[lmi] Complete replacement of help engine [Re: wxHTML table header alignment]

[Greg Chicares]

On 2009-03-11 15:19Z, Vadim Zeitlin wrote:
> On Wed, 11 Mar 2009 15:05:13 +0000 Greg Chicares <address@hidden> wrote:
> 
> GC> On 2009-03-11 14:17Z, Vadim Zeitlin wrote:
> GC> > On Wed, 11 Mar 2009 10:04:29 +0000 Greg Chicares <address@hidden> wrote:
> GC> > 
> GC> > GC> | - Show busy cursor in wxLaunchDefaultBrowser and add 
> wxBROWSER_NOBUSYCURSOR.
> GC> > GC> 
> GC> > GC> A stray thought: maybe we should use wxLaunchDefaultBrowser instead
> GC> > GC> of wxHtmlHelp.
> GC> > 
> GC> >  It was possible to do this before, wxLaunchDefaultBrowser() just makes 
> it
> GC> > simpler, so this shouldn't really change anything for this question. But
> GC> > it's indeed much more common to show help in the (users default) browser
> GC> > rather than doing it inside the application itself nowadays.
> GC> 
> GC> Would wxExtHelpController take care of that in the best way?
> 
>  This would probably be the simplest way when switching from
> wxHtmlHelpController as they have almost the same API. But I don't know
> whether it has any advantages for LMI compared to just opening the external
> browser on the help contents page as LMI doesn't seem to use neither help
> index nor provide a search command in help menu as one of the main added
> benefits of this class is support for these commands.

USA: Do you see any disadvantage in this change, which otherwise
I will apply this weekend?

Index: main_wx.cpp
===================================================================
RCS file: /sources/lmi/lmi/main_wx.cpp,v
retrieving revision 1.129
diff -U 3 -r1.129 main_wx.cpp
--- main_wx.cpp 11 Mar 2009 15:59:39 -0000      1.129
+++ main_wx.cpp 12 Mar 2009 19:27:06 -0000
@@ -526,7 +526,7 @@

 void Skeleton::UponHelp(wxCommandEvent&)
 {
-    help_controller_->DisplayContents();
+    wxLaunchDefaultBrowser("http://www.nongnu.org/lmi/user_manual.html";);
 }

 /// Rethrow an exception caught by wx into a local catch clause.

Europe: I know about "file:"; and I'd check the return code;
and I realize it can change the floating-point control word.
This is only a quick "proof of concept", not a polished
revision that I'd actually commit.