[rfc] avoid assert statements?

John Arbash Meinel john at arbash-meinel.com
Mon Oct 1 19:28:09 BST 2007


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Aaron Bentley wrote:
> Nicholas Allen wrote:
>> I would think that you would want to keep the assertions in bzr.dev as
>> an extra safety check when running selftest. The test suite would need
>> to complete without any assertions failing. When the user is using
>> bzr.dev though the assertions would be disabled for performance.
> 
> It's hard to have some commands run with -O while selftest runs without it.
> 
> And since we have no control over how bzr selftest is run, it is best if
> -O does not disable safety checks.
> 
> Aaron

Well, 'make check' actually runs with -O. So the pqm *always* (and only)
runs that way. (It used to run both, but Martin changed it because it
seemed redundant and just made things take a lot longer.)

John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHATw5JdeBCYSNAAMRAk8EAKDItvx9Jstx37J4W4nVabjY5AgvOgCfdZp/
e7kKap4oyTIb2lDkQA6Wm1o=
=XGHX
-----END PGP SIGNATURE-----



More information about the bazaar mailing list