AFAIK, both these issues are resolved:

     - jbj: RPM_I18NSTRING_TYPE is still wonky, one test needed is to  
see
         whether rpm-python h['summary'] returns a string (which is  
correct)
         or an argv tuple (which is broken). There's more to do here:
         $ rpm -qa --yaml
rpm: header.c:1897: rpmheRealloc: Assertion `he->c == 1' failed.
         You need packages with Summary(lo): to see the assertion  
failure.

     - jbj: HEAD segfaults under "devtool standalone"'s test-drive as it
            builds RPM with prefix="/tmp/rpm" (see line 309 in
            devtool.conf). When one uses a longer prefix, e.g.
            prefix="/tmp/rpm-xxx", the segfault disappears!!!
            20071215: rse: I currently can no longer reproduce this  
segfault with
                      the latest HEAD as of today, even with prefix=/ 
tmp/rpm.

I will remove if/when you wish. I have only a single instance in a  
single
package as a test case, or my hand-crafted "time" test package, to  
back up
my claim of "resolved".

Certainly
	rpm -qa --yaml
WORKSFORME with a "time" test package containing

     Summary: A GNU utility for monitoring a program's use of system  
resources
     Summary(de): DE: A GNU utility for monitoring a program's use of  
system resources
     Summary(es): ES: A GNU utility for monitoring a program's use of  
system resources
     Summary(fr): FR: A GNU utility for monitoring a program's use of  
system resources

Because of (my belief) that the same RPM_I18NSTRING_TYPE issue was  
responsible
for the devtool segfault, which I have never seen and can't reproduce  
in my environment,
I've left both in the TODO list.

Again, AFAIK, the RPM_I18NSTRING_TYPE problems are resolved.

73 de Jeff

On Dec 15, 2007, at 6:12 AM, Ralf S. Engelschall wrote:

>   RPM Package Manager, CVS Repository
>   http://rpm5.org/cvs/
>    
> ______________________________________________________________________ 
> ______
>
>   Server: rpm5.org                         Name:   Ralf S. Engelschall
>   Root:   /v/rpm/cvs                       Email:  rse@rpm5.org
>   Module: rpm                              Date:   15-Dec-2007  
> 12:12:56
>   Branch: HEAD                             Handle: 2007121511125600
>
>   Modified files:
>     rpm                     TODO
>
>   Log:
>     5.0b2 is coming today and the segfault issue has to be gone at  
> last
>     for 5.0b4 or we have to be really sure that it doesn't happen in
>     regular production environments
>
>   Summary:
>     Revision    Changes     Path
>     1.75        +7  -5      rpm/TODO
>    
> ______________________________________________________________________ 
> ______
>
>   patch -p0 <<'@@ .'
>   Index: rpm/TODO
>    
> ====================================================================== 
> ======
>   $ cvs diff -u -r1.74 -r1.75 TODO
>   --- rpm/TODO	14 Dec 2007 16:49:35 -0000	1.74
>   +++ rpm/TODO	15 Dec 2007 11:12:56 -0000	1.75
>   @@ -48,11 +48,7 @@
>      resolved during release engineering. Please do not just list any
>      *CANDO* items.
>
>   -  o to be resolved before RPM 5.0b2:
>   -    - jbj: HEAD segfaults under "devtool standalone"'s test- 
> drive as it
>   -           builds RPM with prefix="/tmp/rpm" (see line 309 in
>   -           devtool.conf). When one uses a longer prefix, e.g.
>   -           prefix="/tmp/rpm-xxx", the segfault disappears!!!
>   +  o to be resolved before RPM 5.0b3:
>        - jbj: RPM_I18NSTRING_TYPE is still wonky, one test needed  
> is to see
>    	whether rpm-python h['summary'] returns a string (which is  
> correct)
>    	or an argv tuple (which is broken). There's more to do here:
>   @@ -60,6 +56,12 @@
>    rpm: header.c:1897: rpmheRealloc: Assertion `he->c == 1' failed.
>    	You need packages with Summary(lo): to see the assertion failure.
>
>   +  o to be resolved before RPM 5.0b4:
>   +    - jbj: HEAD segfaults under "devtool standalone"'s test- 
> drive as it
>   +           builds RPM with prefix="/tmp/rpm" (see line 309 in
>   +           devtool.conf). When one uses a longer prefix, e.g.
>   +           prefix="/tmp/rpm-xxx", the segfault disappears!!!
>   +
>      o to be resolved before RPM 5.0.0:
>        - rse: xar/ has weak (think RPM_CHECK_LIB) support its third- 
> party
>               libraries LibXML, OpenSSL, etc. But we at least  
> support building
>   @@ .
> ______________________________________________________________________
> RPM Package Manager                                    http://rpm5.org
> CVS Sources Repository                                rpm-cvs@rpm5.org