Commit graph

1426 commits

Author SHA1 Message Date
Enno Rehling
59069ae342 CID 26260: Resource leaks (RESOURCE_LEAK) 2015-10-29 16:28:59 +01:00
Enno Rehling
db6e9444d6 CID 26263: Memory - illegal accesses (UNINIT) 2015-10-29 16:27:10 +01:00
Enno Rehling
5f4c31af1c bad variable reuse leads to gcc confusion 2015-10-29 12:04:52 +01:00
Enno Rehling
8b6da79984 64 bit compatibility, thank you gcc. 2015-10-29 11:32:04 +01:00
Enno Rehling
4f25831407 coverity scan CID 22576: arguments in wrong order
I believe that was a false positive, and will try to aquelch it by naming the arguments better.
2015-10-29 11:24:01 +01:00
Enno Rehling
4384183ab8 coverity scan CID 22504: argument cannot be negative
handle error cases for ftell, just because they could happen, I guess?
2015-10-29 11:20:09 +01:00
Enno Rehling
976b6aaea1 coverity CID 22593: write to pointer after free
github issue #324
use free_land where we need it.
2015-10-29 11:17:31 +01:00
Enno Rehling
c298b7fd74 coverity CID 22501 imroper use of negative value
explicitly use SK_MAGIC instead of a lookup. Faster, easier to reason about.
2015-10-29 11:06:25 +01:00
Enno Rehling
8675002e68 coverity scan CIDs 22540 22541 22546: resource leak
the calendar should only ever be initialized once, but based on static analysis, I agree that this was sloppy.
2015-10-29 09:56:53 +01:00
Enno Rehling
2372d3aacd coverity scan CID 22516: out-of-bounds write
potentially a bug with bad XML files? only matters if we have those, but we shouldn't
2015-10-29 09:49:26 +01:00
Enno Rehling
8b92003fff coverity scan CID 22487: dereference after null check
we don't use many differnt calendars, so this has never happened.
2015-10-29 09:43:51 +01:00
Enno Rehling
e27c4b4164 coverity scan CID 22457: logically dead code
the original TEACH_FRIENDS feature could not possibly have worked?
2015-10-29 09:39:33 +01:00
Enno Rehling
f4c32acd05 coverity scan CID 22515: out-of-bounds access
increase buffer to stop coverity false positive
2015-10-29 09:32:31 +01:00
Enno Rehling
55b3bfd90e coverity scan CID 22574: copy into fixed size buffer
zOrder *should* probably be big enough, but static analysis says safe is safe.
2015-10-29 09:28:02 +01:00
Enno Rehling
6ebfd33478 coverity scan CID 22503: negative array index write 2015-10-29 09:24:58 +01:00
Enno Rehling
c7aa8c89b1 coverity scan CID 22551: dereference before null check 2015-10-29 09:22:00 +01:00
Enno Rehling
4da658584c coverity scan CID 22511: dereference null return value 2015-10-29 09:20:58 +01:00
Enno Rehling
178a740ecf coverity scan CID 22573: copy into fixed-size buffer
in theory, all IDs in the game should be no longer than 4 characters, but coverity doesn't know this.
2015-10-29 09:17:23 +01:00
Enno Rehling
78899ca801 coverity scan CID 22433: buffer not null terminated, prefer strlcpy over strncpy 2015-10-29 09:06:53 +01:00
Enno Rehling
04bf07a526 coverity scan (multiple CID) is confuced about a_remove, trying to help it. 2015-10-29 09:06:52 +01:00
Enno Rehling
65429a12c3 coverity scan CID 22585, 22586: result is not floating point 2015-10-29 09:06:51 +01:00
Enno Rehling
70d4a6af03 coverity scan CID 22553: dereference before null check 2015-10-29 09:06:50 +01:00
Enno Rehling
e60b739ea4 coverity scan CID 22448: logically dead code 2015-10-29 09:06:48 +01:00
Enno Rehling
f72314e915 coverity scan CID 22451 (logically dead code) 2015-10-29 09:06:47 +01:00
Enno Rehling
6cccdec4a2 coverity scan CID 22454 (logically dead code) 2015-10-29 09:06:46 +01:00
Enno Rehling
0328e84c6c coverity scan CID 22555: sizeof not portable
wrong type in sizeof (pointer instead of struct)
2015-10-29 09:06:43 +01:00
Enno Rehling
185a2ee29c Merge pull request #321 from ennorehling/feature/fixme-base36-static
eliminate static variable leak and allocation from base36
2015-10-27 13:30:12 +01:00
Enno Rehling
f80f37def3 eliminate static variable leak and allocation from base36 2015-10-27 12:48:34 +01:00
Enno Rehling
feb915fe4a eliminate deprecated reporting function, static variable 2015-10-27 12:40:04 +01:00
Enno Rehling
e3572ef14e eliminate deprecated call 2015-10-27 12:40:04 +01:00
Enno Rehling
e23a7fc87d deprecate report_kampfstatus and implement a replacement function. 2015-10-27 12:40:04 +01:00
Enno Rehling
f2b0ba1ec6 eliminate find_spell warning for games where nobody is cursed, and you cannot summon igjarjuk 2015-10-27 12:40:03 +01:00
Enno Rehling
b771ec9aab test-only leak: do some json surgery and delete the entire tree from the export 2015-10-15 09:06:44 +02:00
Enno Rehling
47e925473e always use a macro, not magical constants 2015-10-15 08:40:52 +02:00
Enno Rehling
c9e7a76d9b minor leak: release memory held by curse_type structures 2015-10-15 08:38:43 +02:00
Enno Rehling
d4655ee88a leak: do not create new message_type objects in test_cleanup 2015-10-15 08:29:20 +02:00
Enno Rehling
d79745812a unit needs to be in a region or test_cleanup will not free it 2015-10-14 22:19:15 +02:00
Enno Rehling
467ce2a719 visual studio compilation fix 2015-10-14 22:09:11 +02:00
Enno Rehling
ef1a2dfc4f fixed a crash 2015-10-14 22:06:40 +02:00
Enno Rehling
3200dd2dc8 leak: free locale_data 2015-10-14 21:51:43 +02:00
Enno Rehling
b5455531e7 do not leak terrain.herbs 2015-10-14 21:37:27 +02:00
Enno Rehling
c04bda102b call free_order for each remaining reference 2015-10-14 20:56:09 +02:00
Enno Rehling
a3b27ad6d2 leak: remember to call test_cleanup 2015-10-14 20:49:30 +02:00
Enno Rehling
03ff8f612d leaks: remember to call test_cleanup 2015-10-14 20:48:01 +02:00
Enno Rehling
284a74c878 leak: deleting a config entry did not free its name 2015-10-14 20:45:43 +02:00
Enno Rehling
f740f2829a leak: in tests, when correct message_type doesn't exist, arguments should not leak 2015-10-14 20:41:42 +02:00
Enno Rehling
144aeb23ac leaks: call cleanup, free triggers 2015-10-14 20:29:07 +02:00
Enno Rehling
0f32630590 leak: release message in test 2015-10-14 20:22:51 +02:00
Enno Rehling
5018bbb0c0 correctly release messages created in test_merge_split 2015-10-14 20:21:38 +02:00
Enno Rehling
203b058166 avoid memory leaks when running with --list 2015-10-14 20:04:03 +02:00
Enno Rehling
47eee29813 correct argument count 2015-10-14 18:40:37 +02:00
Enno Rehling
5d192cb4c9 fix compile: gcc is extra picky about const char** conversions. 2015-10-14 18:38:10 +02:00
Enno Rehling
ba0828afd4 rather complex implementation of selective test running (CuTest does not have this feature). 2015-10-14 18:34:41 +02:00
Enno Rehling
e6f2864327 leak: duplicate assignment to building_type.name in tests 2015-10-14 14:10:35 +02:00
Enno Rehling
2abdba9e8a refactor free_messagelist, fix leak in test_merge_split 2015-10-14 14:08:50 +02:00
Enno Rehling
1c2922aafa leak: spell names and other strings need to be released. 2015-10-14 13:58:55 +02:00
Enno Rehling
5f9c2e13a8 I hate umlaut.c, it's basically magic, and the magic doesn't work right (it leaks memory from line 158) 2015-10-14 13:56:19 +02:00
Enno Rehling
cd9cb24b48 extend test_umlaut a bit, which doesn't trigger issue #316 2015-10-14 13:42:12 +02:00
Enno Rehling
78a8795305 there is a bug in umlaut.c, adding a FIXME.
this happens during test_umlaut, and valgrind detects it.
2015-10-14 13:33:10 +02:00
Enno Rehling
254cda86e8 short commands speedup wasn't working (never got assigned) 2015-10-14 13:20:27 +02:00
Enno Rehling
4fc5ac7740 there is a memory leak in test_long_order_hungry, but the root cause lies deeper 2015-10-14 12:51:45 +02:00
Enno Rehling
469d312339 leak: test must free castorder and call cleanup 2015-10-14 12:23:58 +02:00
Enno Rehling
bdf4834dc2 leak: region_owner data was never released 2015-10-14 12:22:37 +02:00
Enno Rehling
8e9e9c9c0f eliminate leak from duplicate call to test_create_buildingtype 2015-10-14 12:21:13 +02:00
Enno Rehling
ae3210ec87 pervious commit did not compile: small copy/paste error 2015-10-14 12:14:33 +02:00
Enno Rehling
2c659c1f47 memory leak: free building maintenance data 2015-10-14 12:13:05 +02:00
Enno Rehling
1fc5273f63 memory leak: bad call to register_buildings, replace test_create_world. 2015-10-14 12:09:35 +02:00
Enno Rehling
2b59922d8d memory leak: tests should call cleanup instead of trying to restore original state 2015-10-14 12:05:29 +02:00
Enno Rehling
acda22ae1b leak: free route at the end of the test 2015-10-14 12:03:19 +02:00
Enno Rehling
3cfb22876b memory leak: free params after test 2015-10-14 12:01:33 +02:00
Enno Rehling
20c9fb60dc release additional reference to recruit order 2015-10-14 11:58:54 +02:00
Enno Rehling
f0671aafda leak: some tests were leaking seen_factions becaus they don't call destroyfaction or render reports 2015-10-13 23:33:11 +02:00
Enno Rehling
66dd7e2320 eliminate unused variable warning (gcc) 2015-10-13 23:27:13 +02:00
Enno Rehling
a3210aaed7 leak: free per-locale cached order data (has additional refcount) 2015-10-13 23:25:44 +02:00
Enno Rehling
377b0d0305 leak: teardown build tests, free construction data 2015-10-13 23:09:50 +02:00
Enno Rehling
f4f038c23c leak: free strlist result of split_paragraph 2015-10-13 23:06:33 +02:00
Enno Rehling
24f1f82e93 release order reference used to create castorder, message 2015-10-13 23:02:01 +02:00
Enno Rehling
a6ca489d9a free multi-level construction correctly, avoid duplication 2015-10-13 22:56:16 +02:00
Enno Rehling
26f5b592bd leak: free construction materials 2015-10-13 22:41:54 +02:00
Enno Rehling
457465b4a2 memory leak: free message_list objects 2015-10-13 22:37:45 +02:00
Enno Rehling
333c27403f memory leaks: ship_type.construction, give_cmd return values 2015-10-13 22:34:22 +02:00
Enno Rehling
cd1467f668 cleanup: test_create_world is lazy 2015-10-13 22:31:03 +02:00
Enno Rehling
c70d987c09 memory leak: free weapon data 2015-10-13 22:27:43 +02:00
Enno Rehling
0facb386a8 memory leaks: call test_cleanup, free strlist, free message 2015-10-13 22:21:05 +02:00
Enno Rehling
d264c454b8 memory leak: building_type::construction was never released 2015-10-13 22:13:50 +02:00
Enno Rehling
1f1850dafc memory leak: always cleanup at the end of a test that creates game data. 2015-10-13 22:11:45 +02:00
Enno Rehling
7c2c520911 memory leak: at_travelunit contains a quicklist, needs custom cleanup code. 2015-10-13 22:09:41 +02:00
Enno Rehling
7e9c36b599 memory leak: free spell components and spells at exit 2015-10-13 22:02:58 +02:00
Enno Rehling
67135c8ea3 avoid triggering null-pointer assert 2015-10-13 21:50:23 +02:00
Enno Rehling
2f35246f8c memory leak: when freeing factions, also free their spellbooks 2015-10-13 21:47:45 +02:00
Enno Rehling
ff0bbce4f9 weird things happen when you name locales something else? 2015-10-13 21:30:10 +02:00
Enno Rehling
9f6fa5b948 Merge branch 'feature/tests-leaks' of github.com:ennorehling/eressea into feature/tests-leaks 2015-10-13 21:26:56 +02:00
Enno Rehling
6de8e4466a stop memory leak from function registry. 2015-10-13 21:26:48 +02:00
Enno Rehling
3741a19437 prevent memory leak of cJSON structures in tests 2015-10-13 21:21:31 +02:00
Enno Rehling
422476409c eliminate memory leak from locales 2015-10-13 20:58:03 +02:00
Enno Rehling
5c8a68e1de always call seen_done to eliminate memory leaks in tests. 2015-10-13 16:05:32 +02:00
Enno Rehling
7854684d8b fix memory leaks in all tests that create an order without calling free_order at the end. 2015-10-13 15:45:38 +02:00
Enno Rehling
96be12218e test rename with missing name (not crashing, error message) 2015-10-13 14:27:42 +02:00
Enno Rehling
6de604701d add some test coverage for name_cmd (not covering all error cases yet) 2015-10-13 13:56:58 +02:00
Enno Rehling
177f8f5288 re-enable a test that supposedly breaks on travis (find out why)
fix a potential null-pointer crash in rename_cmd
2015-10-13 13:32:27 +02:00