net: Provide better HTTP User-Agent strings

Previously, macOS used “User-Agent: crashpad_util_test (unknown version)
CFNetwork/807.2.14 Darwin/16.4.0 (x86_64)” and Windows gave results like
“User-Agent: Crashpad/0.8.0”.

Now, macOS uses “User-Agent: Crashpad/0.8.0 CFNetwork/807.2.14
Darwin/16.4.0 (x86_64)” and Windows uses “User-Agent: Crashpad/0.8.0
WinHTTP/10.0.14393.351 Windows_NT/10.0.14393.0 (x64)”

Change-Id: I578b44734cf59d79e3d9b6136b4b92f05acefe71
Reviewed-on: https://chromium-review.googlesource.com/447796
Reviewed-by: Robert Sesek <rsesek@chromium.org>
Reviewed-by: Scott Graham <scottmg@chromium.org>

snapshot/win/system_snapshot_win.cc

@@ -106,17 +106,17 @@ void SystemSnapshotWin::Initialize(ProcessReaderWin* process_reader) {
     os_version_major_ = ffi.dwFileVersionMS >> 16;
     os_version_minor_ = ffi.dwFileVersionMS & 0xffff;
     os_version_bugfix_ = ffi.dwFileVersionLS >> 16;
-    os_version_build_ =
-        base::StringPrintf("%d", ffi.dwFileVersionLS & 0xffff);
+    os_version_build_ = base::StringPrintf("%u", ffi.dwFileVersionLS & 0xffff);
     os_version_full_ = base::StringPrintf(
-        "%s %d.%d.%d.%s%s",
+        "%s %u.%u.%u.%s%s",
         os_name.c_str(),
         os_version_major_,
         os_version_minor_,
         os_version_bugfix_,
         os_version_build_.c_str(),
-        flags_string.empty() ? "" : (std::string(" (") + flags_string + ")")
-                                        .c_str());
+        flags_string.empty()
+            ? ""
+            : (std::string(" (") + flags_string + ")").c_str());
   }
 
   INITIALIZATION_STATE_SET_VALID(initialized_);

util/net/http_transport_mac.mm

@@ -16,11 +16,14 @@
 
 #include <CoreFoundation/CoreFoundation.h>
 #import <Foundation/Foundation.h>
+#include <sys/utsname.h>
 
 #include "base/mac/foundation_util.h"
 #import "base/mac/scoped_nsobject.h"
 #include "base/strings/stringprintf.h"
 #include "base/strings/sys_string_conversions.h"
+#include "build/build_config.h"
+#include "package.h"
 #include "third_party/apple_cf/CFStreamAbstract.h"
 #include "util/file/file_io.h"
 #include "util/misc/implicit_cast.h"
@@ -30,6 +33,80 @@ namespace crashpad {
 
 namespace {
 
+NSString* AppendEscapedFormat(NSString* base,
+                              NSString* format,
+                              NSString* data) {
+  return [base stringByAppendingFormat:
+                   format,
+                   [data stringByAddingPercentEncodingWithAllowedCharacters:
+                             [[NSCharacterSet
+                                 characterSetWithCharactersInString:
+                                     @"()<>@,;:\\\"/[]?={} \t"] invertedSet]]];
+}
+
+// This builds the same User-Agent string that CFNetwork would build internally,
+// but it uses PACKAGE_NAME and PACKAGE_VERSION in place of values obtained from
+// the main bundle’s Info.plist.
+NSString* UserAgentString() {
+  NSString* user_agent = [NSString string];
+
+  // CFNetwork would use the main bundle’s CFBundleName, or the main
+  // executable’s filename if none.
+  user_agent = AppendEscapedFormat(
+      user_agent, @"%@", [NSString stringWithUTF8String:PACKAGE_NAME]);
+
+  // CFNetwork would use the main bundle’s CFBundleVersion, or the string
+  // “(unknown version)” if none.
+  user_agent = AppendEscapedFormat(
+      user_agent, @"/%@", [NSString stringWithUTF8String:PACKAGE_VERSION]);
+
+  // Expected to be CFNetwork.
+  NSBundle* nsurl_bundle = [NSBundle bundleForClass:[NSURLRequest class]];
+  NSString* bundle_name = base::mac::ObjCCast<NSString>([nsurl_bundle
+      objectForInfoDictionaryKey:base::mac::CFToNSCast(kCFBundleNameKey)]);
+  if (bundle_name) {
+    user_agent = AppendEscapedFormat(user_agent, @" %@", bundle_name);
+
+    NSString* bundle_version = base::mac::ObjCCast<NSString>([nsurl_bundle
+        objectForInfoDictionaryKey:base::mac::CFToNSCast(kCFBundleVersionKey)]);
+    if (bundle_version) {
+      user_agent = AppendEscapedFormat(user_agent, @"/%@", bundle_version);
+    }
+  }
+
+  utsname os;
+  if (uname(&os) != 0) {
+    PLOG(WARNING) << "uname";
+  } else {
+    user_agent = AppendEscapedFormat(
+        user_agent, @" %@", [NSString stringWithUTF8String:os.sysname]);
+    user_agent = AppendEscapedFormat(
+        user_agent, @"/%@", [NSString stringWithUTF8String:os.release]);
+
+    // CFNetwork just uses the equivalent of os.machine to obtain the native
+    // (kernel) architecture. Here, give the process’ architecture as well as
+    // the native architecture. Use the same strings that the kernel would, so
+    // that they can be de-duplicated.
+#if defined(ARCH_CPU_X86)
+    NSString* arch = @"i386";
+#elif defined(ARCH_CPU_X86_64)
+    NSString* arch = @"x86_64";
+#else
+#error Port
+#endif
+    user_agent = AppendEscapedFormat(user_agent, @" (%@", arch);
+
+    NSString* machine = [NSString stringWithUTF8String:os.machine];
+    if (![machine isEqualToString:arch]) {
+      user_agent = AppendEscapedFormat(user_agent, @"; %@", machine);
+    }
+
+    user_agent = [user_agent stringByAppendingString:@")"];
+  }
+
+  return user_agent;
+}
+
 // An implementation of CFReadStream. This implements the V0 callback
 // scheme.
 class HTTPBodyStreamCFReadStream {
@@ -171,6 +248,13 @@ bool HTTPTransportMac::ExecuteSynchronously(std::string* response_body) {
                             timeoutInterval:timeout()];
     [request setHTTPMethod:base::SysUTF8ToNSString(method())];
 
+    // If left to its own devices, CFNetwork would build a user-agent string
+    // based on keys in the main bundle’s Info.plist, giving ugly results if
+    // there is no Info.plist. Provide a User-Agent string similar to the one
+    // that CFNetwork would use, but with appropriate values in place of the
+    // Info.plist-derived strings.
+    [request setValue:UserAgentString() forHTTPHeaderField:@"User-Agent"];
+
     for (const auto& pair : headers()) {
       [request setValue:base::SysUTF8ToNSString(pair.second)
           forHTTPHeaderField:base::SysUTF8ToNSString(pair.first)];

util/net/http_transport_win.cc

@@ -28,15 +28,59 @@
 #include "base/strings/string_number_conversions.h"
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversions.h"
+#include "build/build_config.h"
 #include "package.h"
 #include "util/file/file_io.h"
 #include "util/numeric/safe_assignment.h"
 #include "util/net/http_body.h"
+#include "util/win/module_version.h"
 
 namespace crashpad {
 
 namespace {
 
+const wchar_t kWinHttpDll[] = L"winhttp.dll";
+
+std::string UserAgent() {
+  std::string user_agent =
+      base::StringPrintf("%s/%s WinHTTP", PACKAGE_NAME, PACKAGE_VERSION);
+
+  VS_FIXEDFILEINFO version;
+  if (GetModuleVersionAndType(base::FilePath(kWinHttpDll), &version)) {
+    user_agent.append(base::StringPrintf("/%u.%u.%u.%u",
+                                         version.dwFileVersionMS >> 16,
+                                         version.dwFileVersionMS & 0xffff,
+                                         version.dwFileVersionLS >> 16,
+                                         version.dwFileVersionLS & 0xffff));
+  }
+
+  if (GetModuleVersionAndType(base::FilePath(L"kernel32.dll"), &version) &&
+      (version.dwFileOS & VOS_NT_WINDOWS32) == VOS_NT_WINDOWS32) {
+    user_agent.append(base::StringPrintf(" Windows_NT/%u.%u.%u.%u (",
+                                         version.dwFileVersionMS >> 16,
+                                         version.dwFileVersionMS & 0xffff,
+                                         version.dwFileVersionLS >> 16,
+                                         version.dwFileVersionLS & 0xffff));
+#if defined(ARCH_CPU_X86)
+    user_agent.append("x86");
+#elif defined(ARCH_CPU_X86_64)
+    user_agent.append("x64");
+#else
+#error Port
+#endif
+
+    BOOL is_wow64;
+    if (!IsWow64Process(GetCurrentProcess(), &is_wow64)) {
+      PLOG(WARNING) << "IsWow64Process";
+    } else if (is_wow64) {
+      user_agent.append("; WoW64");
+    }
+    user_agent.append(1, ')');
+  }
+
+  return user_agent;
+}
+
 // PLOG doesn't work for messages from WinHTTP, so we need to use
 // FORMAT_MESSAGE_FROM_HMODULE + the dll name manually here.
 std::string WinHttpMessage(const char* extra) {
@@ -45,7 +89,7 @@ std::string WinHttpMessage(const char* extra) {
   DWORD flags = FORMAT_MESSAGE_FROM_SYSTEM | FORMAT_MESSAGE_IGNORE_INSERTS |
                 FORMAT_MESSAGE_MAX_WIDTH_MASK | FORMAT_MESSAGE_FROM_HMODULE;
   DWORD len = FormatMessageA(flags,
-                             GetModuleHandle(L"winhttp.dll"),
+                             GetModuleHandle(kWinHttpDll),
                              error_code,
                              0,
                              msgbuf,
@@ -93,12 +137,11 @@ HTTPTransportWin::~HTTPTransportWin() {
 }
 
 bool HTTPTransportWin::ExecuteSynchronously(std::string* response_body) {
-  ScopedHINTERNET session(
-      WinHttpOpen(base::UTF8ToUTF16(PACKAGE_NAME "/" PACKAGE_VERSION).c_str(),
-                  WINHTTP_ACCESS_TYPE_DEFAULT_PROXY,
-                  WINHTTP_NO_PROXY_NAME,
-                  WINHTTP_NO_PROXY_BYPASS,
-                  0));
+  ScopedHINTERNET session(WinHttpOpen(base::UTF8ToUTF16(UserAgent()).c_str(),
+                                      WINHTTP_ACCESS_TYPE_DEFAULT_PROXY,
+                                      WINHTTP_NO_PROXY_NAME,
+                                      WINHTTP_NO_PROXY_BYPASS,
+                                      0));
   if (!session.get()) {
     LOG(ERROR) << WinHttpMessage("WinHttpOpen");
     return false;