Various minor extension fixes

chrome/browser/profile.cc

Index: chrome/browser/profile.cc
diff --git a/chrome/browser/profile.cc b/chrome/browser/profile.cc
index 085f8146e3bd0108c53ebdb0cbc0f053f875c0a7..3842ac7887fd839535e6838c01fde04578167c77 100644
--- a/chrome/browser/profile.cc
+++ b/chrome/browser/profile.cc
@@ -568,6 +568,18 @@ void ProfileImpl::InitExtensions() {
       g_browser_process->file_thread()->message_loop());
 
   extensions_service_->Init();
+
+  // Load any extensions specified with --load-extension.
+  if (command_line->HasSwitch(switches::kLoadExtension)) {

[Erik does not do reviews, 2009/07/14 20:22:38]

By moving the load into here, this means you can't invoke chrome.exe with
--load-extension after chrome has already been launched.  Definitely a minor
edge case, but just wanted to make sure it was a conscious decision.

[Aaron Boodman, 2009/07/14 20:35:44]

Hmm. No, not intentional. I moved the code here so that it could be tested.
InProcessBrowserTest was not triggering it in the old location. I guess it seems
like an ok tradeoff?

[Erik does not do reviews, 2009/07/14 20:46:39]

Yep.  I think it's fine like this.

+    std::wstring path_string =
+        command_line->GetSwitchValue(switches::kLoadExtension);
+    FilePath path = FilePath::FromWStringHack(path_string);
+    extensions_service_->LoadExtension(path);
+
+    // Tell UserScriptMaser to watch this extension's directory for changes so
+    // you can live edit content scripts during development.
+    user_script_master_->AddWatchedPath(path);
+  }
 }
 
 void ProfileImpl::InitWebResources() {