17 replies to this topic

[Miloch]

I mentioned this elsewhere, but I guess I should sort it out now since the mod using it is completing final testing.

The mod externalises race string references using Taimon's original code. While we're at it, we enable the added race's paperdoll via Erephine's code (a subset of IA/1PP I presume).

Spoiler

OUTER_PATCH_SAVE orig_bytes_1 ~~ BEGIN
    INSERT_BYTES 0x000 0xc
    WRITE_LONG   0x000 0xfe38bd81
    WRITE_LONG   0x004 0x0098ffff
    WRITE_LONG   0x008 0x5d770000
  END

  OUTER_PATCH_SAVE patch_bytes_1 ~~ BEGIN
    INSERT_BYTES 0x000 0xc
    WRITE_LONG   0x000 0x68cee850
    WRITE_LONG   0x004 0x45890022
    WRITE_LONG   0x008 0x9065eb80
  END

  OUTER_PATCH_SAVE orig_bytes_2 ~~ BEGIN
    INSERT_BYTES 0x000 0x12
    WRITE_LONG   0x000 0xfe4cbd81
    WRITE_LONG   0x004 0x0098ffff
    WRITE_LONG   0x008 0x75770000
    WRITE_LONG   0x00c 0xfe4c958b
    WRITE_SHORT  0x010 0xffff
  END

  OUTER_PATCH_SAVE patch_bytes_2 ~~ BEGIN
    INSERT_BYTES 0x000 0x12
    WRITE_LONG   0x000 0x14a6e852
    WRITE_LONG   0x004 0x85890003
    WRITE_LONG   0x008 0xffffff78
    WRITE_LONG   0x00c 0x90907deb
    WRITE_SHORT  0x010 0x9090
  END

  OUTER_PATCH_SAVE orig_bytes_3 ~~ BEGIN
    INSERT_BYTES 0x000 0xe
    WRITE_LONG   0x000 0x99e47d81
    WRITE_LONG   0x004 0x77000000
    WRITE_LONG   0x008 0xe4558b63
    WRITE_SHORT  0x00c 0xc933
  END

  OUTER_PATCH_SAVE patch_bytes_3 ~~ BEGIN
    INSERT_BYTES 0x000 0xe
    WRITE_LONG   0x000 0x00f1e850
    WRITE_LONG   0x004 0x45890000
    WRITE_LONG   0x008 0x008ae9f0
    WRITE_SHORT  0x00c 0x0000
  END

  OUTER_PATCH_SAVE orig_bytes_4 ~~ BEGIN
    INSERT_BYTES 0x000 0xc2
    WRITE_LONG   0x000 0x006d9000
    WRITE_LONG   0x004 0x006d8fc1
    WRITE_LONG   0x008 0x006d8fca
    WRITE_LONG   0x00c 0x006d8fd3
    WRITE_LONG   0x010 0x006d8fdc
    WRITE_LONG   0x014 0x006d8fee
    WRITE_LONG   0x018 0x006d8fe5
    WRITE_LONG   0x01c 0x006d8ff7
    WRITE_LONG   0x020 0x006d9009
    WRITE_LONG   0x024 0x006d9012
    WRITE_LONG   0x028 0x03020100
    WRITE_LONG   0x02c 0x07060504
    WRITE_LONG   0x030 0x09090909
    WRITE_LONG   0x034 0x09090909
    WRITE_LONG   0x038 0x09090909
    WRITE_LONG   0x03c 0x09090909
    WRITE_LONG   0x040 0x09090909
    WRITE_LONG   0x044 0x09090909
    WRITE_LONG   0x048 0x09090909
    WRITE_LONG   0x04c 0x09090909
    WRITE_LONG   0x050 0x09090909
    WRITE_LONG   0x054 0x09090909
    WRITE_LONG   0x058 0x09090909
    WRITE_LONG   0x05c 0x09090909
    WRITE_LONG   0x060 0x09090909
    WRITE_LONG   0x064 0x09090909
    WRITE_LONG   0x068 0x09090909
    WRITE_LONG   0x06c 0x09090909
    WRITE_LONG   0x070 0x09090909
    WRITE_LONG   0x074 0x09090909
    WRITE_LONG   0x078 0x09090909
    WRITE_LONG   0x07c 0x09090909
    WRITE_LONG   0x080 0x09090909
    WRITE_LONG   0x084 0x09090909
    WRITE_LONG   0x088 0x09090909
    WRITE_LONG   0x08c 0x09090909
    WRITE_LONG   0x090 0x09090909
    WRITE_LONG   0x094 0x09090909
    WRITE_LONG   0x098 0x09090909
    WRITE_LONG   0x09c 0x09090909
    WRITE_LONG   0x0a0 0x09090909
    WRITE_LONG   0x0a4 0x09090909
    WRITE_LONG   0x0a8 0x09090909
    WRITE_LONG   0x0ac 0x09090909
    WRITE_LONG   0x0b0 0x09090909
    WRITE_LONG   0x0b4 0x09090909
    WRITE_LONG   0x0b8 0x09090909
    WRITE_LONG   0x0bc 0x09090909
    WRITE_SHORT  0x0c0 0x0809
  END

  OUTER_PATCH_SAVE patch_bytes_4 ~~ BEGIN
    INSERT_BYTES 0x000 0xc2
    WRITE_LONG   0x000 0x83ec8b55
    WRITE_LONG   0x004 0x4d8d30ec
    WRITE_LONG   0x008 0x5185e8dc
    WRITE_LONG   0x00c 0x4f68fff6
    WRITE_LONG   0x010 0x8d006d91
    WRITE_LONG   0x014 0xcde8dc4d
    WRITE_LONG   0x018 0x8bffd29c
    WRITE_LONG   0x01c 0xc085dc45
    WRITE_LONG   0x020 0x10a16274
    WRITE_LONG   0x024 0x8900b720
    WRITE_LONG   0x028 0x75ffd845
    WRITE_LONG   0x02c 0x1cb86808
    WRITE_LONG   0x030 0x558d00b1
    WRITE_LONG   0x034 0x0de852d8
    WRITE_LONG   0x038 0x8300374a
    WRITE_LONG   0x03c 0x58680cc4
    WRITE_LONG   0x040 0x8d006d91
    WRITE_LONG   0x044 0x53e8d44d
    WRITE_LONG   0x048 0x8d003778
    WRITE_LONG   0x04c 0x8d52d855
    WRITE_LONG   0x050 0x8d50d445
    WRITE_LONG   0x054 0xf7e8dc4d
    WRITE_LONG   0x058 0x8bffd2b6
    WRITE_LONG   0x05c 0xd0558d00
    WRITE_LONG   0x060 0x1cb86852
    WRITE_LONG   0x064 0xe85000b1
    WRITE_LONG   0x068 0x0035fa03
    WRITE_LONG   0x06c 0x8d0cc483
    WRITE_LONG   0x070 0xb9e8d84d
    WRITE_LONG   0x074 0x8d003777
    WRITE_LONG   0x078 0xb1e8d44d
    WRITE_LONG   0x07c 0x8b003777
    WRITE_LONG   0x080 0x05ebd045
    WRITE_LONG   0x084 0xffffffb8
    WRITE_LONG   0x088 0x5de58bff
    WRITE_LONG   0x08c 0x000004c2
    WRITE_LONG   0x090 0x00000000
    WRITE_LONG   0x094 0x00000000
    WRITE_LONG   0x098 0x00000000
    WRITE_LONG   0x09c 0x00000000
    WRITE_LONG   0x0a0 0x00000000
    WRITE_LONG   0x0a4 0x00000000
    WRITE_LONG   0x0a8 0x00000000
    WRITE_LONG   0x0ac 0x00000000
    WRITE_LONG   0x0b0 0x41520000
    WRITE_LONG   0x0b4 0x45544543
    WRITE_LONG   0x0b8 0x53005458
    WRITE_LONG   0x0bc 0x45525254
    WRITE_SHORT  0x0c0 0x0046
  END

  OUTER_PATCH_SAVE orig_bytes_5 ~~ BEGIN
    INSERT_BYTES 0x000 0xf
    WRITE_LONG   0x000 0xff78bd83
    WRITE_LONG   0x004 0x7706ffff
    WRITE_LONG   0x008 0x78958b4c
    WRITE_SHORT  0x00c 0xffff
    WRITE_BYTE   0x00e 0xff
  END

  OUTER_PATCH_SAVE patch_bytes_5 ~~ BEGIN
    INSERT_BYTES 0x000 0xf
    WRITE_LONG   0x000 0x19e85141
    WRITE_LONG   0x004 0x89fffa4d
    WRITE_LONG   0x008 0x72eb8045
    WRITE_SHORT  0x00c 0x9090
    WRITE_BYTE   0x00e 0x90
  END

  OUTER_PATCH_SAVE orig_bytes_6 ~~ BEGIN
    INSERT_BYTES 0x000 0x12
    WRITE_LONG   0x000 0xfcd8bd81
    WRITE_LONG   0x004 0x0098ffff
    WRITE_LONG   0x008 0x75770000
    WRITE_LONG   0x00c 0xfcd8958b
    WRITE_SHORT  0x010 0xffff
  END

  OUTER_PATCH_SAVE patch_bytes_6 ~~ BEGIN
    INSERT_BYTES 0x000 0x12
    WRITE_LONG   0x000 0x4f02e852
    WRITE_LONG   0x004 0x8589ffef
    WRITE_LONG   0x008 0xffffff78
    WRITE_LONG   0x00c 0x90907deb
    WRITE_SHORT  0x010 0x9090
  END

  COPY ~bgmain.exe~ ~bgmain.exe~
    READ_ASCII 0x0b27c9 target_bytes_1 ELSE 0 (0xc)
    READ_ASCII 0x2a7bf1 target_bytes_2 ELSE 0 (0x12)
    READ_ASCII 0x2d8fa6 target_bytes_3 ELSE 0 (0xe)
    READ_ASCII 0x2d909d target_bytes_4 ELSE 0 (0xc2)
    READ_ASCII 0x33437d target_bytes_5 ELSE 0 (0xf)
    READ_ASCII 0x3e4195 target_bytes_6 ELSE 0 (0x12)
    PATCH_IF (~%target_bytes_1%~ STRING_EQUAL ~%orig_bytes_1%~)
      AND (~%target_bytes_2%~ STRING_EQUAL ~%orig_bytes_2%~)
      AND (~%target_bytes_3%~ STRING_EQUAL ~%orig_bytes_3%~)
      AND (~%target_bytes_4%~ STRING_EQUAL ~%orig_bytes_4%~)
      AND (~%target_bytes_5%~ STRING_EQUAL ~%orig_bytes_5%~)
      AND (~%target_bytes_6%~ STRING_EQUAL ~%orig_bytes_6%~)
    THEN BEGIN
      PATCH_PRINT ~Patching .exe to externalize race strrefs ...~
      WRITE_ASCIIE 0x000b27c9 ~%patch_bytes_1%~ (0xc)
      WRITE_ASCIIE 0x002a7bf1 ~%patch_bytes_2%~ (0x12)
      WRITE_ASCIIE 0x002d8fa6 ~%patch_bytes_3%~ (0xe)
      WRITE_ASCIIE 0x002d909d ~%patch_bytes_4%~ (0xc2)
      WRITE_ASCIIE 0x0033437d ~%patch_bytes_5%~ (0xf)
      WRITE_ASCIIE 0x003e4195 ~%patch_bytes_6%~ (0x12)
    END ELSE PATCH_PRINT ~Target bytes not found for .exe race strref patch.~
    READ_BYTE 0x41ae67 b1
    READ_LONG 0x41ae68 b2
    READ_BYTE 0x41ae86 b3
    READ_LONG 0x41ae87 b4
    PATCH_IF (b1 = 0xe9) AND (b2 = 0x1ec) AND (b3 = 0xe9) AND (b4 = 0x1cd) BEGIN
      PATCH_PRINT ~Patching .exe to enable creature paperdolls ...~
      WRITE_BYTE 0x41ae67 0xeb
      WRITE_LONG 0x41ae68 0x90909042
      WRITE_BYTE 0x41ae86 0xeb
      WRITE_LONG 0x41ae87 0x90909023
    END ELSE BEGIN
      PATCH_PRINT ~Target bytes not found for .exe paperdoll patch.~
    END
  BUT_ONLY

Now I assume at least the race part will conflict with TobEx, at least that's what initial testing seems to suggest (via crashing). I guess we could skip it if TobEx's race component is detected, but that doesn't stop someone from tacking on TobEx after this, does it? Or would that still work due to whatever TobEx has in its DLLs? Presumably, the paperdoll patch should still be ok as I don't think TobEx does anything there.

This is our racetext.2da which also gets skipped if it exists. So there shouldn't be any conflicts there... we use lowercase labels for the race (as the game does) but I don't think that makes a real difference (unless the game uses large caps vs. small caps for upper and lowercase - it might use all caps, but a minor detail anyway).

Spoiler

2DA V1.0
-1
    STRREF
1    7193
2    7194
3    7197
4    7182
5    7195
6    7196
7    53186
108  15946
153  8332

Is there a way to code this to avoid conflicts with a subsequent TobEx install, or is the best way just to enforce an install order (TobEx goes first)?

[Ascension64]

Well, if we still are going with this 'standard' idea, TobEx will not check for compatibility, so someone installing TobEx after Infinity Animations is going to suffer.

If you plan to keep this code, then suggest putting TobEx before Infinity Animations.
Otherwise, I can pop the paperdoll code into TobEx and you could just embed TobEx (without suggesting that TobEx is taking over the world...)

[the bigg]

If you plan to keep this code, then suggest putting TobEx before Infinity Animations.
Otherwise, I can pop the paperdoll code into TobEx and you could just embed TobEx (without suggesting that TobEx is taking over the world...)

I think that, long-term, ToBEx should absorb the other EXE patches, to save you from having to juggle compatibility with stand-alone EXE patches.

[cmorgan]

I suggest ToBEX take over the world.

Seriously.

At least all the .exe changes that can be folded in - it would make things much more stable, and much easier to diagnose problems. Updating older mods would be relatively simple.

[Miloch]

Well, this is actually used in a different mod than IA that just wants to enable the display text for a race and its paperdoll. Some slots have paperdolls enabled automatically in the engine; others do not. I suppose that at least is something fairly basic you could TobEx-ise, but it would probably require a bunch of mods to account for the possibility of TobEx being installed (IA and several 1PP components for starters). Then again, it sounds like cmorgan is offering to handle that .

Compatibility seems a bit dodgy... ordinarily, we would check if the .exe has already been patched for this feature and skip it if so. But TobEx doesn't patch the .exe. So I put this check in to skip the strref stuff if it's enabled in TobEx. Would that suffice? Of course, someone can change that .ini at any time. I don't know how to account for that, except put it in the readme under "don't do it" or something.

COPY ~bgmain.exe~ ~bgmain.exe~
    PATCH_IF FILE_EXISTS ~TobEx.ini~ AND FILE_CONTAINS ~TobEx.ini~ ~Externalise Race Selection StrRef=1~ BEGIN
      PATCH_PRINT ~Skipping race strref patching (TobEx already does this).~
    END ELSE BEGIN
      ... (stuff to enable racial strref)
    END
    ... (stuff to enable racial paperdoll)
  BUT_ONLY

As for externalising all of IA's .exe patch, in theory it sounds great but I suspect it would be a lot of work, and then there's the issue of compatibility and maintenance. I doubt it's something you want to maintain every time we change the patch (as we are currently working on). At its basic level though, it's more or less an "Externalise animation references to .ids files" (animate and anisnd... they exist in the game but mainly as static reference rather than dynamic tables - IA changes this).

Edit: incidentally, Taimon's hack reverses the gnome and dwarf races. So if you create a gnome fighter/illusionist, everything looks good in character creation but your character record says "dwarf" once you're in the game, and vice versa. I seem to recall you fixed this when you put in your version, could you point out where in the code? I don't see a technical spec for this particular hack.

Edited by Miloch, 04 February 2011 - 12:58 PM.

[Ascension64]

Compatibility seems a bit dodgy... ordinarily, we would check if the .exe has already been patched for this feature and skip it if so. But TobEx doesn't patch the .exe. So I put this check in to skip the strref stuff if it's enabled in TobEx. Would that suffice? Of course, someone can change that .ini at any time. I don't know how to account for that, except put it in the readme under "don't do it" or something.

Well, from version beta 0018 onwards, mods will need to assume that Externalise Race Selection StrRef=1. People who tamper with TobExCore.ini (unless they have been told to or know what they are doing) will get themselves kicked up the backside.

I have suggested in TobEx_redist that mods check for TobEx install by FILE_EXISTS ~TobEx_ini/TobExVer.txt~.

Edit: incidentally, Taimon's hack reverses the gnome and dwarf races. So if you create a gnome fighter/illusionist, everything looks good in character creation but your character record says "dwarf" once you're in the game, and vice versa. I seem to recall you fixed this when you put in your version, could you point out where in the code? I don't see a technical spec for this particular hack.

Hacky fix. Basically, everywhere there was switch (nRace), I substituted in my own procedure below.

STRREF __stdcall CInfGame_GetRaceText(unsigned int nRace) {
	int Eip;
	GetEip(Eip);

	if (Eip == 0x734392) { //this is where the races are switched around because of the buttons being in the wrong order
		if (nRace == 5) {
			nRace = 3;
		} else {
			if (nRace == 3) {
				nRace = 5;
			}
		}
	}

	if (pRuleEx->m_RaceText.m_2da.bLoaded) {
		IECString sStrRef;

		int nCol = 0;
		int nRow = nRace;
		if (nCol < pRuleEx->m_RaceText.nCols &&
			nRow < pRuleEx->m_RaceText.nRows &&
			nCol >= 0 &&
			nRow >= 0) {
			sStrRef = *((pRuleEx->m_RaceText.pDataArray) + (pRuleEx->m_RaceText.nCols * nRow + nCol));
		} else {
			sStrRef = pRuleEx->m_RaceText.defaultVal;
		}

		STRREF strref;
		sscanf_s((LPCTSTR)sStrRef, "%d", &strref);
		return strref;
	} else {
		STRREF strref;
		switch (nRace) {
		case 0:
			strref = 0x1C19; //human
			break;
		case 1:
			strref = 0x1C1A; //elf
			break;
		case 2:
			strref = 0x1C1D; //half-elf
			break;
		case 3:
			strref = 0x1C0E; //dwarf
			break;
		case 4:
			strref = 0x1C1B; //halfling
			break;
		case 5:
			strref = 0x1C1C; //gnome
			break;
		case 6:
			strref = 0xCFC2; //half-orc
			break;
		case 152:
			strref = 0x208C; //tiefling
			break;
		default:
			strref = -1;
			break;
		}

		return strref;
	}

}

[Miloch]

I have suggested in TobEx_redist that mods check for TobEx install by FILE_EXISTS ~TobEx_ini/TobExVer.txt~.

Eh. So um... when is that coming out? Because that looks a lot different than what's in b017 or whatever. Since the mod with the code above is coming out "presently" (when we get the last few glitches ironed out, this being the main if not only one remaining) it seems kind of like juggling to be compatible, but the BWP will sure rant and rave if we aren't.

if (nRace == 5) {
			nRace = 3;
		} else {
			if (nRace == 3) {
				nRace = 5;
			}

Hmm... I see (I think). Any way you could (re-reverse) the pseudocode into a revised .exe patch? I'm tempted to just require (or suggest at least) TobEx and phase out our patch in light of the potential difficulties, but I don't like requiring other mods if an alternative is feasible.

[Ascension64]

Well, if BWP still uses TobEx, then it will explode...

Hmm... I see (I think). Any way you could (re-reverse) the pseudocode into a revised .exe patch? I'm tempted to just require (or suggest at least) TobEx and phase out our patch in light of the potential difficulties, but I don't like requiring other mods if an alternative is feasible.

This would be annoying, because:
1. I'd have to re-look at Taimon's original code
2. TobEx's current implementation cannot be put into a tob_hacks format.
I can have a look, but it's rather divergent for me.

[DavidWallace]

Well, if BWP still uses TobEx, then it will explode...

Hmm... I see (I think). Any way you could (re-reverse) the pseudocode into a revised .exe patch? I'm tempted to just require (or suggest at least) TobEx and phase out our patch in light of the potential difficulties, but I don't like requiring other mods if an alternative is feasible.

This would be annoying, because:
1. I'd have to re-look at Taimon's original code
2. TobEx's current implementation cannot be put into a tob_hacks format.
I can have a look, but it's rather divergent for me.

Miloch, the idea for the next version of ToBEx is that you just ship it with your mod and auto-install it - so in practice, you're not requiring other mods. (I had parallel worries with SCS, which is what led to the ship-ToBEx proposal.)

[the bigg]

If you're absorbing other exe patches, tb#tweaks contains this one:

Spoiler

OUTER_PATCH_SAVE orig_bytes_1 ~~ BEGIN
	INSERT_BYTES 0x000 0x8
	WRITE_LONG   0x000 0x0d8a016a
	WRITE_LONG   0x004 0x00aa5d6f
END

OUTER_PATCH_SAVE patch_bytes_1 ~~ BEGIN
	INSERT_BYTES 0x000 0x8
	WRITE_LONG   0x000 0x000125e9
	WRITE_LONG   0x004 0x90909000
END

OUTER_PATCH_SAVE orig_bytes_2 ~~ BEGIN
	INSERT_BYTES 0x000 0x5
	WRITE_LONG   0x000 0xba5ee2e8
	WRITE_BYTE   0x004 0xff
END

OUTER_PATCH_SAVE patch_bytes_2 ~~ BEGIN
	INSERT_BYTES 0x000 0x5
	WRITE_LONG   0x000 0xba577fe8
	WRITE_BYTE   0x004 0xff
END

OUTER_PATCH_SAVE orig_bytes_3 ~~ BEGIN
	INSERT_BYTES 0x000 0x6
	WRITE_LONG   0x000 0x0152840f
	WRITE_SHORT  0x004 0x0000
END

OUTER_PATCH_SAVE patch_bytes_3 ~~ BEGIN
	INSERT_BYTES 0x000 0x6
	WRITE_LONG   0x000 0xff0c840f
	WRITE_SHORT  0x004 0xffff
END

COPY bgmain.exe bgmain.exe
	READ_ASCII 0x004ce8af target_bytes_1 ELSE 0 (0x8)
	READ_ASCII 0x004ce91e target_bytes_2 ELSE 0 (0x5)
	READ_ASCII 0x004ce9f1 target_bytes_3 ELSE 0 (0x6)
	PATCH_IF (~%target_bytes_1%~ STRING_EQUAL ~%orig_bytes_1%~)
		 AND (~%target_bytes_2%~ STRING_EQUAL ~%orig_bytes_2%~)
		 AND (~%target_bytes_3%~ STRING_EQUAL ~%orig_bytes_3%~)
	THEN BEGIN
		WRITE_ASCIIE 0x004ce8af ~%patch_bytes_1%~ (0x8)
		WRITE_ASCIIE 0x004ce91e ~%patch_bytes_2%~ (0x5)
		WRITE_ASCIIE 0x004ce9f1 ~%patch_bytes_3%~ (0x6)
	END ELSE
	PATCH_IF (~%target_bytes_1%~ STRING_EQUAL ~%patch_bytes_1%~)
		 AND (~%target_bytes_2%~ STRING_EQUAL ~%patch_bytes_2%~)
		 AND (~%target_bytes_3%~ STRING_EQUAL ~%patch_bytes_3%~)
	THEN BEGIN
		PATCH_PRINT ~Patch already applied. Skipping ...~
	END ELSE INNER_ACTION BEGIN FAIL ~Target bytes don't match. Aborting ...~ END

it makes all creatures get the WSPATCK attack bonuses, rather than only Warriors (according to Taimon, it replaces a call to calculate the Warrior level with one that calculates the generic character level, while preserving current behavior for Monks); as such, it's clearly a Tweak.

[Ascension64]

Well, this is actually used in a different mod than IA that just wants to enable the display text for a race and its paperdoll. Some slots have paperdolls enabled automatically in the engine; others do not. I suppose that at least is something fairly basic you could TobEx-ise, but it would probably require a bunch of mods to account for the possibility of TobEx being installed (IA and several 1PP components for starters). Then again, it sounds like cmorgan is offering to handle that .

Can you tell me what the name of the mod is that has the patch to display race text and paperdoll. Needs to go in compatibility information for tobex_ini.txt.

Anyway, I have implemented similar in TobEx:

-----980 Enable More Paperdolls [C, M]
(identical to, but also extends from a 1PP mod)
Allows paperdolls to show in inventory and character customisation for animations 0x12XX, 0x2XXX, 0x3XXX, and non-vanilla 0x7XXX (except 0x73XX, which is already shown in vanilla)
Note that paperdolls for the 0x7XXX group of animation must have a be named <prefix>0INV.BAM

Options:
-0: disabled
-1: enabled

Additionally, I have allowed paperdolls for dragon animations (0x12**), the 0x2*** series (Sirine, Volo, Ogre Mage, Death Knight), and the 0x3*** series (not really used, unless people decide to make ankheg-like animations).
I hope you have been made aware by Taimon that the paperdolls for the 0x7*** series animations must have a 0INV suffix (note the 0) because of the way the hack is applied. I don't intend to change this for compatibility reasons.

As for externalising all of IA's .exe patch, in theory it sounds great but I suspect it would be a lot of work, and then there's the issue of compatibility and maintenance. I doubt it's something you want to maintain every time we change the patch (as we are currently working on). At its basic level though, it's more or less an "Externalise animation references to .ids files" (animate and anisnd... they exist in the game but mainly as static reference rather than dynamic tables - IA changes this).

Not really. Infinity Animations actually messes up the system by deferring ids in one group of animations (say 0x5***) to another group (say 0xE***). This makes it very difficult for me to re-implement because the assembly code is rather messed up.

[ScuD]

Just a question. Are you planning to add Erephine's palette extension EXE hack?

[Shaitan]

Well, this is actually used in a different mod than IA that just wants to enable the display text for a race and its paperdoll. Some slots have paperdolls enabled automatically in the engine; others do not. I suppose that at least is something fairly basic you could TobEx-ise, but it would probably require a bunch of mods to account for the possibility of TobEx being installed (IA and several 1PP components for starters). Then again, it sounds like cmorgan is offering to handle that .

Can you tell me what the name of the mod is that has the patch to display race text and paperdoll. Needs to go in compatibility information for tobex_ini.txt.

It's FFT (Fishing for Troubles) over here

Cheers

[Ascension64]

Just a question. Are you planning to add Erephine's palette extension EXE hack?

I don't know much about it. Will take a peek.

Edited by Ascension64, 07 February 2011 - 04:33 AM.

[Miloch]

Not really. Infinity Animations actually messes up the system by deferring ids in one group of animations (say 0x5***) to another group (say 0xE***).

I don't know exactly why we did that if there's a virtually unlimited extension to the slots available, but it probably sounded like a good idea at the time.

The palette extender makes more colours available for palettable animations. Might need to extend randcolr.2da also - I forget whether there was a limit to that like there is to the other random tables.

It might not be a bad idea to consolidate all the 1PP exe patches, which probably wouldn't be nearly as difficult as IA. But there are currently some install order glitches with the patches that could be sorted out/avoided if done early on (but again, I suppose the mods would have to be updated).

What would the code be to detect the paperdoll patch and skip it if it's already in? I'm a bit uncertain how to handle this what with all the "beta" stuff you've got going on, but I've asked for access for you in the mod's dev forum, if you want to take a look at it (or I could PM you the tp2 at least).

[Ascension64]

There's a lot of unused code in the constructor for CAnimation5000 (and CAnimation7000), so I suppose that just took the cake.

The palette extender makes more colours available for palettable animations. Might need to extend randcolr.2da also - I forget whether there was a limit to that like there is to the other random tables.

It might not be a bad idea to consolidate all the 1PP exe patches, which probably wouldn't be nearly as difficult as IA. But there are currently some install order glitches with the patches that could be sorted out/avoided if done early on (but again, I suppose the mods would have to be updated).

What is the issue, exactly?

What would the code be to detect the paperdoll patch and skip it if it's already in? I'm a bit uncertain how to handle this what with all the "beta" stuff you've got going on, but I've asked for access for you in the mod's dev forum, if you want to take a look at it (or I could PM you the tp2 at least).

Yeah, again it would be ACTION_IF FILE_EXISTS "TobEx_ini/TobExVer.txt". I don't plan to change anything else from the current test release (0.18.2.8 ) apart from the bug fix to Learn Spell Mod and the WEIGHTALLOWANCEMOD stat, so feel free to test with that. Release planned at the end of this week, pending further bug reports.

Edited by Ascension64, 08 February 2011 - 10:45 PM.

[Miloch]

What is the issue, exactly?

Here is a brief description. Basically, several mods include the same patching code (or similar bits of it) but they aren't all smart enough to account for the possibility of other mods already having done the relevant patching, so they'll fail to install (they assume to find only the bytes of an unmodded ToB .exe before patching).

At a larger level, what you've done might even cause an issue for mods that (currently) do check for the possibility of the .exe being patched already. For example, IA and Aurora include the paperdoll snippet already, as it is embedded in the .exe patch. If you've "externalised" it via DLL or whatever, won't that conflict with mods that are already patching the .exe to do it?

[Ascension64]

No, it won't. the virtual image of BGMain.exe is patched at run-time remember, so even if you install TobEx, all other mods won't even detect it.

With these other mods, instead of using FAIL, you could just PRINT or PATCH_PRINT instead and continue, although the latter does come with caveats if BGMain.exe got modded weirdly.

Edited by Ascension64, 09 February 2011 - 04:03 PM.