Wireshark-dev: Re: [Wireshark-dev] SVN 24839 build failures
From: "Maynard, Chris" <[email protected]>
Date: Tue, 8 Apr 2008 01:24:14 -0400
Running a distclean now.  As Stephen pointed out in another e-mail, he ran into this as well, so I'm sure it'll be resolved, especially since the buildbot isn't complaining.  On Saturday I noticed about 20+ of these "unresolved externals" errors, but the number of errors has been slowly coming down so I thought eventually all of them would be gone, so that's why I waited and didn't think I would need a distclean.  Plus I hate doing it if I don't need to because it's so time-consuming.
 
Anyway ... yes I'm back home ... thanks to Jaap who helped me navigate my way from the restaurant back to the hotel.  I would like to add that I really enjoyed meeting all of you.  I wish I had the chance to meet and speak with everyone, but it was a pretty busy couple of days.  Stephen, I thought I would see you later in the day on Monday but obviously we only spoke briefly in the morning.  I should have asked you then what you did, etc. at that time.  I guess I'll have to read about you on the wiki.  Speaking of the wiki, is Gerald ever going to create a homepage?  Or am I really going to have to file a bug report for that? Critical/High priority, of course! :)  For anyone who wasn't at Sharkfest, you really missed out on a great chance to meet some fantastic, incredibly talented people.  Don't miss Sharkfest '09!  And if Anders is reading this, let us know how we can get some pictures.  Andrew, thanks for the IT-crowd.  Good stuff.  Thanks for everything CACE!
 
- Chris


From: Ulf Lamping
Sent: Tue 4/8/2008 12:42 AM
To: Developer support list for Wireshark
Subject: Re: [Wireshark-dev] SVN 24839 build failures

Maynard, Chris schrieb:
> The wimax plugin fails to compile on Windows (XP, 32-bit) because of 
> the following:
>  
> msg_sbc.c
> msg_sbc.c(525) : error C2491: 'tfs_yes_no' : definition of dllimport 
> data not allowed
> msg_sbc.c(567) : error C2099: initializer is not a constant
> msg_sbc.c(567) : warning C4047: 'initializing' : 'const void *' 
> differs in levels of indirection from 'const int '
>  
> The attached patch resolves the problem by renaming tfs_yes_no to 
> simply yes_no and it also removes the #ifdef _WIN32 ... #endif wrapper 
> around it.
I fixed it by replacing it with tfs_yes_no_sbc
>  
> However, there are 8 unresolved externals that show up later on - all 
> with respect to the recent GTK changes - nothing to do with the wimax 
> plugin.  I haven't bothered reporting these problems since I know at 
> least a few developers are still in the middle of making changes, and 
> I figured one or more buildbots would complain anyway, but the 
> buildbots don't seem to be complaining, so here are the unresolved 
> externals I'm seeing (duplicates not listed):
>  
> Linking wireshark.exe
> libui.lib(print_prefs.obj) : error LNK2001: unresolved external symbol 
> _gtk_signal_connect
> libui.lib(capture_prefs.obj) : error LNK2001: unresolved external 
> symbol _gtk_type_check_object_cast
> libui.lib(menu.obj) : error LNK2001: unresolved external symbol 
> _eo_http_cb
> libui.lib(airpcap_gui_utils.obj) : error LNK2001: unresolved external 
> symbol _gtk_signal_handler_unblock_by_func
> libui.lib(airpcap_gui_utils.obj) : error LNK2001: unresolved external 
> symbol _gtk_signal_handler_block_by_func
> libui.lib(sctp_assoc_analyse.obj) : error LNK2001: unresolved external 
> symbol _gtk_button_box_set_spacing
> libui.lib(scsi_stat.obj) : error LNK2001: unresolved external symbol 
> _gtk_menu_append
> wireshark.exe : fatal error LNK1120: 8 unresolved externals
>  
> Hmm, that's only 7 unique errors.  Not sure why 8 are reported, but I 
> can't find any others?  Oh well.
> - Chris
I have no problems here, maybe trying a distclean?

Let's see what the buildbot thinks ...

Regards, ULFL

P.S: Seems you're back home and fine :-)