[Junior_Djjr]

I also checked the link for 'head.cs' you gave me, which is useless as it is because the script tries to load audio files which aren't part of the package:

0AC1: 11@ = load_audiostream_with_3d_support "Audio\sfx\HONDA_LAMP_START.wav" // IF and SET
0AC1: 12@ = load_audiostream_with_3d_support "Audio\sfx\HONDA_LAMP_END.wav" // IF and SET
...
0AAD: set_mp3 11@ perform_action 1

Perhaps that's why it crashes?

omg, i aready installed IVLM at one time in my game and don't see this files (and yes, this .cs is from IVLM, but i don't published in my blog, i don't know what version of ivlm)

thats a problem, cause now the scripts with this "errors" will crash, but cleo 4.1 not, and then still little more hard to know what is problem with cleo and script


now a diferent question and "problem" : cleo 4.3 have more size/compatibility of string?
cause i create THIS mod using cleo 4.3 to test... first function worked for all people, but the second function not...
then i tested in cleo 4.1 and don't worked!

CODE

00000375&1: [0A92] START_NEW_CUSTOM_SCRIPT "EnhFuncs\betatest\Script.BMS" 257 236624160 26 23679630
00000419&1: [D100] NOT COMMAND_5100

seems a bug of compilation (i'm using the latest sanny)
sorry if i didn't see, but, cleo 4.1 and cleo 4.3 have diferents compatibity of strings? more size? don't need "cleo\"? i don't read about this in the change log... but in cleo 4.1 crashed and cleo 4.3 worked @_@
this problem is little old, i don't remember, but i put "cleo\EnhFuncs\slidecover\Script.BMS" and crashed in cleo 4.1 but "cleo\EnhFuncs\slide\Script.BMS" worked .-. ...now almost people sayind about this worked in cleo 4.1, but here don't worked, only in cleo 4.3

have diferences in the versions of cleo?

This post has been edited by Junior_Djjr: Feb 25 2014, 06:55 PM

[Deji]

CLEO 4.3 indeed increases the maximum string size, from 4.2a, at least. 4.2 had some weird logic going on, where it had a buffer for 128 characters (the limit on SCM strings is 127 due to data size and signed conversion to int) yet only requested 100 characters from the script engine. I didn't see any valid reason for this, so I "corrected" it so it used the entire SCM limit. I thought this would have been an error in the 4.2 source, and not particularly effective in 4.1.

So yes, the size is increased, but the code you posted doesn't break either limit and should work anyway.

[Seemann]

There's definitely something weird with Enhanced Functions by BMS.cs. I've tried to decompile it and failed. A hex editor shows me where a possible problem could be present.

02 00 01 FE FF FF FF 00 01 0E FF...

it's an opcode 0002 (jump) which is correct. An opcode 0100 (?) is next to it with the first parameter being a long string with 255 characters (??). Very unlikely it was supposed to be so.
I need to know how the script is looked in original, to find out if there is a SB's bug or not.

CODE

:NONAME_177
{177} 30@ += 256
{185}   30@ > 35584
{195} jf @NONAME_93
{202} jump @NONAME_2
{209} 0100:   actor "Æ

-ûÿÿ" in_sphere 27@ 10@ in_car
{475} jump @NONAME_539

This post has been edited by Seemann: Feb 26 2014, 06:19 AM

[mfisto]

Oups. Problems with saving solved! I found conflicting plugin. This plugin laid out in sannybuilder.com forum - icons_adder.asi from author Vitlgt. Bad and good news...

Oups again. Up info is wrong...

I enclose two scripts, while maintaining they are buggy. Namely, if we pick up pickup and save, at booting turns nonsense. In Stop list is not the same script, and the other did not in the save game list.

This is scripts
This is log

This post has been edited by mfisto: Feb 26 2014, 02:36 PM

[Junior_Djjr]

There's definitely something weird with Enhanced Functions by BMS.cs. I've tried to decompile it and failed. A hex editor shows me where a possible problem could be present.


it's an opcode 0002 (jump) which is correct. An opcode 0100 (?) is next to it with the first parameter being a long string with 255 characters (??). Very unlikely it was supposed to be so.
I need to know how the script is looked in original, to find out if there is a SB's bug or not.

CODE

:NONAME_177
{177} 30@ += 256
{185}   30@ > 35584
{195} jf @NONAME_93
{202} jump @NONAME_2
{209} 0100:   actor "Æ

-ûÿÿ" in_sphere 27@ 10@ in_car
{475} jump @NONAME_539

haha, thats a "block" for script can remove this and open

but, i don't have here the version of the old problem with cleo 4.1 .-.
the new version, in this part is this:

CODE

0AA7: call_return 0x4C5400 2 2 "sawbarl" 10@ -> 31@
        if 31@ > 0
        then
            30@ = true
            0470: 9@ = actor 11@ current_weapon
            if 0AAB:   file_exists "cleo\EnhFuncs\sawbarl\Script.BMS"
            then 0A92: create_custom_thread "EnhFuncs\sawbarl\Script.BMS" 11@ {actorhandle}  31@ {pframe}  9@ {weapon type} 27@ {label offset} 10@ {RwObject}
            else 0ACD: show_text_highpriority "~r~Erro ao carregar~n~Verifique se falta arquivos." time 3000
            end
        end

[Junior_Djjr]

failbit's error again.......

http://pastebin.com/bcW0rLLk

thats rare to appears this error now, appears 2 times with me in last weak (i think more, but some times don't crash, only don't load the script)
in the other time, i think of i was with cleo 4.3.14 (i don't know), then i installed .15 to test, now with .15 give this error again, the gta crashed with this, days ago much mods don't loaded in game, but don't crashed
my friend said this is a error of stream of scripts .-.

edit: and again: http://pastebin.com/bnNrxWeh

This post has been edited by Junior_Djjr: Mar 3 2014, 05:51 PM

[Deji]

The file is simply not opening, which could be for a huge number of reasons. Try disabling your anti-virus and anything else using the file when running the game.

[mfisto]

What about my problem in last post? I need solve my problem. Please

[Deji]

What about my problem in last post? I need solve my problem. Please

I have no idea what the problem is, the 50 page log file is not helpful


EDIT

Okay, wow, another 4.2a source bug:

CODE

WriteBinary(ss, header);
WriteBinary(ss, CleoVariables, 0x400);
std::for_each(savedThreads.begin(), savedThreads.end(),
    [&savedThreads, &ss](CCustomScript *cs)
{
    ThreadSavingInfo savingInfo(cs);
    WriteBinary(ss, savingInfo);
});
std::for_each(stoppedThreadHashes.begin(), stoppedThreadHashes.end(),
    [&ss](unsigned long hash)
{
    WriteBinary(ss, hash);
});

Writes the saved script info first, then the stopped script info...

CODE

ReadBinary(ss, CleoVariables, 0x400);
ReadBinary(ss, stopped_info, safe_header.n_stopped_threads);
ReadBinary(ss, safe_info, safe_header.n_saved_threads);

Some minor inconsistencies there, I think.

Updated in 4.3.16 so they're loaded the same way they're written.

[mfisto]

Nice work, man...Ok for now...

[mfisto]

I am notice that bug with scm_func not repair. That ScrLog ->

CODE

[0AB1] CALL -2480 2 12647 -1
[2F72] COMMAND_2F72

But in script instead of 27f2 -> 00A0

[Deji]

I am notice that bug with scm_func not repair. That ScrLog ->

CODE

[0AB1] CALL -2480 2 12647 -1
[2F72] COMMAND_2F72

But in script instead of 27f2 -> 00A0

It's probably a script error - number of params doesn't match the count or number of returns isn't matched in variables. What exactly are you reporting?

[fastman92]

I just tested a latest CLEO 4.3
My game is a bit modded. CLEO stopped loading of the game at 50%.
Game doesn't crash, the music is playing.

[Junior_Djjr]

i think of i found a solution for failbit error (and others errors) edit: not failbit error, but other big problem
not me, but a friend...

my pc is in portuguese, then the local is here in the print:

he said about a crash with my "Tuning Mod", i said about to use crack 1.0, cleo 4.3 etc, but don't fixed the crash, much others peoples have the same crash and dont fix in anyway, but he said to me about this and worked
(i think of the windows blocked the mod cause i used memalloc, or i could be writing nonsense)

today other guy was with problem in steering(dashboard mod) in cleo 4.3, was in on/off the engine of car without stop and press a key
i thought of the problem is the load of .cfg file with the keys, i said about this soluction, and worked!


btw: thats not failbit error in the load of game ↑ this occurs in ALL time of the mod will make such function, in the case... the tuning mod, when you go in the street, the mod gets a vehicle, alloc a memory and make a copy handling struct and put in the allocated memory, then, crash...
not for me, but some others peoples, the failbit error continues for me, i make this soluction and now i'm waiting for the error again

edit: may be possible of the problem is by a update of windows?
i don't actived my windows update and don't have this problems (only with failbit error)
i don't know if this occurs in cleo 4.1 too, and i don't can test cause i don't have this problems .-.

This post has been edited by Junior_Djjr: Mar 31 2014, 04:10 PM

[Deji]

Well the failbit error is a file access problem. Data execution prevention is to prevent parts of an executable which are not marked as code from being run as such, which could kick in for CLEO memory hooks, however (another reason I'm not keen on hooking memory in CLEO any more) - but I doubt it is what's causing this occasional error. It's a bit of a security measure against malware injecting itself into executables.

Now, even if the problem was to do with some Windows security measure, it wouldn't work even some of the time. Windows doesn't mess around and only force such security measures on occasion. So unless an inefficient anti-virus is causing that problem, it must be something else. The CLEO scripts are opened using the same standard C++ file opening methods as many other programs, but with the exception that this error is handled whereas an older method would at most provide the error "file not found".

I just tested a latest CLEO 4.3
My game is a bit modded. CLEO stopped loading of the game at 50%.
Game doesn't crash, the music is playing.

Got a CLEO 3 only script without the script compatibility mode (.cs3) set?

[Junior_Djjr]

right↑

i get new info about this:
the problem isn't from cleo 4.3/cleo 4.1... IS .ASI TOO, a guy tested to me

thats not occured in any time, anyone have this problem, in the lasts days this problem appeared to much guys in my blog, can explain this? the windows update's updated with a new """better""" security?

[uokka]

Sorry for interrupting - just curious.
That workaround, excluding gta_sa.exe from DEP, is it only needed if DEP is already enabled, or does it still 'fix' those crashes even if DEP was disabled when GTA crashed?

Meaning, is "the way to fix this via DEP" to enable DEP and then exclude gta_sa.exe?

[Junior_Djjr]

Sorry for interrupting - just curious.
That workaround, excluding gta_sa.exe from DEP, is it only needed if DEP is already enabled, or does it still 'fix' those crashes even if DEP was disabled when GTA crashed?

Meaning, is "the way to fix this via DEP" to enable DEP and then exclude gta_sa.exe?

i think the gta_sa.exe entered in the list of "essentials programs of windows" in the first circle in the print... (OR NOT?)
my friend said of he "put the prevention mode", don't said more infos and i don't asked, i will asked now @_@
(anyway, if other people have more infos about this will help)

[uokka]

Apart from the earlier failbit errors that are already fixed, I don't seem to get anymore crashes like this.

I have DEP turned off (My anti-virus is already messing with all opened files and even my firewall tryes to handle potential malware) and UAC is set to not ask the user (me) to allow programs to run.
If I would create evil programs, I would probably hide them inside some kind of entertainment software (games/cracks/plugins). Excluding those from being analized just makes the hole security thing irrelevant.

I'd suggest to try offline. Disconnect from the net and disable security software before playing hacked GTA. If that works, alternate security software might be used to be able to stay connected to the net while running GTA.

I'm personally using Avira anti virus (free). I also emailed them two or three ASI plugins, made by well known GTA modders, which were wrongly identified as trojans. Those files got analized by Avira and added to the white-list (false alert).
On another PC, I'm using Avast. Identifying which files causing problems there was much less intuitive as the error message was just telling nothing (so I had to use my brain, damn!).