[PATCH v3 0/1] MR7494: msxml3: IXMLDOMDocument3::preserveWhiteSpace fix for a non VARIANT_BOOL value.

[Alistair Leslie-Hughes (＠alesliehughes)]

0024:trace:msxml:domdoc_put_preserveWhiteSpace (068F0520)->(1)

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=21940

--
  v3: msxml3: IXMLDOMDocument3::preserveWhiteSpace fix for a non VARIANT_BOOL value.

https://gitlab.winehq.org/wine/wine/-/merge_requests/7494

[PATCH v3 1/1] msxml3: IXMLDOMDocument3::preserveWhiteSpace fix for a non VARIANT_BOOL value.

[Alistair Leslie-Hughes]

From: Alistair Leslie-Hughes

0024:trace:msxml:domdoc_put_preserveWhiteSpace (068F0520)->(1)

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=21940
---
 dlls/msxml3/domdoc.c       |  2 +-
 dlls/msxml3/tests/domdoc.c | 21 +++++++++++++++++++++
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/dlls/msxml3/domdoc.c b/dlls/msxml3/domdoc.c
index 8407f914c4f..e1e8e6c803f 100644
--- a/dlls/msxml3/domdoc.c
+++ b/dlls/msxml3/domdoc.c
@@ -2746,7 +2746,7 @@ static HRESULT WINAPI domdoc_put_preserveWhiteSpace(
 {
     domdoc *This = impl_from_IXMLDOMDocument3( iface );
     TRACE("(%p)->(%d)\n", This, isPreserving);
-    This->properties->preserving = isPreserving;
+    This->properties->preserving = isPreserving == VARIANT_TRUE ? VARIANT_TRUE : VARIANT_FALSE;
     return S_OK;
 }

diff --git a/dlls/msxml3/tests/domdoc.c b/dlls/msxml3/tests/domdoc.c
index 76f0c827c2a..93201724ffd 100644
--- a/dlls/msxml3/tests/domdoc.c
+++ b/dlls/msxml3/tests/domdoc.c
@@ -4859,6 +4859,16 @@ static void test_whitespace(void)
         check_ws_ignored(class_ptr->name, doc3, NULL);
         check_ws_preserved(class_ptr->name, doc4, NULL);

+        hr = IXMLDOMDocument2_put_preserveWhiteSpace(doc4, 1);
+        ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+        hr = IXMLDOMDocument2_get_preserveWhiteSpace(doc4, &b);
+        ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+        ok(b == VARIANT_FALSE, "expected true\n");
+        check_ws_ignored(class_ptr->name, doc3, NULL);
+
+        hr = IXMLDOMDocument2_put_preserveWhiteSpace(doc4, VARIANT_TRUE);
+        ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+
         /* setting after loading xml affects trimming of leading/trailing ws only */
         hr = IXMLDOMDocument2_put_preserveWhiteSpace(doc1, VARIANT_TRUE);
         ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
@@ -4925,6 +4935,17 @@ static void test_whitespace(void)
         IXMLDOMNodeList_Release(list);
         IXMLDOMElement_Release(root);

+        hr = IXMLDOMDocument2_put_preserveWhiteSpace(doc1, VARIANT_TRUE);
+        ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+        hr = IXMLDOMDocument2_get_preserveWhiteSpace(doc1, &b);
+        ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+        ok(b == VARIANT_TRUE, "expected true %d\n", b);
+        hr = IXMLDOMDocument2_put_preserveWhiteSpace(doc1, 1);
+        ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+        hr = IXMLDOMDocument2_get_preserveWhiteSpace(doc1, &b);
+        ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
+        ok(b == VARIANT_FALSE, "expected true %d\n", b);
+
         IXMLDOMDocument2_Release(doc1);

free_bstrs();

[Nikolay Sivov (＠nsivov)]

Nikolay Sivov (@nsivov) commented about dlls/msxml3/tests/domdoc.c:
>      check_ws_ignored(class_ptr->name, doc3, NULL);
>      check_ws_preserved(class_ptr->name, doc4, NULL);
>
> -
>    hr = IXMLDOMDocument2_put_preserveWhiteSpace(doc4, 1);
> -
>    ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
> -
>    hr = IXMLDOMDocument2_get_preserveWhiteSpace(doc4, &b);
> -
>    ok(hr == S_OK, "Unexpected hr %#lx.\n", hr);
> -
>    ok(b == VARIANT_FALSE, "expected true\n");
> -
>    check_ws_ignored(class_ptr->name, doc3, NULL);

Sorry, my diff was wrong. This line should be using `doc4`. And once you do that both check_ws_ignored() and check_ws_preserved() fail. So setting it to invalid value in terms of VARIANT_BOOL produces third result.

[Alistair Leslie-Hughes (＠alesliehughes)]

On Fri Mar  7 08:13:42 2025 +0000, Nikolay Sivov wrote:
> Sorry, my diff was wrong. This line should be using `doc4`. And once you
> do that both check_ws_ignored() and check_ws_preserved() fail. So
> setting it to invalid value in terms of VARIANT_BOOL produces third result.

So, what would you suggest?

[Nikolay Sivov (＠nsivov)]

On Sat Mar  8 09:07:38 2025 +0000, Alistair Leslie-Hughes wrote:
> So, what would you suggest?

I suggest to figure out what's different in this case, and see if we can support that.