a position array for the current
position, more for readability below
var currentPos =
[parseInt(Element.getStyle(draggable,''top'')),parseInt(
Element.getStyle(draggable,''left''))]
//This makes the element ''snap'' (its more of a slide) to the
correct
position in the dropped square
revertEffect(draggable, currentPos[0]-newTop, currentPos[1]-newLeft);
//We''ve moved the item so update the property on the object to state
where
it is now, used later for an invalid move
draggable.lastKnownPosition = [parseInt(newTop),parseInt(newLeft)]
} else {
//This is an invalid drop
//We create a 2 element array to store a position array for the
current position, more for readability below
var currentPos = [parseInt(Element.getStyle
(draggable,''top'')),parseInt(Element.getStyle(draggable,''left''))]
//Retrieve last known pos
var lastKnownPos = draggable.lastKnownPosition;
//Move the element back to the last known valid location/square
revertEffect(draggable, currentPos[0]-lastKnownPos[0],
currentPos[1]-lastKnownPos[1]);
}
//This is as per dragdrop.js but moved to be a top level function not a
class method.
function revertEffect(element, top_offset, left_offset) {
var dur = Math.sqrt(Math.abs(top_offset^2)+Math.abs(left_offset^2))*
0.02;
new Effect.Move(element, { x: -left_offset, y: -top_offset,
duration: dur,
queue: {scope:''_draggable'',
position:''end''}
});
}
You may also want to store which square the item is on, as well as the X/Y
because you know what square it has been dropped on when it is a valid drop
and its more work to calculate its position every time.
This way you can retreive it for your local lookup/ajax call.
Hope this helps.
Gareth
On 7/6/07, Red <redlocyellow-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> Well, I''d want the same kind of snapping action that scriptaculous
drag
> and drop gives. I guess I could write a function that keeps calling itself
> to slowly move the piece back. Not sure if this is the way to go.
>
> And now that I think about it, here''s the issue with creating a
list of
> droppables for each piece. If I do that, the list for each piece has to be
> updated with each and every move. I guess it''s just a little bit
of text, so
> bandwidth wouldn''t be that much of an issue. But, for example,
when black
> moves a piece, I''d have to fire off a script to get
white''s latest
> draggables and vice versa. I guess it''s not that bad, but I
don''t know. I
> guess if nothing else, I can go this route.
>
>
> On 7/5/07, Diodeus <diodeus-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
wrote:
> >
> >
> > I believe revert=true has to be set before the dragging occurs.
> >
> > Why not just move the chess piece back to the square it came from
> > using a function if the move is illegal?
> >
> > On Jul 5, 3:44 pm, Red <
redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> > > Whether I do that or do it the way I''m doing it, it
would still
> > require that
> > > I set revert to true or false depending on whether or not the
move is
> > legal.
> > > So I''d have to set revert programatically. I know it can
be done, but
> > I
> > > don''t know how to do it.
> > >
> > > It shouldn''t matter whether I use ajax to verify the
move or store the
> > legal
> > > moves locally. The ajax verification happens pretty
instantaneously.
> > The
> > > issue becomes making it either stay at revert=true or
revert=false
> > depending
> > > on whether or not the move is legal. Because no matter what, I
need
> > the
> > > piece to stay there for a sec or so while the verification
function
> > checks
> > > to see if the move is legal regardless of whether or not
it''s checking
> >
> > > locally or using ajax.
> > >
> > > So I need it to basically go to revert=false (and center itself
in the
> > > square, which is a div) if the move is legal and revert if the
move is
> > > false. I''m not even sure if this is possible, unless I
come up with
> > > something custom.
> > >
> > > Thanks for the response.
> > >
> > > On 7/5/07, Jon Trelfa
<jtre...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> > >
> > >
> > >
> > > > On 7/5/07, Red <
redlocyel...-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> > > > > I can''t figure out an intelligent way to do
this. If I set revert
> > to
> > > > true,
> > > > > it will snap back to it''s original position
when I want it to stay
> > at
> > > > the
> > > > > square while checking to see if the move is legit. If I
set revert
> > to
> > > > false
> > > > > and the move is illegal, it''s not going to
know where to snap back
> > to.
> > >
> > > > > Anybody have any ideas how to go about this? Thanks in
advance.
> > >
> > > > How about pre-fetching your legal moves for the piece using
the XHR
> > call?
> > > > Then you can validate the move against a local set of data.
For
> > example:
> > >
> > > > In a grid where the rows are numbered and the columns are
lettered,
> > > > we''ll say a pawn is in square 4A
> > > > The pre-fetch for that piece says that it can only legally
move to
> > > > square 5A (if there were an opponent''s piece in 5B,
that square
> > would
> > > > also be returned)
> > > > The user moves the piece forward 1 square (legally) to
square 5A (If
> > > > they tried 6A, the event would return false and snap the
piece
> > back.)
> > > > If the ondrop event says it''s "OK" for
that piece, you fire off an
> > XHR
> > > > to get the next list of legal moves for that piece
> > >
> > > > I''m sure it will get more complicated with regards
to taking another
> >
> > > > person''s piece, etc - but using Ajax before the
person actually has
> > to
> > > > make the move would be more sensible rather than making the
person
> > > > wait for the response for each move.
> > >
> > > > You could then maintain an object with 32 (16 black, 16
white)
> > > > elements with arrays of "legal" moves.
> > >
> > > > My two cents :)
> > >
> > > > Jon
> >
> >
> > > >
> >
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Ruby on Rails: Spinoffs" group.
To post to this group, send email to
rubyonrails-spinoffs-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
To unsubscribe from this group, send email to
rubyonrails-spinoffs-unsubscribe-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org
For more options, visit this group at
http://groups.google.com/group/rubyonrails-spinoffs?hl=en
-~----------~----~----~----~------~----~------~--~---