1 files changed, 538 insertions, 0 deletions

diff --git a/src/dutil/uriutil.cpp b/src/dutil/uriutil.cpp
new file mode 100644
index 00000000..fc192b3f
--- /dev/null
+++ b/src/dutil/uriutil.cpp
@@ -0,0 +1,538 @@
+// Copyright (c) .NET Foundation and contributors. All rights reserved. Licensed under the Microsoft Reciprocal License. See LICENSE.TXT file in the project root for full license information.
+
+#include "precomp.h"
+
+
+//
+// UriCanonicalize - canonicalizes a URI.
+//
+extern "C" HRESULT DAPI UriCanonicalize(
+    __inout_z LPWSTR* psczUri
+    )
+{
+    HRESULT hr = S_OK;
+    WCHAR wz[INTERNET_MAX_URL_LENGTH] = { };
+    DWORD cch = countof(wz);
+
+    if (!::InternetCanonicalizeUrlW(*psczUri, wz, &cch, ICU_DECODE))
+    {
+        ExitWithLastError(hr, "Failed to canonicalize URI.");
+    }
+
+    hr = StrAllocString(psczUri, wz, cch);
+    ExitOnFailure(hr, "Failed copy canonicalized URI.");
+
+LExit:
+    return hr;
+}
+
+
+//
+// UriCrack - cracks a URI into constituent parts.
+//
+extern "C" HRESULT DAPI UriCrack(
+    __in_z LPCWSTR wzUri,
+    __out_opt INTERNET_SCHEME* pScheme,
+    __deref_opt_out_z LPWSTR* psczHostName,
+    __out_opt INTERNET_PORT* pPort,
+    __deref_opt_out_z LPWSTR* psczUser,
+    __deref_opt_out_z LPWSTR* psczPassword,
+    __deref_opt_out_z LPWSTR* psczPath,
+    __deref_opt_out_z LPWSTR* psczQueryString
+    )
+{
+    HRESULT hr = S_OK;
+    URL_COMPONENTSW components = { };
+    WCHAR wzHostName[INTERNET_MAX_HOST_NAME_LENGTH + 1];
+    WCHAR wzUserName[INTERNET_MAX_USER_NAME_LENGTH + 1];
+    WCHAR wzPassword[INTERNET_MAX_PASSWORD_LENGTH + 1];
+    WCHAR wzPath[INTERNET_MAX_PATH_LENGTH + 1];
+    WCHAR wzQueryString[INTERNET_MAX_PATH_LENGTH + 1];
+
+    components.dwStructSize = sizeof(URL_COMPONENTSW);
+
+    if (psczHostName)
+    {
+        components.lpszHostName = wzHostName;
+        components.dwHostNameLength = countof(wzHostName);
+    }
+
+    if (psczUser)
+    {
+        components.lpszUserName = wzUserName;
+        components.dwUserNameLength = countof(wzUserName);
+    }
+
+    if (psczPassword)
+    {
+        components.lpszPassword = wzPassword;
+        components.dwPasswordLength = countof(wzPassword);
+    }
+
+    if (psczPath)
+    {
+        components.lpszUrlPath = wzPath;
+        components.dwUrlPathLength = countof(wzPath);
+    }
+
+    if (psczQueryString)
+    {
+        components.lpszExtraInfo = wzQueryString;
+        components.dwExtraInfoLength = countof(wzQueryString);
+    }
+
+    if (!::InternetCrackUrlW(wzUri, 0, ICU_DECODE | ICU_ESCAPE, &components))
+    {
+        ExitWithLastError(hr, "Failed to crack URI.");
+    }
+
+    if (pScheme)
+    {
+        *pScheme = components.nScheme;
+    }
+
+    if (psczHostName)
+    {
+        hr = StrAllocString(psczHostName, components.lpszHostName, components.dwHostNameLength);
+        ExitOnFailure(hr, "Failed to copy host name.");
+    }
+
+    if (pPort)
+    {
+        *pPort = components.nPort;
+    }
+
+    if (psczUser)
+    {
+        hr = StrAllocString(psczUser, components.lpszUserName, components.dwUserNameLength);
+        ExitOnFailure(hr, "Failed to copy user name.");
+    }
+
+    if (psczPassword)
+    {
+        hr = StrAllocString(psczPassword, components.lpszPassword, components.dwPasswordLength);
+        ExitOnFailure(hr, "Failed to copy password.");
+    }
+
+    if (psczPath)
+    {
+        hr = StrAllocString(psczPath, components.lpszUrlPath, components.dwUrlPathLength);
+        ExitOnFailure(hr, "Failed to copy path.");
+    }
+
+    if (psczQueryString)
+    {
+        hr = StrAllocString(psczQueryString, components.lpszExtraInfo, components.dwExtraInfoLength);
+        ExitOnFailure(hr, "Failed to copy query string.");
+    }
+
+LExit:
+    return hr;
+}
+
+
+//
+// UriCrackEx - cracks a URI into URI_INFO.
+//
+extern "C" HRESULT DAPI UriCrackEx(
+    __in_z LPCWSTR wzUri,
+    __in URI_INFO* pUriInfo
+    )
+{
+    HRESULT hr = S_OK;
+
+    hr = UriCrack(wzUri, &pUriInfo->scheme, &pUriInfo->sczHostName, &pUriInfo->port, &pUriInfo->sczUser, &pUriInfo->sczPassword, &pUriInfo->sczPath, &pUriInfo->sczQueryString);
+    ExitOnFailure(hr, "Failed to crack URI.");
+
+LExit:
+    return hr;
+}
+
+
+//
+// UriInfoUninitialize - frees the memory in a URI_INFO struct.
+//
+extern "C" void DAPI UriInfoUninitialize(
+    __in URI_INFO* pUriInfo
+    )
+{
+    ReleaseStr(pUriInfo->sczHostName);
+    ReleaseStr(pUriInfo->sczUser);
+    ReleaseStr(pUriInfo->sczPassword);
+    ReleaseStr(pUriInfo->sczPath);
+    ReleaseStr(pUriInfo->sczQueryString);
+    memset(pUriInfo, 0, sizeof(URI_INFO));
+}
+
+
+//
+// UriCreate - creates a URI from constituent parts.
+//
+extern "C" HRESULT DAPI UriCreate(
+    __inout_z LPWSTR* psczUri,
+    __in INTERNET_SCHEME scheme,
+    __in_z_opt LPWSTR wzHostName,
+    __in INTERNET_PORT port,
+    __in_z_opt LPWSTR wzUser,
+    __in_z_opt LPWSTR wzPassword,
+    __in_z_opt LPWSTR wzPath,
+    __in_z_opt LPWSTR wzQueryString
+    )
+{
+    HRESULT hr = S_OK;
+    WCHAR wz[INTERNET_MAX_URL_LENGTH] = { };
+    DWORD cch = countof(wz);
+    URL_COMPONENTSW components = { };
+
+    components.dwStructSize = sizeof(URL_COMPONENTSW);
+    components.nScheme = scheme;
+    components.lpszHostName = wzHostName;
+    components.nPort = port;
+    components.lpszUserName = wzUser;
+    components.lpszPassword = wzPassword;
+    components.lpszUrlPath = wzPath;
+    components.lpszExtraInfo = wzQueryString;
+
+    if (!::InternetCreateUrlW(&components, ICU_ESCAPE, wz, &cch))
+    {
+        ExitWithLastError(hr, "Failed to create URI.");
+    }
+
+    hr = StrAllocString(psczUri, wz, cch);
+    ExitOnFailure(hr, "Failed copy created URI.");
+
+LExit:
+    return hr;
+}
+
+
+//
+// UriGetServerAndResource - gets the server and resource as independent strings from a URI.
+//
+// NOTE: This function is useful for the InternetConnect/HttpRequest APIs.
+//
+extern "C" HRESULT DAPI UriGetServerAndResource(
+    __in_z LPCWSTR wzUri,
+    __out_z LPWSTR* psczServer,
+    __out_z LPWSTR* psczResource
+    )
+{
+    HRESULT hr = S_OK;
+    INTERNET_SCHEME scheme = INTERNET_SCHEME_UNKNOWN;
+    LPWSTR sczHostName = NULL;
+    INTERNET_PORT port = INTERNET_INVALID_PORT_NUMBER;
+    LPWSTR sczUser = NULL;
+    LPWSTR sczPassword = NULL;
+    LPWSTR sczPath = NULL;
+    LPWSTR sczQueryString = NULL;
+
+    hr = UriCrack(wzUri, &scheme, &sczHostName, &port, &sczUser, &sczPassword, &sczPath, &sczQueryString);
+    ExitOnFailure(hr, "Failed to crack URI.");
+
+    hr = UriCreate(psczServer, scheme, sczHostName, port, sczUser, sczPassword, NULL, NULL);
+    ExitOnFailure(hr, "Failed to allocate server URI.");
+
+    hr = UriCreate(psczResource, INTERNET_SCHEME_UNKNOWN, NULL, INTERNET_INVALID_PORT_NUMBER, NULL, NULL, sczPath, sczQueryString);
+    ExitOnFailure(hr, "Failed to allocate resource URI.");
+
+LExit:
+    ReleaseStr(sczQueryString);
+    ReleaseStr(sczPath);
+    ReleaseStr(sczPassword);
+    ReleaseStr(sczUser);
+    ReleaseStr(sczHostName);
+
+    return hr;
+}
+
+
+//
+// UriFile - returns the file part of the URI.
+//
+extern "C" HRESULT DAPI UriFile(
+    __deref_out_z LPWSTR* psczFile,
+    __in_z LPCWSTR wzUri
+    )
+{
+    HRESULT hr = S_OK;
+    WCHAR wz[MAX_PATH + 1];
+    DWORD cch = countof(wz);
+    URL_COMPONENTSW uc = { };
+
+    uc.dwStructSize = sizeof(uc);
+    uc.lpszUrlPath = wz;
+    uc.dwUrlPathLength = cch;
+
+    if (!::InternetCrackUrlW(wzUri, 0, ICU_DECODE | ICU_ESCAPE, &uc))
+    {
+        ExitWithLastError(hr, "Failed to crack URI.");
+    }
+
+    // Copy only the file name. Fortunately, PathFile() understands that
+    // forward slashes can be directory separators like backslashes.
+    hr = StrAllocString(psczFile, PathFile(wz), 0);
+    ExitOnFailure(hr, "Failed to copy file name");
+
+LExit:
+    return hr;
+}
+
+
+/*******************************************************************
+ UriProtocol - determines the protocol of a URI.
+
+********************************************************************/
+extern "C" HRESULT DAPI UriProtocol(
+    __in_z LPCWSTR wzUri,
+    __out URI_PROTOCOL* pProtocol
+    )
+{
+    Assert(wzUri && *wzUri);
+    Assert(pProtocol);
+
+    HRESULT hr = S_OK;
+
+    if ((L'h' == wzUri[0] || L'H' == wzUri[0]) &&
+        (L't' == wzUri[1] || L'T' == wzUri[1]) &&
+        (L't' == wzUri[2] || L'T' == wzUri[2]) &&
+        (L'p' == wzUri[3] || L'P' == wzUri[3]) &&
+        (L's' == wzUri[4] || L'S' == wzUri[4]) &&
+         L':' == wzUri[5] &&
+         L'/' == wzUri[6] &&
+         L'/' == wzUri[7])
+    {
+        *pProtocol = URI_PROTOCOL_HTTPS;
+    }
+    else if ((L'h' == wzUri[0] || L'H' == wzUri[0]) &&
+             (L't' == wzUri[1] || L'T' == wzUri[1]) &&
+             (L't' == wzUri[2] || L'T' == wzUri[2]) &&
+             (L'p' == wzUri[3] || L'P' == wzUri[3]) &&
+              L':' == wzUri[4] &&
+              L'/' == wzUri[5] &&
+              L'/' == wzUri[6])
+    {
+        *pProtocol = URI_PROTOCOL_HTTP;
+    }
+    else if ((L'f' == wzUri[0] || L'F' == wzUri[0]) &&
+             (L't' == wzUri[1] || L'T' == wzUri[1]) &&
+             (L'p' == wzUri[2] || L'P' == wzUri[2]) &&
+              L':' == wzUri[3] &&
+              L'/' == wzUri[4] &&
+              L'/' == wzUri[5])
+    {
+        *pProtocol = URI_PROTOCOL_FTP;
+    }
+    else if ((L'f' == wzUri[0] || L'F' == wzUri[0]) &&
+             (L'i' == wzUri[1] || L'I' == wzUri[1]) &&
+             (L'l' == wzUri[2] || L'L' == wzUri[2]) &&
+             (L'e' == wzUri[3] || L'E' == wzUri[3]) &&
+              L':' == wzUri[4] &&
+              L'/' == wzUri[5] &&
+              L'/' == wzUri[6])
+    {
+        *pProtocol = URI_PROTOCOL_FILE;
+    }
+    else
+    {
+        *pProtocol = URI_PROTOCOL_UNKNOWN;
+    }
+
+    return hr;
+}
+
+
+/*******************************************************************
+ UriRoot - returns the root of the path specified in the URI.
+
+ examples:
+    file:///C:\path\path             -> C:\
+    file://server/share/path/path    -> \\server\share
+    http://www.example.com/path/path -> http://www.example.com/
+    ftp://ftp.example.com/path/path  -> ftp://www.example.com/
+
+ NOTE: This function should only be used on cannonicalized URIs.
+       It does not cannonicalize itself.
+********************************************************************/
+extern "C" HRESULT DAPI UriRoot(
+    __in_z LPCWSTR wzUri,
+    __out LPWSTR* ppwzRoot,
+    __out_opt URI_PROTOCOL* pProtocol
+    )
+{
+    Assert(wzUri && *wzUri);
+    Assert(ppwzRoot);
+
+    HRESULT hr = S_OK;
+    URI_PROTOCOL protocol = URI_PROTOCOL_UNKNOWN;
+    LPCWSTR pwcSlash = NULL;
+
+    hr = UriProtocol(wzUri, &protocol);
+    ExitOnFailure(hr, "Invalid URI.");
+
+    switch (protocol)
+    {
+    case URI_PROTOCOL_FILE:
+        if (L'/' == wzUri[7]) // file path
+        {
+            if (((L'a' <= wzUri[8] && L'z' >= wzUri[8]) || (L'A' <= wzUri[8] && L'Z' >= wzUri[8])) && L':' == wzUri[9])
+            {
+                hr = StrAlloc(ppwzRoot, 4);
+                ExitOnFailure(hr, "Failed to allocate string for root of URI.");
+                *ppwzRoot[0] = wzUri[8];
+                *ppwzRoot[1] = L':';
+                *ppwzRoot[2] = L'\\';
+                *ppwzRoot[3] = L'\0';
+            }
+            else
+            {
+                hr = E_INVALIDARG;
+                ExitOnFailure(hr, "Invalid file path in URI.");
+            }
+        }
+        else // UNC share
+        {
+            pwcSlash = wcschr(wzUri + 8, L'/');
+            if (!pwcSlash)
+            {
+                hr = E_INVALIDARG;
+                ExitOnFailure(hr, "Invalid server name in URI.");
+            }
+            else
+            {
+                hr = StrAllocString(ppwzRoot, L"\\\\", 64);
+                ExitOnFailure(hr, "Failed to allocate string for root of URI.");
+
+                pwcSlash = wcschr(pwcSlash + 1, L'/');
+                if (pwcSlash)
+                {
+                    hr = StrAllocConcat(ppwzRoot, wzUri + 8, pwcSlash - wzUri - 8);
+                    ExitOnFailure(hr, "Failed to add server/share to root of URI.");
+                }
+                else
+                {
+                    hr = StrAllocConcat(ppwzRoot, wzUri + 8, 0);
+                    ExitOnFailure(hr, "Failed to add server/share to root of URI.");
+                }
+
+                // replace all slashes with backslashes to be truly UNC.
+                for (LPWSTR pwc = *ppwzRoot; pwc && *pwc; ++pwc)
+                {
+                    if (L'/' == *pwc)
+                    {
+                        *pwc = L'\\';
+                    }
+                }
+            }
+        }
+        break;
+
+    case URI_PROTOCOL_FTP:
+        pwcSlash = wcschr(wzUri + 6, L'/');
+        if (pwcSlash)
+        {
+            hr = StrAllocString(ppwzRoot, wzUri, pwcSlash - wzUri);
+            ExitOnFailure(hr, "Failed allocate root from URI.");
+        }
+        else
+        {
+            hr = StrAllocString(ppwzRoot, wzUri, 0);
+            ExitOnFailure(hr, "Failed allocate root from URI.");
+        }
+        break;
+
+    case URI_PROTOCOL_HTTP:
+        pwcSlash = wcschr(wzUri + 7, L'/');
+        if (pwcSlash)
+        {
+            hr = StrAllocString(ppwzRoot, wzUri, pwcSlash - wzUri);
+            ExitOnFailure(hr, "Failed allocate root from URI.");
+        }
+        else
+        {
+            hr = StrAllocString(ppwzRoot, wzUri, 0);
+            ExitOnFailure(hr, "Failed allocate root from URI.");
+        }
+        break;
+
+    default:
+        hr = E_INVALIDARG;
+        ExitOnFailure(hr, "Unknown URI protocol.");
+    }
+
+    if (pProtocol)
+    {
+        *pProtocol = protocol;
+    }
+
+LExit:
+    return hr;
+}
+
+
+extern "C" HRESULT DAPI UriResolve(
+    __in_z LPCWSTR wzUri,
+    __in_opt LPCWSTR wzBaseUri,
+    __out LPWSTR* ppwzResolvedUri,
+    __out_opt const URI_PROTOCOL* pResolvedProtocol
+    )
+{
+    UNREFERENCED_PARAMETER(wzUri);
+    UNREFERENCED_PARAMETER(wzBaseUri);
+    UNREFERENCED_PARAMETER(ppwzResolvedUri);
+    UNREFERENCED_PARAMETER(pResolvedProtocol);
+
+    HRESULT hr = E_NOTIMPL;
+#if 0
+    URI_PROTOCOL protocol = URI_PROTOCOL_UNKNOWN;
+
+    hr = UriProtocol(wzUri, &protocol);
+    ExitOnFailure(hr, "Failed to determine protocol for URL: %ls", wzUri);
+
+    ExitOnNull(ppwzResolvedUri, hr, E_INVALIDARG, "Failed to resolve URI, because no method of output was provided");
+
+    if (URI_PROTOCOL_UNKNOWN == protocol)
+    {
+        ExitOnNull(wzBaseUri, hr, E_INVALIDARG, "Failed to resolve URI - base URI provided was NULL");
+
+        if (L'/' == *wzUri || L'\\' == *wzUri)
+        {
+            hr = UriRoot(wzBaseUri, ppwzResolvedUri, &protocol);
+            ExitOnFailure(hr, "Failed to get root from URI: %ls", wzBaseUri);
+
+            hr = StrAllocConcat(ppwzResolvedUri, wzUri, 0);
+            ExitOnFailure(hr, "Failed to concat file to base URI.");
+        }
+        else
+        {
+            hr = UriProtocol(wzBaseUri, &protocol);
+            ExitOnFailure(hr, "Failed to get protocol of base URI: %ls", wzBaseUri);
+
+            LPCWSTR pwcFile = const_cast<LPCWSTR> (UriFile(wzBaseUri));
+            if (!pwcFile)
+            {
+                hr = E_INVALIDARG;
+                ExitOnFailure(hr, "Failed to get file from base URI: %ls", wzBaseUri);
+            }
+
+            hr = StrAllocString(ppwzResolvedUri, wzBaseUri, pwcFile - wzBaseUri);
+            ExitOnFailure(hr, "Failed to allocate string for resolved URI.");
+
+            hr = StrAllocConcat(ppwzResolvedUri, wzUri, 0);
+            ExitOnFailure(hr, "Failed to concat file to resolved URI.");
+        }
+    }
+    else
+    {
+        hr = StrAllocString(ppwzResolvedUri, wzUri, 0);
+        ExitOnFailure(hr, "Failed to copy resolved URI.");
+    }
+
+    if (pResolvedProtocol)
+    {
+        *pResolvedProtocol = protocol;
+    }
+
+LExit:
+#endif
+    return hr;
+}