Skip to content

Fix three files that stopped compiling after the eef6f667a header sync - #368

Merged
gynt merged 2 commits into
sourcehold:mainfrom
kotijekriv:fix/sync-eef6f667a-renames
Nov 7, 2026
Merged

gynt merged 2 commits into
sourcehold:mainfrom
kotijekriv:fix/sync-eef6f667a-renames

Conversation

@kotijekriv

Copy link
Copy Markdown
Contributor

Main (eef6f66) has three .cpp files that no longer compile in a clean build: the Ghidra sync eef6f66 renamed DAT_00df5564 to INT_00df5564 (and declared it int) and INT_00ed279c to BOOLEnum_00ed279c, and these files still include and name the old ones. This PR moves them to main's names and changes nothing else: no header and no status line. The three functions stay at 100%, the same as their status lines. They're the only uses of the old names left on main.

File Was Main's name now
UI/Helpers/SetTutorialHintActiveWithTimestamp.cpp DAT_00df5564 INT_00df5564
UI/Helpers/UpdateTutorialStepAndProgress.cpp DAT_00df5564 INT_00df5564
UI/MenuItems/MissionEndscreen/MenuItemActionHandler_MissionEndscreen_Main.cpp INT_00ed279c = 1 BOOLEnum_00ed279c = TRUE

The include lines follow the names, sorted by clang-format. SetTutorialHintActiveWithTimestamp's first include is now written "../Helpers.func.hpp", as the style guide's rule 14 asks; it's the same header.

Built with the bundled MSVC1400-SP1, natively on Windows, against the Steam Stronghold Crusader.exe (3bb0a8c1…5387ac5a). Every .cpp under src/OpenSHC compiled (all 1,809), and only three others still fail, each fixed by an open PR whose header changes the sync took in:

With those three taken from their PRs for the measurement only, the DLL links, and all three functions here read 100.0% with sourcehold's reccmp (--resolve-wrapped-calls --ignore-security-cookie, the struct resolvers enabled locally and not committed). status_regression.py finds nothing below a known score.

The same kind of fix as #343 and #350.

Something for the next Ghidra sync

A full link of main also stops on two entries in src/symbols/data.asm (they're generated, so this is yours):

  • DAT_PerformanceCounterFrequency (0x00DF33E8) is declared as U_LARGE_INTEGER, a struct, but _LARGE_INTEGER is a union in the Windows headers, so the compiler asks for T_LARGE_INTEGER (TicksStartCounter.cpp, TicksSinceCounterStart.cpp);
  • DPLAY_InterfacePointer (0x00DF3D18) is declared in a dplay namespace (PAUIDirectPlay4@dplay@@), but IDirectPlay4 from dplay.h is in the global one (initializeDirectPlayAndCreateOrJoinSession.cpp).

Both have been that way since the syncs of 2026-11-04 and 11-05. For the measurement I left those three files out of the build.

Other PRs: the ready #348, #349, #352, #353, #354, #357, #358, #359 and #364, and the drafts #313, #318, #321, #327, #328, #356, #360 and #363. Several of them will get main merged in for the same sync.

🤖 Generated with Claude Code

kotijekriv and others added 2 commits November 7, 2026 17:40
The Ghidra sync eef6f66 renamed DAT_00df5564 to INT_00df5564 (an int now) and INT_00ed279c to
BOOLEnum_00ed279c. SetTutorialHintActiveWithTimestamp (0x004BC770), UpdateTutorialStepAndProgress and
MenuItemActionHandler_MissionEndscreen_Main still included and named the old ones. They move to main's
names; the BOOLEnum store is written TRUE. No header or status line changes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Style guide rule 14, as most files on main write it; the same header.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gynt
gynt added this pull request to the merge queue Nov 7, 2026
Merged via the queue into sourcehold:main with commit 6023fe5 Nov 7, 2026
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants