[lsb-discuss] error in POSIX.os

[Wichmann, Mats D]

Thanks Andrew - I had promised to email some infomration on
how to rerun just failed tests and you have saved me the
work of looking it up!!!!

-- mats



-----Original Message-----
From: lsb-discuss-bounces at lists.freestandards.org on behalf of Andrew Josey
Sent: Mon 7/17/2006 3:47 AM
To: lsb-discuss at lists.freestandards.org
Subject: Re: [lsb-discuss] error in POSIX.os
 

On Mon, 2006-07-17 at 05:25 +0100, nirupama rout wrote:
> I did the runtime test for 3.0.The tjreport is 
> 
> 
> /tset/LI18NUX2K.L1/utils/uniq/T.uniq 3 FAIL
> /tset/POSIX.os/dataform/tar/T.tar 17 UNRESOLVED
> /tset/POSIX.os/dataform/tar/T.tar 23 UNRESOLVED
> 
> 
It would be useful if you could use the vrpt report writer to output the
full supporting information with the test result (assertion and
strategy).

A good technique is to re-run just the failures and unresolved tests
after the initial run, so you get those results in a self contained
journal, so if for example your first run created the journal named
results/0001e/journal

in the vsx0 home directory type

    tcc -p -e -r FAIL,UNRESOLVED results/0001e/journal

The -p argument is to have tcc report on progress, so it emits a line
for each test executed
The -e argument is the mode of operation, in this case e for execute
The -r argument to tcc is rerun mode,
FAIL and UNRESOLVED are result codes that are being selected for the
rerun in this example
the last argument "results/0001e/journal" is the path to the old journal
file that you want to use for determining test selection

Assuming it put the new results into the file results/0002e/journal ,
the following will generate a more verbose journal file and save it in
the file jout 

   vrpt results/0002e/journal > jout

Andrew


_______________________________________________
lsb-discuss mailing list
lsb-discuss at lists.freestandards.org
http://lists.freestandards.org/mailman/listinfo/lsb-discuss