Re: [Evolution-hackers] More e-d-s ABI breakage ?

[Harish Krishnaswamy <kharish novell com>]

On Fri, 2006-08-11 at 10:00 +0100, Michael Meeks wrote:
> So,
> --- source/drivers/evoab2/EApi.cxx
> +++ source/drivers/evoab2/EApi.cxx
> @@ -41,9 +41,10 @@
>  #include "EApi.h"
>  #endif
>  static char *eBookLibNames[] = {
> -	"libebook.so.8",	 // evolution-2.0
> +	"libebook-1.2.so.9", // ****** evolution-2.8 *****
> +	"libebook-1.2.so.5", // evolution-2.4 and 2.6+
>  	"libebook-1.2.so.3", // evolution-2.2
> -	"libebook-1.2.so.5"	 // evolution-2.4 and 2.6+
> +	"libebook.so.8"      // evolution-2.0
>  };
> 
> 	WTF is going on !? why was this done ? what was the reason for this
> [ FWIW it seems that no other code changes were necessary (so far) ].


We discussed about the calendar changes in IRC but I believe it is
libebook that you really care about.

The changes in question are as follows :
http://cvs.gnome.org/viewcvs/evolution-data-server/addressbook/libebook/e-contact.h?r1=1.20&r2=1.21
http://cvs.gnome.org/viewcvs/evolution-data-server/addressbook/libebook/e-contact.h?r1=1.21&r2=1.22

The bugs that drove this change are :

      * http://bugzilla.gnome.org/show_bug.cgi?id=313533
   [ Allow EContactPhoto to refer to images with URIs]

      * http://bugzilla.gnome.org/show_bug.cgi?id=259536 
   [Missing Gadu-Gadu Messenger in Contact entry]

This was not a trigger-happy change. 

The patch for the first bug was submitted on 15 Aug 2005 (almost a year
ago) and that for the second on 3 Mar 2004 ( more than 2 years ago) and
had been in the back-burner only because we had wanted to preserve the
ABI.


The #313533 patch was vital for Ross Burton's dbus-based EDS and running
EDS on devices (say Nokia 770) would not be possible w/o this change.
The #259536 patch had been blocking a large number of users in Poland
where Gadu Gadu is the defacto IM standard.

In Bugzilla (cf. above links) and elsewhere, there have been vocal
demands for absorbing these patches. See my comment -
http://bugzilla.gnome.org/show_bug.cgi?id=259536#c22 We did aggregate
the changes before the break.

IMO, a single break after two major releases against such gains is
justifiable. These patches were reviewed and approved by the addressbook
hackers before the bump. 

Could these improvements have been absorbed w/o breaking the ABI and
with not too much additional costs in the process?

I do not know - nor did the patch authors or the reviewers. But if you
could show us, I am all willing to learn.

Can you also help me understand the severity of the break in OO.o ?
Why are the library files with SONAMEs hard-coded in a source file ?
Why would you not use (say) pkg-config instead ?


--Harish