\nSome people believe that the COM rules on interfaces\nare needlessly strict.\nBut the rules are there for a reason.\n<\/P>\n
\nSuppose you ship some interface in version N of your product.\nIt’s an internal interface,\nnot documented to outsiders.\nTherefore, you are free to change it any time you want without\nhaving to worry about breaking compatibility with any third-party\nplug-ins.\n<\/P>\n
\nBut remember that if you change an interface, you need to generate\na new Interface Identifier (IID).\nBecause an interface identifier uniquely identifies the interface.\n(That’s sort of implied by its name, after all.)\n<\/P>\n
\nAnd this rule applies even to internal interfaces.\n<\/P>\n
\nSuppose you decide to violate this rule and use the same\nIID to represent a slightly different interface in\nversion N+1 of your program.\nSince this is an internal interface,\nyou have no qualms about doing this.\n<\/P>\n
\nUntil you have to write a patch that services both versions.\n<\/P>\n
\nNow your patch is in trouble.\nIt can call\n \nDebugging this problem is not fun.\nNeither is fixing it.\nYour patch has to use some other cues to decide which\ninterface it actually got back.\nIf your program has been patched previously,\nyou need to have the version numbers of every single patch\nso that you can determine which version of the interface you have.\n<\/P>\n \nNote that this dependency can be hidden behind other interfaces.\nConsider:\n<\/P>\n [\n uuid(“XYZ”)\n]\ninterface IGraphicImage\n{\n …\n HRESULT GetColorInfo([out] IColorInfo **ppci);\n};\n<\/PRE>\n \nSuppose you want to add a new method to the\n [\n uuid(“XYZ”)\n]\ninterface IGraphicImage\n{\n …\n HRESULT GetColorInfo([out] IColorInfo **ppci);\n};\n<\/PRE>\n \nYou changed the interface, but you also changed the IID,\nso everything is just fine, right?\n<\/P>\n \nNo, it isn’t.\n<\/P>\n \nThe \nConsider a patch written with the\nversion N+1 header files.\n<\/P>\n \nIf run against version N, the call to\n \nThe quick solution is to change the IID for the\n \nA more robust fix would be to change\nthe \nThis allows interfaces on which Some people believe that the COM rules on interfaces are needlessly strict. But the rules are there for a reason. Suppose you ship some interface in version N of your product. It’s an internal interface, not documented to outsiders. Therefore, you are free to change it any time you want without having to worry about breaking […]<\/p>\n","protected":false},"author":1069,"featured_media":111744,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[1],"tags":[25],"class_list":["post-33533","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-oldnewthing","tag-code"],"acf":[],"blog_post_summary":" Some people believe that the COM rules on interfaces are needlessly strict. But the rules are there for a reason. Suppose you ship some interface in version N of your product. It’s an internal interface, not documented to outsiders. Therefore, you are free to change it any time you want without having to worry about breaking […]<\/p>\n","_links":{"self":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts\/33533","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/users\/1069"}],"replies":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/comments?post=33533"}],"version-history":[{"count":0,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts\/33533\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/media\/111744"}],"wp:attachment":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/media?parent=33533"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/categories?post=33533"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/tags?post=33533"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}IUnknown::QueryInterface<\/CODE>\nand ask for that IID, and it will get something back.\nBut you don’t know whether this is the version N interface\nor the version N+1 interface.\nIf you’re not even aware that this has happened,\nyour patch will probably just assume it has the\nversion N+1 interface,\nand strange things happen when it is run on version N.\n<\/P>\n\n[\n uuid(“ABC”)\n]\ninterface IColorInfo\n{\n HRESULT GetBackgroundColor([out] COLORREF *pcr);\n …\n};<\/p>\nIColorInfo<\/CODE> interface:\n<\/P>\n\n[\n uuid(“DEF<\/FONT>“)\n]\ninterface IColorInfo\n{\n HRESULT GetBackgroundColor([out] COLORREF *pcr);\n …\n HRESULT AdjustColor(COLORREF clrOld,\n COLORREF clrNew);<\/FONT>\n};<\/p>\nIGraphicImage<\/CODE> interface is dependent upon the\nIColorInfo<\/CODE> interface.\nWhen you changed the IColorInfo<\/CODE> interface,\nyou implicitly changed the IGraphicImage::GetColorInfo<\/CODE> method,\nsince the returned interface is now the\nversion N+1 IColorInfo<\/CODE> interface.\n<\/P>\n\nvoid AdjustGraphicColorInfo(IGraphicImage* pgi,\n COLORREF clrOld, COLORREF clrNew)\n{\n IColorInfo *pci;\n if (SUCCEEDED(pgi->GetColorCount(&pci)) {\n pci->AdjustColor(clrOld, clrNew);\n pci->Release();\n }\n}\n<\/PRE>\nIGraphicImage::GetColorCount<\/CODE> will return a\nversion N IColorInfo<\/CODE>, and that version\ndoesn’t support the IColorInfo::AdjustColor<\/CODE>\nmethod.\nBut you’re going to call it anyway.\nResult: Walking off the end of the version N vtable\nand calling into space.\n<\/P>\nIGraphicImage<\/CODE> function to reflect the\nchange on the IColorInfo<\/CODE> interface\non which it depends.\n<\/P>\n\n[\n uuid(“UVW<\/FONT>“)\n]\ninterface IGraphicImage\n{\n …\n HRESULT GetColorInfo([out] IColorInfo **ppci);\n};\n<\/PRE>\nIGraphicImage::GetColorInfo<\/CODE> method\nso that you pass the interface you want to receive.\n<\/P>\n\n[\n uuid(“RST<\/FONT>“)\n]\ninterface IGraphicImage\n{\n …\n HRESULT GetColorInfo([in] REFIID riid,\n [iid_is(riid), out] void** ppv<\/FONT>);\n};\n<\/PRE>\nIGraphicImage<\/CODE> depends\nto change without requiring a change to\nthe IGraphicImage<\/CODE> interface itself.\nOf course, the implementation needs to change to respond to the\nnew value of IID_IColorInfo<\/CODE>.\nBut now the caller can feel safe in the knowledge that when it asks\nfor an interface, it’s actually getting it and not something else\nthat coincidentally has the same name.\n<\/P><\/p>\n","protected":false},"excerpt":{"rendered":"