DDraceNetwork - developer

DDraceNetwork

Development / developer

Development discussion. Logged to https://ddnet.tw/irclogs/ Connected with DDNet's IRC channel, Matrix room and GitHub repositories — IRC: #ddnet on Quakenet | Matrix: #ddnet-developer:matrix.org GitHub: https://github.com/ddnet

Between 2020-03-24 00:00:00Z and 2020-03-25 00:00:00Z

@heinrich5991 yes its fixed, good job

@heinrich5991 what was it?

dunno, crash was in antibot, disabled it for now

01:42

probably my "fix"

I have a feeling it has something to do with the fact that we "shutdown" the gameserver when changing maps

[ddnet-maps:master] 1 new commit

974b002 A Sky High - ddnet-maps

@heinrich5991 can u either make it use open() instead of fopen()? or entirely remove the code that tries to output to file if ur just discarding it anyway?

@noby it'd just be masking the actual issue with the glue code

what is the actual issue?

well the issue isn't the fopen anymore anyway

what is it

I think it's what I said above but I'm not the one that debugged it so idk

was this shutdown a recent change

12:22

cus my server is based on ddnet 10.8 and i dont see this problem

the problem is probably with the glue code and you aren't running the glue code

ooo

[ddnet/ddnet] Pull request opened: #2096 Fix the double-free of the antibot

I could not find an easy way to ensure that AntibotDestroy is only being called once for each AntibotInit but still happening after all the CPlayer destructors.

[ddnet:master] 2 new commits

6efcc47 Fix cl_showrecord with dummy (fixes #2077) - def- 1029926 Merge #2078 - bors[bot]

14:29

[ddnet/ddnet] Pull request closed: #2078 Fix cl_showrecord with dummy (fixes #2077)

[ddnet/ddnet] Pull request closed: #2096 Fix the double-free of the antibot

hey im trying to make sense of the map file format. while the datafile part of the file format is documented, i didnt find any documentation on how to make sense of the data. is anyone here familiar with the variables used in the different item types?

17:26

i am making progress, but not on the pace i was hoping for. the main problem being that im not used to directing through the ddnet source (edited)

is it specific ones?

17:27

if so, could you post them?

currently im trying to make sense of the Layer item type

17:28

source:

17:28

struct CMapItemLayer
{
    int m_Version;
    int m_Type; 
    int m_Flags;
} ;

struct CMapItemLayerTilemap
{
    CMapItemLayer m_Layer;
    int m_Version;
    
    int m_Width; 
    int m_Height;
    int m_Flags;
    
    CColor m_Color;
    int m_ColorEnv;
    int m_ColorEnvOffset;
    
    int m_Image;
    int m_Data;
    
    int m_aName[3];
    
    // DDRace
    
    int m_Tele;   
    int m_Speedup;
    int m_Front; 
    int m_Switch;
    int m_Tune;
} ;

struct CMapItemLayerQuads
{
    CMapItemLayer m_Layer;
    int m_Version;

    int m_NumQuads;
    int m_Data;
    int m_Image;

    int m_aName[3];
} ;

most fun one

17:28

rip irc

consisting of those structs

17:29

im not used to c++ so its hard for me to figure out how those are used

17:30

interesting for me would be what the different variables mean, for example the content of the m_Flagss

17:30

or the ints m_Tele to m_Tune

17:31

or the m_aName ints, tho these are probably the easiest ones for me to figure out

m_Tele, m_Tune point to the data items that have the {tele,tune} tiles

17:33

then there's TILESLAYERFLAG_* for the tile layer's m_Flag, and LAYERFLAG_DETAIL for the layer's m_Flag

ah nice

m_aName weirdly packs a null-terminated string

one?

17:34

thats odd

there is also names made up of ints somewhere

This is one part of the code that I couldn't make any sense of for the longest time, some of the decisions made no sense to me at all

you tell me, i wrote all this https://github.com/Ryozuki/ddnet-map-lib/blob/master/src/parser.ts just to surrender later cuz i got rly annoyed

nice rq commit message

2

@fokkonaut it seems that you won't lose you ninja when going through shield anymore: https://github.com/ddnet/ddnet/pull/2086/files#diff-80a3da4e88c6e4fb0e9298f92ddbc809R71

Remove setting ammo for freeze to 0 by fokkonaut · Pull Request #20...

did you try?

19:52

for me it worked when i tried in my mod

wait

did not try

does this mean you didnt lose ninja before when you are frozen?

I don't know

i guess

ah no

20:12

pChr->SetNinjaActivationDir(vec2(0,0));
                    pChr->SetNinjaActivationTick(-500);
                    pChr->SetNinjaCurrentMoveTime(0);

20:12

ninja gets set there, should be removed and replaced by RemoveNinja()

20:12

ah no

20:12

removeninja gets called when this is 0

20:13

yea no works like this

but there is another bug with freezetime and shield, gonna search for that

ok fixed everything, you were right :)

20:44

but there is another bug with freezetime and shield, gonna search for that

this was a wrong assumption

[ddnet/ddnet7] Issue opened: #86 specvoted and pausevoted dont work

Exported 62 message(s)