… | |
… | |
2 | |
2 | |
3 | TODO: picking up the torches in nimbus seemingly works, but after |
3 | TODO: picking up the torches in nimbus seemingly works, but after |
4 | a while (used up?) they seem to be gone from the inv but the client doesn't know about that. |
4 | a while (used up?) they seem to be gone from the inv but the client doesn't know about that. |
5 | TODO: dragons should not be able to change their title via the settings tab in playerbook |
5 | TODO: dragons should not be able to change their title via the settings tab in playerbook |
6 | |
6 | |
|
|
7 | - the two handed weapon skill cna no longe rbe used as unarmed combat |
|
|
8 | skill. |
7 | - append [drop xxx coins] to altar/trigger_altar/identify_table names |
9 | - append [drop xxx coins] to altar/trigger_altar/identify_table/converter |
8 | (also update all maps to use money instead of coins). |
10 | names (also update all maps to use money instead of coins). |
9 | - add uuid accessors to perl. |
11 | - add uuid accessors to perl. |
10 | - add coin names and archetype accessors to perl. |
12 | - add coin names and archetype accessors to perl. |
11 | - fix runtime uuid file format to use "." instead of ",". |
13 | - fix runtime uuid file format to use "." instead of ",". |
12 | - change perl extensions to not use hardcoded coins and values. |
14 | - change perl extensions to not use hardcoded coins and values. |
13 | - write uuids asynchronously (no hiccups due to uuid writes) |
15 | - write uuids asynchronously (no hiccups due to uuid writes) |
… | |
… | |
19 | - major cleanup of memory management in random map generator. |
21 | - major cleanup of memory management in random map generator. |
20 | - solve the "slice_alloc becomes negative" mystery - no allocation bug, |
22 | - solve the "slice_alloc becomes negative" mystery - no allocation bug, |
21 | bookkeeping bug. |
23 | bookkeeping bug. |
22 | - garbage collect shstr's more evenly and adaptively only when |
24 | - garbage collect shstr's more evenly and adaptively only when |
23 | there have in fact been allocations. |
25 | there have in fact been allocations. |
|
|
26 | - fix object::insert to do as documented and insert into the env |
|
|
27 | if necessary. |
|
|
28 | - fix cf::map::unique_maps. |
|
|
29 | - fix nrof overflow check in can_merge_slow. |
|
|
30 | - fix converter code to handle >31 bit values. |
|
|
31 | - added support for the player_sold flag |
|
|
32 | - converters no longer accept unpaid items. |
24 | |
33 | |
25 | 2.43 Sun Apr 6 20:35:29 CEST 2008 |
34 | 2.43 Sun Apr 6 20:35:29 CEST 2008 |
26 | - work around a glibc bug that causes calloc to return non-zero memory |
35 | - work around a glibc bug that causes calloc to return non-zero memory |
27 | when mlockall has been used. |
36 | when mlockall has been used. |
28 | - pass environment to perl_sys_init3, as required by newer perls. |
37 | - pass environment to perl_sys_init3, as required by newer perls. |
29 | - use separate thread for ticker timing (and more in the future). |
38 | - use separate thread for ticker timing (and more in the future). |
30 | this greatly increases timing accuracy and gets rid of an enourmous |
39 | this greatly increases timing accuracy and gets rid of an enourmous |
31 | number of gettimeofday calls. |
40 | number of gettimeofday calls. |
32 | - use a seperate thread to signal aio completion, for lower latency. |
41 | - use a separate thread to signal aio completion, for lower latency. |
33 | - require Coro 4.47, which fixes a data corruption bug with its C API. |
42 | - require Coro 4.47, which fixes a data corruption bug with its C API. |
34 | |
43 | |
35 | 2.42 Sat Mar 15 12:05:21 CET 2008 |
44 | 2.42 Sat Mar 15 12:05:21 CET 2008 |
36 | - rebalanced default experience table to be more exponential |
45 | - rebalanced default experience table to be more exponential |
37 | and much easier up to level ~40, then moderately harder. |
46 | and much easier up to level ~40, then moderately harder. |