My favorites | Sign in
Logo
                
Details: Show all Hide all

Today

  • 6 hours ago
    issue 369 (new FICS Seek dialog) commented on by lobais   -   Igor, While it is true, that PyChess might currently be using a few too many windows, grouping them together is not something we can do. That is the task of the windowmanager. And btw. nice work image-commenting everything :D
    Igor, While it is true, that PyChess might currently be using a few too many windows, grouping them together is not something we can do. That is the task of the windowmanager. And btw. nice work image-commenting everything :D
  • 6 hours ago
    issue 369 (new FICS Seek dialog) commented on by lobais   -   Happy New Year everyone! Matt, this is great work indeed! Good to see you have found ways around most of the difficulties, I know this issue contains. I think the idea of a configuration window is good. The alternative would be the small popout things, which we have in the NewGameMenu, but here the configuration buttons are not in a dialog but a mainwindow, and thus a dialog feels good. There is as well too much information put it in anything less than a dialog. I agree with every single point of your TODO. Only I'm not 100% sure what you mean by 'rating point increments should be smaller'. Does this refer to the step of the sliders? I'm impressed with the amount of activity in this thread. My opinions on the discussed topics are as follows: [On the Time Settings] The SeekDialog differs from the NewGameDialog in that it is a configuration panel and not an active object. Thus the SeekDialog really shouldn't have any sub-configuration panels itself. The is no need for the SeekDialog to remember time settings, the outer seek panel already does exactly that. I think therefore, that the time-spinners should be moved directly into the dialog, replacing the current radiobuttons. [On the Tolerance Slider] First of all gtk contains no double slider, or anything similar with more than very basic features. Secondly, I think a rating window is most intuitive, when it has a position and a width. No need of fiddling with upper and lower bounds. Possibly the tolerance could even be turned into a link 'Change tolerance' which on click shows the slider. Most of the time a standard width of 200 or so will work just fine. [On the Chess Variants] Similar to the Time Settings, we don't want memory in the configuration panel. A single combobox should do. I don't think we need a 'don't care' option. As far as I remember, that is not even possible on FICS. [On Tabs] Tabs could be a fine solution, if there was only one "Gear" button in the SeekPanel, but in keeping consistency with the NewGameDialog I think it is better to have one "Gear" button for each item, and thus no tabs. [On the Rating Checkbox] The rating Checkbox is a bit problematic because it is a lone option, and we don't have any 'miscellaneous' options to group it with. Maybe the best solution is to bring it out right next to the Manual checkbox. As with the variants it is not possible in fics to 'not care' about rating. At the end of the day your opinion will be strongest, since you code it. :) I'm looking forward to see this in trunk ;) Cheers, Thomas
    Happy New Year everyone! Matt, this is great work indeed! Good to see you have found ways around most of the difficulties, I know this issue contains. I think the idea of a configuration window is good. The alternative would be the small popout things, which we have in the NewGameMenu, but here the configuration buttons are not in a dialog but a mainwindow, and thus a dialog feels good. There is as well too much information put it in anything less than a dialog. I agree with every single point of your TODO. Only I'm not 100% sure what you mean by 'rating point increments should be smaller'. Does this refer to the step of the sliders? I'm impressed with the amount of activity in this thread. My opinions on the discussed topics are as follows: [On the Time Settings] The SeekDialog differs from the NewGameDialog in that it is a configuration panel and not an active object. Thus the SeekDialog really shouldn't have any sub-configuration panels itself. The is no need for the SeekDialog to remember time settings, the outer seek panel already does exactly that. I think therefore, that the time-spinners should be moved directly into the dialog, replacing the current radiobuttons. [On the Tolerance Slider] First of all gtk contains no double slider, or anything similar with more than very basic features. Secondly, I think a rating window is most intuitive, when it has a position and a width. No need of fiddling with upper and lower bounds. Possibly the tolerance could even be turned into a link 'Change tolerance' which on click shows the slider. Most of the time a standard width of 200 or so will work just fine. [On the Chess Variants] Similar to the Time Settings, we don't want memory in the configuration panel. A single combobox should do. I don't think we need a 'don't care' option. As far as I remember, that is not even possible on FICS. [On Tabs] Tabs could be a fine solution, if there was only one "Gear" button in the SeekPanel, but in keeping consistency with the NewGameDialog I think it is better to have one "Gear" button for each item, and thus no tabs. [On the Rating Checkbox] The rating Checkbox is a bit problematic because it is a lone option, and we don't have any 'miscellaneous' options to group it with. Maybe the best solution is to bring it out right next to the Manual checkbox. As with the variants it is not possible in fics to 'not care' about rating. At the end of the day your opinion will be strongest, since you code it. :) I'm looking forward to see this in trunk ;) Cheers, Thomas
  • 8 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   #18 -> ok, you are right tabs would complicate too much. So probably better solution like in #18, but without tabs. #16 -> ok, you are right, chess player should know what kind of game he/she would like to play. #20 -> you are right, what I don't like is that "Rated Game" option looks like it belongs to "Chess Variant" section. So maybe there should be section like "Rating Variant" (the same logic as "Chess Variant") and then check button bellow "Rated Game". So some kind of separation between "Chess Variant" and "Rated Game".
    #18 -> ok, you are right tabs would complicate too much. So probably better solution like in #18, but without tabs. #16 -> ok, you are right, chess player should know what kind of game he/she would like to play. #20 -> you are right, what I don't like is that "Rated Game" option looks like it belongs to "Chess Variant" section. So maybe there should be section like "Rating Variant" (the same logic as "Chess Variant") and then check button bellow "Rated Game". So some kind of separation between "Chess Variant" and "Rated Game".
  • 8 hours ago
    issue 369 (new FICS Seek dialog) commented on by gbtami   -   Comments on #14 If there is a double slider widget in pygtk, it's a good idea to use it here. #15 I agree that time control section is redundant, so "No control" and "Time limit" is enough, as shown in #18 #16 I don'n think "Don't Care" in Chess variant section is a viable option. #18 Using tabs for setting same options seems overcomplication for me. #19 Agreed. #20 Using radio buttons for only 2 options instead of checkbox is not a good idea IMO.
    Comments on #14 If there is a double slider widget in pygtk, it's a good idea to use it here. #15 I agree that time control section is redundant, so "No control" and "Time limit" is enough, as shown in #18 #16 I don'n think "Don't Care" in Chess variant section is a viable option. #18 Using tabs for setting same options seems overcomplication for me. #19 Agreed. #20 Using radio buttons for only 2 options instead of checkbox is not a good idea IMO.
  • 11 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   Hi, to make "Rated Game" consistent with other options in the same dialog. Regards
    Hi, to make "Rated Game" consistent with other options in the same dialog. Regards
  • 11 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   Hi, for Opponent Strength any better to simplify this to only two options. This is in my opinion way more simple to understand. Regards
    Hi, for Opponent Strength any better to simplify this to only two options. This is in my opinion way more simple to understand. Regards
  • 11 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   Hi, another idea for Edit Seek dialog. Use tabs instead of "Edit Seek" dialog. Don't know how well does tabs fit in existing program. Regards
    Hi, another idea for Edit Seek dialog. Use tabs instead of "Edit Seek" dialog. Don't know how well does tabs fit in existing program. Regards
  • 11 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   Hi, it looks like there is some "escaping window" problem. See comment in attached picture. Regards
    Hi, it looks like there is some "escaping window" problem. See comment in attached picture. Regards
  • 11 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   Hi, in "Chess Variant" section there is missing "Don't Care" option. Regards
    Hi, in "Chess Variant" section there is missing "Don't Care" option. Regards
  • 11 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   Hi, if I understand correctly, there is for each "Seek Panel" option (Standard, Blitz, Lightning) a button at right site which oppens a "Edit Seek" dialog. What is little bit confusing is "Time Control" option in this dialog. I suggest: 1. remove "Time Controls" option from "Edit Seek" panel. 2. add "No Controls" option to main "Seek Panel" window. But I think it should have better name something like "Any player without any limitation". Regards
    Hi, if I understand correctly, there is for each "Seek Panel" option (Standard, Blitz, Lightning) a button at right site which oppens a "Edit Seek" dialog. What is little bit confusing is "Time Control" option in this dialog. I suggest: 1. remove "Time Controls" option from "Edit Seek" panel. 2. add "No Controls" option to main "Seek Panel" window. But I think it should have better name something like "Any player without any limitation". Regards
  • 11 hours ago
    issue 369 (new FICS Seek dialog) commented on by igor2x   -   Hi, I suggest instead of "Strength Center" and "Tolerance" to use only one double slider. This slider should provide enough info for min and max settings. It is must more logical for average chess player. I am not a python programmer, so I don't know if there is an object with two sliders. But if there is not such an object, then just put two slider objects one on the left for min value and one the right for max value. Regards
    Hi, I suggest instead of "Strength Center" and "Tolerance" to use only one double slider. This slider should provide enough info for min and max settings. It is must more logical for average chess player. I am not a python programmer, so I don't know if there is an object with two sliders. But if there is not such an object, then just put two slider objects one on the left for min value and one the right for max value. Regards
  • 16 hours ago
    issue 369 (new FICS Seek dialog) commented on by mattgatto   -   Here are some screenshots of what the new seek dialog looks like. It incorporates Thomas' idea of having saved seeks. There is still some work left to do and which is in the screenshot of my ToDo list which I've also attached. Feedback on both the code and UI is welcome.
    Here are some screenshots of what the new seek dialog looks like. It incorporates Thomas' idea of having saved seeks. There is still some work left to do and which is in the screenshot of my ToDo list which I've also attached. Feedback on both the code and UI is welcome.

Last 30 days

  • Dec 25, 2009
    ChessClients (Chess clients for the linux desktop) Wiki page edited by gbtami   -   Revision r1564 Edited wiki page through web user interface.
    Revision r1564 Edited wiki page through web user interface.
  • Dec 25, 2009
    ChessClients (Chess clients for the linux desktop) Wiki page edited by gbtami   -   Revision r1563 Edited wiki page through web user interface.
    Revision r1563 Edited wiki page through web user interface.
  • Dec 23, 2009
    r1562 (Fixed regex for starting an adjourned game in FICS) committed by mattgatto   -   Fixed regex for starting an adjourned game in FICS
    Fixed regex for starting an adjourned game in FICS
  • Dec 22, 2009
    issue 510 (Use python-keyring for saving FICS password) Labels changed by gbtami   -  
    Labels: Priority-Low
    Labels: Priority-Low
  • Dec 21, 2009
    issue 510 (Use python-keyring for saving FICS password) commented on by lobais   -   Another option is the gnome-keyring: http://live.gnome.org/GnomeKeyring But I'm not quire sure what using a keyring saves us. The passwords will always be easily accessible, if you have access to the user account anyways. Pidgin has a long discussion of this: http://developer.pidgin.im/wiki/PlainTextPasswords
    Another option is the gnome-keyring: http://live.gnome.org/GnomeKeyring But I'm not quire sure what using a keyring saves us. The passwords will always be easily accessible, if you have access to the user account anyways. Pidgin has a long discussion of this: http://developer.pidgin.im/wiki/PlainTextPasswords
  • Dec 21, 2009
    issue 510 (Use python-keyring for saving FICS password) commented on by lobais   -   Another option is the gnome-keyring: http://live.gnome.org/GnomeKeyring But I'm not quire sure what using a keyring saves us. The passwords will always be easily accessible, if you have access to the user account anyways. Pidgin has a long discussion of this: http://developer.pidgin.im/wiki/PlainTextPasswords
    Another option is the gnome-keyring: http://live.gnome.org/GnomeKeyring But I'm not quire sure what using a keyring saves us. The passwords will always be easily accessible, if you have access to the user account anyways. Pidgin has a long discussion of this: http://developer.pidgin.im/wiki/PlainTextPasswords
  • Dec 21, 2009
    issue 510 (Use python-keyring for saving FICS password) reported by gbtami   -   http://home.python-keyring.org/
  • Dec 13, 2009
    issue 441 (Deprecated warnings) Status changed by gbtami   -  
    Status: Fixed
    Status: Fixed
  • Dec 12, 2009
    issue 233 (Move glades to glade-3) commented on by mattgatto   -   glade-2 isn't even available anymore in Ubuntu 9.10 Karmic (my system) or Debian unstable. https://bugs.launchpad.net/ubuntu/+source/glade/+bug/358502/comments/5
    glade-2 isn't even available anymore in Ubuntu 9.10 Karmic (my system) or Debian unstable. https://bugs.launchpad.net/ubuntu/+source/glade/+bug/358502/comments/5
  • Dec 12, 2009
    issue 233 (Move glades to glade-3) commented on by mattgatto   -   In http://code.google.com/p/pychess/issues/detail?id=441&q=glade#c1 you mentioned we'd move to glade-3 in the next version. I'm assuming you mean before or after 0.10?
    In http://code.google.com/p/pychess/issues/detail?id=441&q=glade#c1 you mentioned we'd move to glade-3 in the next version. I'm assuming you mean before or after 0.10?
  • Dec 12, 2009
    issue 367 (Cannot start game on FICS) Status changed by mattgatto   -   Still seeing this bug?
    Status: Need-Info
    Still seeing this bug?
    Status: Need-Info
  • Dec 12, 2009
    issue 400 (PyChess crashes on loading a game (observing)) Status changed by mattgatto   -   Still seeing this bug?
    Status: Need-Info
    Still seeing this bug?
    Status: Need-Info
  • Dec 12, 2009
    issue 402 (Bug-buddy appears on "Send a seek") commented on by mattgatto   -   Still seeing this in svn trunk or the latest beta?
    Still seeing this in svn trunk or the latest beta?
  • Dec 12, 2009
    issue 497 (pychess-0.8.2-1.fc10.noarch.rpm can't play) Status changed by mattgatto   -   Do you still see this in the latest beta or in svn trunk?
    Status: Need-Info
    Do you still see this in the latest beta or in svn trunk?
    Status: Need-Info
  • Dec 11, 2009
    issue 509 ([PATCH] vanilla package uninstallable under debian) changed by gbtami   -   Applyed in r1561. Thx.
    Status: Fixed
    Labels: Type-Defect
    Applyed in r1561. Thx.
    Status: Fixed
    Labels: Type-Defect
  • Dec 11, 2009
    r1561 (make package installable under debian, patch from Adam Schma...) committed by gbtami   -   make package installable under debian, patch from Adam Schmalhofer
    make package installable under debian, patch from Adam Schmalhofer
  • Dec 11, 2009
    issue 509 ([PATCH] vanilla package uninstallable under debian) reported by blauebi...@gmx.de   -   What steps will reproduce the problem? with python2.5 (or earlier) installed run: $ dpkg-buildpackage $ sudo dpkg -i ../pychess_0.8_all.deb Did you use an installed version of PyChess or did you run in from a tarball/svn? svn: r1560 Patch included: Make package installable under debian * debian/control: "XS-Python-Version: >= 2.5" as with_statement needed * sidepanel/historyPanel.py: "import __future__" to support python 2.5 Unrelated Note: beta version should have used '~' instead of '-': 0.8-beta4 >> 0.8 >> 0.8~beta4
    What steps will reproduce the problem? with python2.5 (or earlier) installed run: $ dpkg-buildpackage $ sudo dpkg -i ../pychess_0.8_all.deb Did you use an installed version of PyChess or did you run in from a tarball/svn? svn: r1560 Patch included: Make package installable under debian * debian/control: "XS-Python-Version: >= 2.5" as with_statement needed * sidepanel/historyPanel.py: "import __future__" to support python 2.5 Unrelated Note: beta version should have used '~' instead of '-': 0.8-beta4 >> 0.8 >> 0.8~beta4
  • Dec 10, 2009
    r1560 (Rough blunder analysis checker script.) committed by lobais   -   Rough blunder analysis checker script.
    Rough blunder analysis checker script.
  • Dec 08, 2009
    ChessEngines (Chess engines) Wiki page edited by gbtami   -   Revision r1559 Edited wiki page through web user interface.
    Revision r1559 Edited wiki page through web user interface.

Older

  • Dec 06, 2009
    issue 507 (Feature request [two-liner]: support the "stockfish" chess e...) commented on by gbtami   -   Btw thx for reporting this!
    Btw thx for reporting this!
  • Dec 06, 2009
    issue 507 (Feature request [two-liner]: support the "stockfish" chess e...) Status changed by gbtami   -   commited in r1558
    Status: Fixed
    commited in r1558
    Status: Fixed
  • Dec 06, 2009
    r1558 (added stockfish to supported engines) committed by gbtami   -   added stockfish to supported engines
    added stockfish to supported engines
  • Dec 06, 2009
    issue 507 (Feature request [two-liner]: support the "stockfish" chess e...) commented on by UncombedCoconut   -   Yes, stockfish is in my PATH (I put it in /usr/local/bin). I may have been unclear about what I wanted, which is for pychess to detect the engine and show it in the GUI automatically. I did check that my two-liner made this happen (in 0.10beta2.) FWIW, the Debian unstable package puts stockfish in /usr/games/bin, same location as crafty and gnuchess. I don't know what other distros do; mine doesn't even package it yet.
    Yes, stockfish is in my PATH (I put it in /usr/local/bin). I may have been unclear about what I wanted, which is for pychess to detect the engine and show it in the GUI automatically. I did check that my two-liner made this happen (in 0.10beta2.) FWIW, the Debian unstable package puts stockfish in /usr/games/bin, same location as crafty and gnuchess. I don't know what other distros do; mine doesn't even package it yet.
  • Dec 06, 2009
    issue 508 ("Ponder" should probably be disabled for computer-vs-compute...) changed by lobais   -   You are right, and probably analyzers should be disabled automaticly as well. Btw. if you'd like to see which of your engines are stronger, you can also use the Arena script from /trunk/utilities/arena.py Thomas
    Status: Accepted
    Labels: Priority-Low Type-Enhancement
    You are right, and probably analyzers should be disabled automaticly as well. Btw. if you'd like to see which of your engines are stronger, you can also use the Arena script from /trunk/utilities/arena.py Thomas
    Status: Accepted
    Labels: Priority-Low Type-Enhancement
  • Dec 06, 2009
    issue 507 (Feature request [two-liner]: support the "stockfish" chess e...) commented on by lobais   -   Stockfish sounds good. If you open a terminal and types 'stockfish' (or what you wrote in the binname attribute), will that start the engine? Otherwise it is probably not in your PATH, and PyChess can't start it either. Cheers, Thomas
    Stockfish sounds good. If you open a terminal and types 'stockfish' (or what you wrote in the binname attribute), will that start the engine? Otherwise it is probably not in your PATH, and PyChess can't start it either. Cheers, Thomas
  • Dec 05, 2009
    issue 508 ("Ponder" should probably be disabled for computer-vs-compute...) reported by UncombedCoconut   -   I think that ponder (think on opponent's turn) should be disabled in computer-vs.-computer games. Reasoning is below. I'm willing to write a patch if devs agree but are busy. (I'd add a param "forcePonderOff" to setOptionStrength & everything up the call chain... admittedly a little hackish.) Thanks. :) Ponder is automatically enabled at strength >= 7. This normally makes sense, but games can be computer-vs.-computer. (I personally wanted to watch two open-source engines duke it out with default settings.) In that case a pondering engine steals system resources on its opponent's turn, which IMO is unfair. If both engines had ponder on, I would expect weaker play: the engines get the same total system resources, but waste more analyzing lines not played -- they may also suffer more cache misses, less accurate clock management, and suboptimal choice of # of threads.
    I think that ponder (think on opponent's turn) should be disabled in computer-vs.-computer games. Reasoning is below. I'm willing to write a patch if devs agree but are busy. (I'd add a param "forcePonderOff" to setOptionStrength & everything up the call chain... admittedly a little hackish.) Thanks. :) Ponder is automatically enabled at strength >= 7. This normally makes sense, but games can be computer-vs.-computer. (I personally wanted to watch two open-source engines duke it out with default settings.) In that case a pondering engine steals system resources on its opponent's turn, which IMO is unfair. If both engines had ponder on, I would expect weaker play: the engines get the same total system resources, but waste more analyzing lines not played -- they may also suffer more cache misses, less accurate clock management, and suboptimal choice of # of threads.
  • Dec 05, 2009
    issue 507 (Feature request [two-liner]: support the "stockfish" chess e...) reported by UncombedCoconut   -   Stockfish is a fork of Glaurung. It is reportedly stronger, and Glaurung's author has indicated that all future development will take place on the Stockfish branch. [ http://wbec-ridderkerk.forumotion.com/wbec-ridderkerk- news-info-f1/stockfish-glaurung-t402.htm#883 ] The program's interface is UCI v.1 just like Glaurung's; copying the "glaurung" lines in lib/pychess/Players/engineNest.py and changing the name was enough to make it work. I'm not really sure what to do about the flag, though, since the engine's developers are from different countries. Thanks for this great program!
    Stockfish is a fork of Glaurung. It is reportedly stronger, and Glaurung's author has indicated that all future development will take place on the Stockfish branch. [ http://wbec-ridderkerk.forumotion.com/wbec-ridderkerk- news-info-f1/stockfish-glaurung-t402.htm#883 ] The program's interface is UCI v.1 just like Glaurung's; copying the "glaurung" lines in lib/pychess/Players/engineNest.py and changing the name was enough to make it work. I'm not really sure what to do about the flag, though, since the engine's developers are from different countries. Thanks for this great program!
  • Dec 05, 2009
    r1557 (Fix insufficient material draws occuring in king pawn ending...) committed by mattgatto   -   Fix insufficient material draws occuring in king pawn endings when the last pawn is munched
    Fix insufficient material draws occuring in king pawn endings when the last pawn is munched
  • Nov 29, 2009
    issue 271 (Add "Premove" support for internet games) commented on by lobais   -   I'm sorry you are right. I forgot what I had written in Comment #2. In that case, I believe this request is fixed for now.
    I'm sorry you are right. I forgot what I had written in Comment #2. In that case, I believe this request is fixed for now.
  • Nov 29, 2009
    issue 271 (Add "Premove" support for internet games) commented on by mattgatto   -   It should be enabled whenever there is at least one local human player. If it's not, that's a bug.
    It should be enabled whenever there is at least one local human player. If it's not, that's a bug.
  • Nov 29, 2009
    issue 271 (Add "Premove" support for internet games) commented on by lobais   -   The fix works great. I think premove should be enabled in human vs. engine games as well. This improves consistency and the pieces are already unlocked while the computer thinks.
    The fix works great. I think premove should be enabled in human vs. engine games as well. This improves consistency and the pieces are already unlocked while the computer thinks.
  • Nov 29, 2009
    issue 506 (strange painting bug) Status changed by lobais   -   Fixed in trunk. There was a stupid 'if not piece.x' instead of 'if piece.x == None' :/
    Status: Fixed
    Fixed in trunk. There was a stupid 'if not piece.x' instead of 'if piece.x == None' :/
    Status: Fixed
  • Nov 29, 2009
    r1556 (Fixed animation glitch for pieces in the a col) committed by lobais   -   Fixed animation glitch for pieces in the a col
    Fixed animation glitch for pieces in the a col
  • Nov 29, 2009
    r1555 (Fixed update glitches in sidepanels) committed by lobais   -   Fixed update glitches in sidepanels
    Fixed update glitches in sidepanels
  • Nov 29, 2009
    r1554 (the engine analyze signal should also send the currently pre...) committed by lobais   -   the engine analyze signal should also send the currently predicted score.
    the engine analyze signal should also send the currently predicted score.
  • Nov 29, 2009
    r1553 (Convenience) committed by lobais   -   Convenience
    Convenience
  • Nov 29, 2009
    ChessClients (Chess clients for the linux desktop) Wiki page edited by lobais   -   Revision r1552 Edited wiki page through web user interface.
    Revision r1552 Edited wiki page through web user interface.
  • Nov 29, 2009
    issue 271 (Add "Premove" support for internet games) commented on by mattgatto   -   Both of those should be fixed in revision 1551
    Both of those should be fixed in revision 1551
 
Hosted by Google Code