600a78cfe1
add skeleton of auth system based on previous work by parz1val, you need to install libcurl4-gnutls-dev (libcurl4-gnutls-dev:i386 if a 64 bit host)
DarthArgus2016-04-16 23:26:43 +00:00
95961e4a74
according to GCC, no it isn't - fixes a warning
DarthArgus2016-03-24 21:54:12 +00:00
8d44908b4b
this seems to have been test code...we can reenable if something feels missing later
DarthArgus2016-03-24 21:52:27 +00:00
5ed117d9c7
having these here just leaves unnecessary server crash risks...we'll just ignore these instead of crashing if we get any
DarthArgus2016-03-24 21:43:51 +00:00
cd997925eb
cleanup all the "#if 1" macro blocks
DarthArgus2016-03-24 21:27:54 +00:00
3b430656ba
remove SOE deprecated code and files - leaving in the #if 0 blocks with TODO and other possible uses for later
DarthArgus2016-03-24 21:07:31 +00:00
89f2b55420
more clear descriptions of the macros being called
DarthArgus2016-03-11 23:06:19 -06:00
b4e6224ce7
hacky but improves debugging output - may need some format cleanups (extra newline at end?) - gives us function, file, and line # for warnings and fatals
DarthArgus2016-03-11 23:01:27 -06:00
8212c2a273
bump hard mem limit...may go up to 3072
DarthArgus2016-02-26 10:11:28 -06:00
4093f2d0f4
disable the memory manager entirely! - server scripts may load _slightly_ slower at least, visibly, but we don't crash this way
DarthArgus2016-02-26 09:54:55 -06:00
1c1cf54a7b
ensure release and debug build properly
DarthArgus2016-02-24 14:11:14 -06:00
7e1c44fcbe
this seems to stabilize things; if it does not we can try setting DO_GUARDS and the other flags back on one by one...
DarthArgus2016-02-24 10:45:31 -06:00
65a79f2c07
force memory manager to always run in production mode, thus reducing usage of memory, thus hopefully making it play nice with java 8
DarthArgus2016-02-24 10:05:15 -06:00
b86eba889f
by messing with free memory and preallocating we break java 8...so we'll fix it
DarthArgus2016-02-23 22:09:52 -06:00
2a388362d6
permgen was 64mb so we'll set that for metaspace too
DarthArgus2016-02-23 18:36:59 -06:00
bf626d68e4
default trap script errors to false as they tend to crash the newer javas at the moment
DarthArgus2016-02-19 19:16:07 -06:00
fe09cb38c1
add the javaUseXcheck option to ServerGame to allow for use of the jni linting/checking system, which spams shitloads of output
DarthArgus2016-02-19 19:01:21 -06:00
d3666f0fa3
ok this plus (potentially) the removal of those flags makes it work
DarthArgus2016-02-19 18:40:46 -06:00
467c73574e
undo the "brackets mothafucka" commit as this should probably still be proper
DarthArgus2016-02-19 18:24:46 -06:00
be3df7efda
this fixes one problem, but we still run into the issue with addAllIfNotPresent
DarthArgus2016-02-19 14:30:32 -06:00
ad8f6d5a99
another flint detected issue - init to 0 instead of with ourselves
DarthArgus2016-02-18 23:40:07 -06:00
9393e6e7af
another flint detected issue - init to 0 instead of with ourselves
DarthArgus2016-02-18 23:40:07 -06:00
3c50042abc
flint++ static analyzer detected items (not all that were likely legit positives, but a start) as well as some clang detected unused vars, cleanup_
DarthArgus2016-02-18 22:55:54 -06:00
877d0a9b62
flint++ static analyzer detected items (not all that were likely legit positives, but a start) as well as some clang detected unused vars, cleanup_
DarthArgus2016-02-18 22:55:54 -06:00
13019a2b14
null -> nullptr conversion i missed
DarthArgus2016-02-18 10:41:03 -06:00
aa891e3d8c
null -> nullptr conversion i missed
DarthArgus2016-02-18 10:41:03 -06:00
50cf4f7486
this warning is annoying and likely just a remnant of the japanese language version of the game - we can fix all the warned creatures if we turn it on, log them, and edit a custom creature_names file, but for now let's skip it
DarthArgus2016-02-15 11:48:37 -06:00
1feb285e77
this warning is annoying and likely just a remnant of the japanese language version of the game - we can fix all the warned creatures if we turn it on, log them, and edit a custom creature_names file, but for now let's skip it
DarthArgus2016-02-15 11:48:37 -06:00
920e8e658d
this warning is annoying and likely just a remnant of the japanese language version of the game - we can fix all the warned creatures if we turn it on, log them, and edit a custom creature_names file, but for now let's skip it
DarthArgus2016-02-15 11:48:37 -06:00
6373cf6d74
in theory this should allow for backward compatibility with the old busted IBM java 1.4, and any other 1.4 java, and above...testers welcome but i prefer to use 1.8 from now on
DarthArgus2016-02-15 00:38:32 -06:00
03dc62efba
newer standards prefer nullptr over NULL - this is most of them but there are others too
DarthArgus2016-02-11 15:16:14 -06:00
752347bc4b
Merge branch 'testing' of stellabellum.net:staff/src into newjava
DarthArgus2016-02-15 00:06:43 -06:00
6bb5137dc4
set mem manager upper limit to more sane value
DarthArgus2016-02-14 23:51:10 -06:00
c22a458c9f
set mem manager upper limit to more sane value
DarthArgus2016-02-14 23:51:10 -06:00
f70229a178
set metaspace size properly for java 8 and above - otherwise you'll run out of ram and crash
DarthArgus2016-02-14 12:27:12 -06:00
dd9db6d350
Revert "Revert "newer standards prefer nullptr over NULL - this is most of them but there are others too""
DarthArgus2016-02-11 21:19:18 -06:00
8e2160f33e
Revert "newer standards prefer nullptr over NULL - this is most of them but there are others too"
DarthArgus2016-02-11 19:44:31 -06:00
3e4cc36e7e
newer standards prefer nullptr over NULL - this is most of them but there are others too
DarthArgus2016-02-11 15:16:14 -06:00
a7f2d0ea6a
the code for trapping script crashes breaks things, apparently... TODO: maybe find the new/proper way to do this since it's kinda important, or just fix our scripts and have no need for crashes
DarthArgus2016-02-10 16:23:27 -06:00
0c84415392
cmake's still not always respecting JAVA_HOME
DarthArgus2016-02-10 10:18:42 -06:00
6e893ad8c8
cmake's still not always respecting JAVA_HOME
DarthArgus2016-02-10 10:18:42 -06:00
b4bc995250
TESTING (but seems to be fine): slight cleanup, allow for using "oracle" or "sun" interchangably in config, force use of JNI 1.4 or better since nobody uses java 1.2 or 1.3 for this anyway
DarthArgus2016-02-09 11:30:53 -06:00
946cddb179
Revert "actually this should be in release and debug"
DarthArgus2016-02-08 17:00:25 -06:00
fc9c045e53
actually this should be in release and debug
DarthArgus2016-02-08 16:52:22 -06:00
267535648d
fix potential memory leak on release builds - although the macro here may already terminate before releasing
DarthArgus2016-02-08 16:42:14 -06:00
4c03d5eec1
Revert "bump up the maximum/initial jvm stack limits since we have PAE - 32 bit java seems limited to 2GB max"
DarthArgus2016-02-03 13:58:03 -06:00
77661f7480
bump up the maximum/initial jvm stack limits since we have PAE - 32 bit java seems limited to 2GB max
DarthArgus2016-02-03 12:09:37 -06:00
10e3ca6a11
remove commented code that isn't used in orig leak either
DarthArgus2016-02-03 11:11:34 -06:00
97feb98d31
i think this bit was ok and it was a false positive - someone prove me wrong if you think otherwise
DarthArgus2016-02-03 02:11:30 -06:00
adf7f67682
fix some cppcheck detected (potential) issues
DarthArgus2016-02-03 02:00:56 -06:00
77ae27abca
Merge branch 'testing' of git.stellabellum.net:staff/src into testing
DarthArgus2016-01-31 22:47:56 -06:00
8e048cc02f
clang friendliness and a string fix
DarthArgus2016-01-31 22:47:28 -06:00
5b9b1e6a97
start cleaning up some things - switches make me happy
DarthArgus2016-01-31 19:08:47 -06:00
8b43a2d34c
if it doesn't have a portal property, it probably isn't supposed to...we should potentially investigate whatever script loads garages and things to remove this, but it really doesn't matter
DarthArgus2016-01-31 17:37:42 -06:00
7c90f74f43
kill another useless warning, but leave another intact for the falcon npe instance - the only one left now is more of an informational message in debug mode, letting us know that the falcon is invulnerable
DarthArgus2016-01-31 16:58:24 -06:00
97e89e2be2
the "special" npe falcon instance throws useless warns, this should be the last of them silenced
DarthArgus2016-01-31 16:16:55 -06:00
9f83b6ab3c
these can fail silently since the default 1.0 is sane
DarthArgus2016-01-31 15:31:24 -06:00
f32c8b4f21
theoretically this will fix our unsigned comparison warnings
DarthArgus2016-01-31 12:51:49 -06:00