Miroirs/xwords
1
0
Fork 0
mirror of git://xwords.git.sourceforge.net/gitroot/xwords/xwords synced 2025-01-09 05:24:44 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
5340 commits 21 branches 243 tags 33 MiB
Commit graph

251 commits

Author SHA1 Message Date
eehouse@eehouse.org
82615f409b Merge branch 'android_branch' into send_in_background 2011-10-04 06:49:19 -07:00
Andy2
f4b8b2806e remove unused parameter 2011-10-04 06:48:53 -07:00
Andy2
5dd21a16ea don't crash if words stream is null, as it will be when turn's a pass. 2011-10-04 06:45:16 -07:00
Andy2
35e4c21f17 add util_informMove to jni. Use it, but not to advantage yet. 2011-10-02 16:42:10 -07:00
Eric House
f0c3e2c5f0 Merge branch 'android_branch' into send_in_background 2011-09-22 17:22:01 -07:00
Andy2
fcdbed0920 fix to work with changed C api 2011-09-22 06:49:34 -07:00
Andy2
92614370ce Merge branch 'android_branch' into send_in_background 
Conflicts:
	xwords4/common/comms.c
 2011-09-20 06:21:54 -07:00
Andy2
dced47e202 export server_listWordsPlayed via jni. 2011-09-18 19:06:23 -07:00
eehouse@eehouse.org
98c2617613 Merge branch 'android_branch' into send_in_background 2011-09-14 06:47:12 -07:00
Andy2
c9c498786e add to game state whether any tiles selected. Use that to disable 
'commit exchange' button when there's nothing to commit, and get rid
of error message used when committing nothing since it's now
impossible (on Android).
 2011-09-09 19:04:25 -07:00
eehouse@eehouse.org
eb0e41e978 Merge branch 'android_branch' into send_in_background 2011-09-08 21:38:50 -07:00
Andy2
2aa5447301 remove some logging 2011-09-08 19:03:08 -07:00
Eric House
84931f34ff Merge branch 'android_branch' into send_in_background 2011-09-07 21:44:36 -07:00
Andy2
8087153b73 log name of dict being set -- as an easy way to confirm that the right 
dicts are getting passed to the jin.
 2011-09-03 20:19:51 -07:00
Andy2
da475347a5 get rid of unused constants 2011-09-02 18:36:26 -07:00
Andy2
2d2f3e23fd change signature of util_playerScoreHeld(): pass player num rather 
than formatted text. Client code can use new
model_getPlayersLastScore() or whatever else it wants to build text to
show the user.
 2011-09-02 18:36:03 -07:00
Andy2
42950c38bd export model_getPlayersLastScore() 2011-09-02 18:33:34 -07:00
Andy2
e94524958f get rid of util_setInTrade(); fix passing of selBits when trading 
(broken with prev checkin.)
 2011-09-01 19:51:26 -07:00
eehouse@eehouse.org
b6e5327b1a Merge branch 'android_branch' into send_in_background 2011-08-31 06:46:42 -07:00
Andy2
671877cbca implement new util_ methods in jni and java; remove old miniwindow 
stuff.  In java, respond to util_playerScoreHeld() and
util_bonusSquareHeld() by putting up toasts with existing strings.
For util_setInTrade(), fix exchange mode. Add two buttons at bottom
that replace toolbar (and corresponding buttons on menu). Redraw
scoreboard and board with high transparency to make it clear they're
disabled.  Still to do: deal with case where ability to switch which
player's tray is visible is enabled.
 2011-08-31 06:37:33 -07:00
Eric House
c01981bc8f Make miniwindow stuff compile-time optional. The goal's to use native 
features where they look better or are easier to manage, e.g. Toasts
and dialogs on Android.
 2011-08-29 20:36:01 -07:00
eehouse@eehouse.org
147687890f Merge branch 'android_branch' into send_in_background 2011-08-24 06:52:51 -07:00
Andy2
84e61f3b3f remove no-op line 2011-08-22 18:43:42 -07:00
Andy2
c41af221b7 fix non-debug compile warning 2011-08-21 11:33:23 -07:00
Andy2
08789c4097 call server_do() up to five times to allow some extra work to get 
done.  Required if a host is also a robot.
 2011-08-17 19:48:55 -07:00
Andy2
8148c7699a don't loop forever on server_do(). It's broken w.r.t. guaranteeing 
that it can always change state.
 2011-08-17 06:52:04 -07:00
Andy2
7ab8c6eca4 re-factor sending code so essentially same message (without cookieID 
that implies a connection) is sent and received by no-conn code.  Use
flags to determine if comms can support no-conn sending without making
up the message only to have it fail to send.
 2011-08-16 19:36:23 -07:00
Andy2
066cef0306 using new send-noconn proc is too expensive if it's only going to fail 
(return false) so add new iface comms can call once to get flags to
tell it whether to use that proc.  One implementation of
TransportProcs returns one flag; the other, the other.
 2011-08-16 19:31:08 -07:00
Andy2
dfab747504 make server_do() take a new param to allow caller to call it until 
there's nothing more to do (without the assumption it's driven from an
event queue.)
 2011-08-16 19:28:59 -07:00
Andy2
f61748369a fix non-debug compile warning 2011-08-16 19:17:47 -07:00
Andy2
ea05c34d8f add the alternative send proc to the java transportprocs interface, 
and call through to it from the jni.  Add another implemenation of the
interface and pass that when opening games in the background.
 2011-08-10 21:06:05 -07:00
Andy2
e88af19e0a toward the ability to send replies to messages received in background 
without user having to open the game, which will e.g. allow a host to
assign tiles, or a robot to move, without the phone's owner noticing
there's a message.  This is on a branch because it may never work.
 2011-08-10 18:28:34 -07:00
Andy2
8adb925cca only load and pass in dict as byte[] when path isn't available 
(i.e. when it's a builtin), and use mmap in that case.  Works!  Still
need to try a speed test.
 2011-08-08 20:27:42 -07:00
Andy2
755d3e5bb2 as first step in using mmap for dictionaries instead of passing byte 
arrays into the jni, pass the full file paths in in addition to the
byte arrays.  This isn't possible with the built-in dicts, but does
work for the downloaded ones (which are usually larger).  This checkin
does the mmap and uses memcmp to verify that the bytes are the same as
passed in.  Next step is to not pass the bytes when the path will do
and to actually use the mmap'd ptr.
 2011-08-08 19:41:45 -07:00
Eric House
1c227c8921 remove code for engine progress -- it's been disabled for a while 
anyway as it gave away whether robot had blanks.
 2011-07-31 20:52:27 -07:00
Andy2
698d549fd1 assert fired for unknown reason. Fix behavior in non-assert-enabled 
build to do the right thing in that case.
 2011-07-21 05:07:41 -07:00
Eric House
febb640049 add name of remote player to notification that remote player moved. 
And remove a few dead strings.
 2011-07-15 18:13:55 -07:00
Andy2
ab8e4e437b call server_do before passing message to server to give it a chance to 
get any pending work done.
 2011-07-04 12:37:23 -07:00
Andy2
7024718c87 add timestamps to jni logs 2011-07-04 09:38:22 -07:00
Andy2
cd844a5bad plug memory leak flushed out by auto-juggle code: null dict doesn't 
mean the end of the list.
 2011-06-29 18:28:55 -07:00
eehouse@eehouse.org
484bb295d0 Merge branch 'android_branch' into android_invite 2011-06-06 22:08:49 -07:00
Andy2
7fc267bb24 google reports a crash in warnIllegalWord that only makes sense if I'm 
passing a 0-length array of bad words.  Which should never happen, and
would be caught by asserts in a debug build, but: when there are no
bad words don't call back into the java world.
 2011-06-06 20:35:34 -07:00
Andy2
232e4ec1e9 track missing players -- players for whom space is reserved locally 
and whose names have not yet been received from host/server -- as part
of summary.  They can be drawn differently to give a clear visual
indication which games are not in play and for which the user might
want to issue an invitation.
 2011-05-19 22:53:04 -07:00
Andy2
4c5c435fb5 add new parameter so client always gets notified on connection 
including how many players are still missing.
 2011-05-19 05:51:00 -07:00
Andy2
f05e02144c move setting gameID (if not already set) to creation of C version of 
the struct so it always happens, and also set it back into the java
version at the same time.  Makes creation of new games easier.
 2011-05-06 06:47:44 -07:00
Andy2
ba026f761d get rid of unused parameter 2011-04-28 18:47:16 -07:00
Andy2
09100dcb2c track lang code of last dict sent and don't send repeats. 2011-04-28 18:41:39 -07:00
Andy2
bf67a88c5a expect the array of dicts passed to include the game dict (as part of 
changes to always have a game dict and to have player dicts only when
they're different.)
 2011-04-25 20:16:38 -07:00
Andy2
6aa0d38c45 don't call through to java draw_dictChanged unless it's for the main 
game dict.
 2011-04-25 20:15:34 -07:00
Andy2
c814226a08 get rid of gi.dictName: don't pass it across the jni boundary 2011-04-21 18:29:37 -07:00
eehouse@eehouse.org
5a21aea2e5 Merge branch 'android_branch' of ssh://xwords.git.sourceforge.net/gitroot/xwords/xwords into android_branch 2011-04-13 06:58:58 -07:00
Eric House
baedbdff2c remove logging 2011-04-12 22:17:56 -07:00
Andy2
002bb40a09 report counts/values using language rather than dictionary name. To 
support that, pass lang name from java into jni.
 2011-04-11 18:55:42 -07:00
Andy2
3fb368692c dammit -- proofread those checkins! 2011-04-11 06:52:07 -07:00
Andy2
c6cd60deef Save and restore per-player dicts; load games that have 'em correctly. 
Robots default to BasEnglish dict and humans to CollegeEng.  Add new
per-game default for robot dict.  Still need to deal with language
changes and non-English case in general.
 2011-04-11 06:42:17 -07:00
Andy2
092dfdd590 not supposed to be checked in with debugging on 2011-04-11 06:33:31 -07:00
Andy2
dc70db5ba4 catch up with API change 2011-04-11 06:16:25 -07:00
Andy2
97d54c86eb snapshot: track dictName per player in addition to per game. Not yet 
used yet; need to pass into jni next.
 2011-04-08 06:50:08 -07:00
Eric House
c77d3039cc add XP_FREEP. Should be no change in behavior. 2011-04-07 18:07:45 -07:00
Andy2
55221c70b0 fix for new API 2011-04-03 21:51:51 -07:00
Andy2
3b7b4802a9 add ability to specify, track and use separate dictionaries for each 
player.  Works for gtk client.  Compiles for Android but there's no UI
yet to specify more than one dict.  Management of dupicate dicts
without duplicating memory -- refcounting -- will be up to the
platforms.
 2011-04-01 19:57:10 -07:00
Andy2
d9bb5197d4 Add two new menus to be available only when in trade mode. Hide and 
show them and another group depending on whether trading at the time.
To make that work, replace the individual calls made to dis/enable
toolbar buttons with a single jin call that takes a struct full of
booleans and make that struct available in BoardActivity where menus
are hidden/shown.  Remove the individual calls from the jin interface.
 2011-03-10 18:13:56 -08:00
Andy2
ce64f79cdd add draw_boardBegin 2011-02-24 22:24:24 -08:00
Andy2
5afb7aadff add dict_getTileValue 2011-02-24 22:23:57 -08:00
Andy2
f382f17d20 make game_receiveMessage return even if server doesn't do anything 
with it -- since caller will use the return value to decide whether to
save state.
 2011-02-24 17:33:10 -08:00
Andy2
f956df853d return "" rather than null when there's no java class backing util -- 
common code is less likely to crash.
 2011-02-09 06:17:38 -08:00
Andy2
c587a11ee2 Add turn and isrobot/islocal flags to summaries DB, and use to 
background in green the player whose turn it is provided that player's
local.
 2011-02-04 06:20:40 -08:00
Andy2
c2be642302 add game_getGi to fetch into java world any changes to the gi in the 
jni world.
 2011-02-01 18:53:29 -08:00
Andy2
383206f5dc move bonus squares into jni since they're constant rather than e.g. in 
an editable resource.
 2011-02-01 18:52:29 -08:00
Andy2
977feada82 use a reference to a byte array in the java heap (locking down the 
object) rather than allocating a new array in the C heap -- for the
DAWG data of a dictionary.  This can use up to 5% of the java heap for
huge dictionaries, but I'm hoping it fixes a problem reported by a
user of the large German dictionary that seems to involve allocation.
If I'm reading correctly, as long as I stay within 16M (24M or more on
newer devices) I'm sure to get my memory in the java world while it's
less a sure thing in the JNI world (where in addition linux's
aggressive overallocation is used, meaning I'll fail when I try to
swap in memory on write rather than get back NULL from malloc.)
 2011-01-31 18:56:20 -08:00
Andy2
d2ddc5cee7 turn off tile selection for android 2011-01-12 19:01:04 -08:00
Andy2
6495bd321f fix to compile with robotiq changes 2011-01-10 06:57:10 -08:00
Andy2
ff6fb7070e Android market reports a few crashes going out-of-bounds on array of 
bad words.  Add an assertion on the only path that seems likely to be
causing that.
 2010-12-17 18:57:43 -08:00
Andy2
53de4eab9b correctly detect empty dict and set top ptr to null. Fixes crash 
using empty dicts.
 2010-12-17 18:54:27 -08:00
eehouse@eehouse.org
64bcb8bb2b Revert "replace call to rand() with android_only function that reads two bytes" 
This reverts commit 1653b9f6e9.
 2010-12-08 05:38:41 -08:00
Andy2
4a12f7a96f Revert "in attempt to work around non-random rand() in jni world, call back" 
This reverts commit 99c92e779d.

Conflicts:

	xwords4/android/XWords4/jni/utilwrapper.c
 2010-12-07 18:43:53 -08:00
eehouse@eehouse.org
fcd179f6b7 Merge remote branch 'origin/android_branch' into dawg_fixup 2010-12-06 18:50:13 -08:00
Andy2
f50e6b3f1b turn dict_getLangCode into dict_getInfo, adding a struct that contains 
lang code and new word count.
 2010-12-06 18:25:23 -08:00
Andy2
ff9e03ef4b read rather than just skipping word count 2010-12-06 07:30:28 -08:00
Andy2
e2bb50ef62 skip new header if present 2010-12-06 07:22:07 -08:00
Eric House
c4cdc24b78 initial changes to add a header to xwd format so that stuff like 
number of words can be included.  Changed to build dicts and linux to
open them.  Android still needs to learn.  Also, some of the tools in
dawg/ need to be fixed to read old-format (pre-utf8) .xwd files.
 2010-12-05 19:33:10 -08:00
Andy2
426f990156 fix infinite loop creating networked game: util_rand() can't call into 
java in some cases so needs to fall back to returning rand().
 2010-12-03 06:43:40 -08:00
Andy2
984a5d6154 Fix yet again what shouldn't have been checked in. 2010-12-02 22:37:21 -08:00
Andy2
99c92e779d in attempt to work around non-random rand() in jni world, call back 
into the java world by making it a util_ctxt function.  Do same on
linux to test.  We'll see how it is -- and can back this commit out if
there's no improvement.
 2010-12-02 19:02:36 -08:00
Andy2
2e956e2f6e debug is not supposed to be checked in turned on 2010-12-01 18:13:55 -08:00
Andy2
1653b9f6e9 replace call to rand() with android_only function that reads two bytes 
from /dev/urandom.  rand() has been returning pretty crappy numbers
resulting in lots of clustering of tiles in trays.  Let's see if this
is any better.
 2010-11-29 17:12:12 -08:00
Andy2
71059e6dff remove nPlayers from game_summarize 2010-11-19 18:23:34 -08:00
Andy2
82876bda06 add seed and fetch it in jni 2010-11-08 17:23:16 -08:00
Andy2
c2472c0cb5 don't call java code if rect width is 0. Fixes crash (but not design 
problem) where first to connect in room specifies a timer and second
does not.  Second eventually gets message saying there's a timer but
has not layed out space in the board and java freaks out when later
asked to make a 0-width image.  This fix leaves the device not drawing
the timer until the game is reopened.  The better fix would trigger a
re-layout as soon as the change is made.
 2010-11-07 14:03:24 -08:00
Andy2
f36b253ae7 add number of players to GameSummary. Changes DB but it's already 
changed since last beta so I'm not upping the version.
 2010-11-06 18:24:44 -07:00
Andy2
b115ce14ed add devOrder (= hostID) to params passed on connect. Fix to pass new 
param from jni; use object to pass params into UI thread since there
are now too many for handler.
 2010-10-29 06:34:39 -07:00
Andy2
4cd15a9c6a Merge branch 'relay_proxy' of ssh://xwords.git.sourceforge.net/gitroot/xwords/xwords into relay_proxy 
Conflicts:
	xwords4/common/comms.c
 2010-10-29 05:27:07 -07:00
Andy2
645a042481 add room name to callback used when device connects: fix jni to compile 2010-10-28 18:46:28 -07:00
Andy2
ddeb744ff7 init room name to empty string 2010-10-28 18:43:06 -07:00
Eric House
4ae58342da test, don't assert: comms can be null 2010-10-28 05:45:56 -07:00
Eric House
eba5b77494 add comms_isConnected; public room variables default to false 2010-10-27 20:50:11 -07:00
Andy2
99dbbb443c add comms_canChat, which knows better than java code what state a 
connected game is in, and call it to determine whether to show the
chat icon.
 2010-10-21 19:14:10 -07:00
Andy2
cb4092bb39 make relayID a String rather than a byte[]; log result of pending 
messages query to confirm working.
 2010-10-06 05:21:19 -07:00
Eric House
b3c7cb3711 start on sending chat messages. Doesn't even compile -- committing to get onto another machine. 2010-10-04 05:38:05 -07:00
Eric House
1e13e92deb support for receiving and displaying chat messages. Compiles but not tested. 2010-10-03 18:32:39 -07:00