On 05/10/2021 22:18, Jacek Caban wrote:
On 10/5/21 4:18 PM, Gabriel Ivăncescu wrote:
Signed-off-by: Gabriel Ivăncescu gabrielopcode@gmail.com
All the names match native, and will be tested later when more are implemented.
dlls/mshtml/dispex.c | 29 +++++++++++++++++++++++++++++ dlls/mshtml/htmlanchor.c | 1 + dlls/mshtml/htmlarea.c | 1 + dlls/mshtml/htmlattr.c | 1 + dlls/mshtml/htmlbody.c | 1 + dlls/mshtml/htmlcomment.c | 1 + dlls/mshtml/htmlcurstyle.c | 1 + dlls/mshtml/htmldoc.c | 2 ++ dlls/mshtml/htmlelem.c | 6 ++++++ dlls/mshtml/htmlelemcol.c | 1 + dlls/mshtml/htmlevent.c | 6 ++++++ dlls/mshtml/htmlform.c | 1 + dlls/mshtml/htmlframe.c | 2 ++ dlls/mshtml/htmlgeneric.c | 1 + dlls/mshtml/htmlhead.c | 4 ++++ dlls/mshtml/htmlimg.c | 2 ++ dlls/mshtml/htmlinput.c | 3 +++ dlls/mshtml/htmllink.c | 1 + dlls/mshtml/htmllocation.c | 1 + dlls/mshtml/htmlnode.c | 2 ++ dlls/mshtml/htmlobject.c | 2 ++ dlls/mshtml/htmlscript.c | 1 + dlls/mshtml/htmlselect.c | 3 +++ dlls/mshtml/htmlstorage.c | 1 + dlls/mshtml/htmlstyle.c | 2 ++ dlls/mshtml/htmlstyleelem.c | 1 + dlls/mshtml/htmlstylesheet.c | 4 ++++ dlls/mshtml/htmltable.c | 3 +++ dlls/mshtml/htmltextarea.c | 1 + dlls/mshtml/htmltextnode.c | 1 + dlls/mshtml/htmlwindow.c | 1 + dlls/mshtml/mshtml_private.h | 2 ++ dlls/mshtml/omnavigator.c | 18 ++++++++++++------ dlls/mshtml/range.c | 2 ++ dlls/mshtml/selection.c | 1 + dlls/mshtml/xmlhttprequest.c | 2 ++ 36 files changed, 106 insertions(+), 6 deletions(-)
diff --git a/dlls/mshtml/dispex.c b/dlls/mshtml/dispex.c index 64ead8f..26aab5e 100644 --- a/dlls/mshtml/dispex.c +++ b/dlls/mshtml/dispex.c @@ -886,6 +886,7 @@ static const dispex_static_data_vtbl_t function_dispex_vtbl = { static const tid_t function_iface_tids[] = {0}; static dispex_static_data_t function_dispex = { + NULL,
The patch looks mostly good, but we're so close to having those names for all objects, that I think it would be best to do it a be able to assume that all objects have name in their description. In this case I believe that this should be "Function". I have suggestions for missing ones bellow.
static dispex_static_data_t HTMLCommentElement_dispex = { + NULL, NULL, DispHTMLCommentElement_tid, HTMLCommentElement_iface_tids,
if(v >= 9) { document.body.innerHTML = "<!--...-->"; test("comment", document.body.firstChild, "Comment"); }
Sounds good to me but I ran into a problem. The comment test is right, of course, but giving it the proper name makes a test fail in dom.c, namely test_doctype.
That one expects its value to always be [object], even in IE9+ mode, implicitly via test_elem_type. However, toString also returns the same thing (see attached diff). And I can't separate the tests in test_elem_type, it already tests for [object].
Now from what I can see, a simple solution would be to add a "value" override in the vtbl for HTMLCommentElement, since toString is based on it, and have it return [object] for doctypes. But how do I check if the comment element is a doctype or a "normal" comment?
Thanks, Gabriel