[PATCH v2 0/1] MR5893: uniscribe: Update ETO_GLYPH_INDEX flag even if script cache is freed.

[Danyil Blyschak (＠blyss)]

If the script cache was already freed before ScriptTextOut() runs and
hence the sfnt member is not available, use the definition from
init_script_cache() to conditionally apply the ETO_GLYPH_INDEX flag.

--
  v2: uniscribe: Update ETO_GLYPH_INDEX flag even if script cache is freed.

https://gitlab.winehq.org/wine/wine/-/merge_requests/5893

[PATCH v2 1/1] uniscribe: Update ETO_GLYPH_INDEX flag even if script cache is freed.

[Danyil Blyschak]

From: Danyil Blyschak

Call init_script_cache() in ScriptTextOut() to ensure that a cache
exists for the logic around the ETO_GLYPH_INDEX flag.
---
 dlls/gdi32/uniscribe/usp10.c | 1 +
 dlls/usp10/tests/usp10.c     | 1 +
 2 files changed, 2 insertions(+)

diff --git a/dlls/gdi32/uniscribe/usp10.c b/dlls/gdi32/uniscribe/usp10.c
index 72e75224069..233f2a3dc83 100644
--- a/dlls/gdi32/uniscribe/usp10.c
+++ b/dlls/gdi32/uniscribe/usp10.c
@@ -3521,6 +3521,7 @@ HRESULT WINAPI ScriptTextOut(const HDC hdc, SCRIPT_CACHE *psc, int x, int y, UIN

if (!hdc || !psc) return E_INVALIDARG;
     if (!piAdvance || !psa || !pwGlyphs) return E_INVALIDARG;
+    if ((hr = init_script_cache(hdc, psc)) != S_OK) return hr;

fuOptions &= ETO_CLIPPED | ETO_OPAQUE;
     fuOptions |= ETO_IGNORELANGUAGE;
diff --git a/dlls/usp10/tests/usp10.c b/dlls/usp10/tests/usp10.c
index 7516bfb2a97..92b26e56392 100644
--- a/dlls/usp10/tests/usp10.c
+++ b/dlls/usp10/tests/usp10.c
@@ -2777,6 +2777,7 @@ static void test_ScriptTextOut(HDC hdc)
     hr = ScriptTextOut(hdc, &psc, 0, 0, 0, NULL, &pItem[0].a, NULL, 0,
             pwOutGlyphs1, pcGlyphs, piAdvance, NULL, pGoffset);
     ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+    ok(!!psc, "Got unexpected psc %p.\n", psc);

/* Test Rect Rgn is acceptable. */
     SetRect(&rect, 10, 10, 40, 20);

[Danyil Blyschak (＠blyss)]

Pushed commit that calls init_script_cache() instead, and also updates the test.

[Huw Davies (＠huw)]

Could you change the commit subject to be something like: `gdi32/uniscribe: Ensure the cache is initialised.`?

There's now no reason to init `hr` to `S_OK`, so that can be removed.