Files

T

CydandClaude Fable 5 5b35eb973c 4.10 reconstruction: BTL4OPT.EXE links clean and BOOTS to the first staged brick

The reconstructed tree now produces a runnable binary with the authentic
1995 toolchain (BC4.52 / tlink32 / DPMI32):

- BTL4.CPP main TU reconstructed from the 4.11 Ghidra decomp (FUN_0040109c)
  + the 4.10 binary's own string pool + surviving RPL4TOOL.CPP house style;
  probe_main.cpp scaffold retired (BTL4.NOTES.md documents every decoded call)
- L4NET.CPP staged: L4NetworkManager ctor/dtor + 10 vtable-pulled virtuals
  (standalone-benign ones no-op, network ones Fail loudly) + NetNub client
  globals (Net_Common_Ptr=NULL routes L4File to its plain-DOS path)
- build410.sh: libs now built in the AUTHENTIC makefile member order
  (MUNGA.MAK / mungal4.mak / BT.MAK / BTL4.MAK). Order is load-bearing:
  tlink emits static-init records in module pull order, and alphabetical
  order booted into a null-vptr crash (IcomManager::ClassDerivations
  constructing before parent NetworkClient::ClassDerivations). Also fixed
  stage_link to the proven 32-bit lib set (SOSDBXC+SOSMBXC, no WATTCPLG)
- BOXTREE.HPP MemoryBlock unify, BTL4GRND notify stubs, remaining engine
  backfills (audio/gauge/resource/stream TUs) that closed the deep ledger

Smoke test (DOSBox-X + 32RTM, copy of the pod BT tree, our exe swapped in):
  BattleTech v4.10
  BTL4Application::BTL4Application
  l4net.cpp(22): L4NetworkManager -- l4net.cpp not yet reconstructed
Static init, main, -egg parse, BTL4.RES load (version 1.0.6 check passes),
ApplicationManager and the BTL4Application ctor chain all execute real
reconstructed code; boot halts at the first staged Fail() as designed.
Next brick: the real l4net.cpp body.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

2026-07-19 19:05:53 -05:00

3.7 KiB

Raw Blame History

BTL4.CPP — reconstruction notes

Status: RECONSTRUCTED (normal launch path); review/camera branch staged with Fail().

The 4.10 main TU. No original survives anywhere in the archive (OPT.MAK links btl4.obj as a bare object next to the libs — it was never a lib member).

Evidence base

Ghidra decomp of the 4.11 binary — C:\VWE\BT411\reference\decomp\all\part_000.c, FUN_0040109c (the 524-byte function the Borland startup calls = main). The 4.10 and 4.11 binaries share the same string pool layout for this TU.
String pool of the original 4.10 BTL4OPT.EXE (ALPHA_1/REL410/BT), file offset ~917884: "btl4" "HEAPSIZE" "BattleTech v4.10" "btl4.res" "SPOOLSIZE" "Spool size = " — banner text and resource name byte-exact.
Surviving 1995 launcher-style TU CODE/RP/RP_L4/RPL4TOOL.CPP — house style for main(int, char*[]) and the ProgName global pattern.
BT411's win32 launcher game/btl4main.cpp (adapted from RP's RPL4.CPP) independently reconstructs the same skeleton and pins version[3] = {1,0,6} against the RESOURCE.cpp version check.

Decoded decomp → source mapping

decomp	source
`FUN_004dbb24(&cout, "BattleTech v4.10", 0); FUN_004d9c38(&cout)`	`DEBUG_STREAM << "BattleTech v4.10" << endl`
`FUN_0047b2ec(argc, argv, &LAB_0047af8c)`	`L4Application::ParseCommandLine(argc, argv, L4Application::ParseToken)`
`FUN_00407218(stack84, "btl4.res", &DAT_004e0070, 3)`	`StreamableResourceFile resources("btl4.res", version)` (match_level=3 is the declared default)
`DAT_004e0070` = 3 bytes	`version[3]` = `{MAJOR_DATA_VERSION, RELEASE_VERSION, MINOR_DATA_VERSION}` = {1,0,6} from BTL4VER.HPP
`DAT_004fd550 < 1`	`L4Application::GetMissionReviewMode() < 1` (inline static accessor)
`new(0x28) FUN_0044f29c(this, DAT_0052140c)`	`new ApplicationManager(GetTicksPerSecond())` (inline friend accessor, TIME.HPP:49) — DAT_0052140c is set to 28.0f by `SystemClock::SystemClock()` (L4TIME.CPP:93) when `L4TIMER=FAST`
`new(0xd4) FUN_004d34c4(this, &resources, 0x3EB, &DAT_0051e8d8)`	`new BTL4Application(&resources)` — 0x3EB/`DefaultData` are the declared default args (BTL4APP.HPP:34)
`FUN_0044f318(mgr, app)`	`application_manager->StartApplication(application)`
`FUN_0044f344(mgr)`	`application_manager->RunMissions()`
vcall slot 0 (mgr, 3)	`delete application_manager`
`return DAT_004efc98`	`return Exit_Code` (APP.HPP:542)

Review/camera branch (NOT yet reconstructed — staged Fail)

Decomp structure for missionReviewMode >= 1 (mode 1 = camera, 2 = review per emulator/CAMERA-REVIEW-NOTES.md):

spool_size = getenv("SPOOLSIZE") ? atoi(...) : 0x800000;   // "Spool size = " printed
mgr  = new(0x38) MissionReviewApplicationManager(ticksPerSecond,
                    (mode != 2) ? 2 : 1, spool_size);
spool_file = <mgr getter, FUN_0044fa14>;
if (spool_file)
    mgr->StartApplication(new(0xd8) BTL4PlaybackApplication(&resources, spool_file));
if (mode == 1)
    mgr->StartApplication(new(0xd8) <camera app, FUN_0047c3fc>(&resources, 1));

Class declarations survive: BTL4PlaybackApplication (CODE/BT/BT_L4/BTL4PB.HPP), spooler bricks in CODE/RP/MUNGA_L4/L4SPLR.HPP (MissionReviewApplicationManager referenced there). Their .CPP bodies are not staged, so the branch Fails loudly instead of linking half a subsystem.

Deliberate differences from the binary

HEAPSIZE string sits in the pool between ProgName and the banner but is not referenced from the decompiled main — it likely belongs to the runtime heap init, not this TU. Not reproduced.
Register_Object/Start_Registering debug bookkeeping is absent from the optimized binary (DEBUG_LEVEL=0 compiles it out), so it is not written here.

3.7 KiB Raw Blame History