Re: [ZODB-Dev] mkzeoinstance

2010-04-22 Thread Jim Fulton
On Thu, Apr 22, 2010 at 2:41 PM, Tres Seaver wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Jim Fulton wrote: >> ZODB includes a script named mkzeoinstance that tries to create >> a somewhat self-contained ZEO server installation. >> >> It only works when ZODB and its dependencies ar

Re: [ZODB-Dev] mkzeoinstance

2010-04-22 Thread Tres Seaver
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jim Fulton wrote: > ZODB includes a script named mkzeoinstance that tries to create > a somewhat self-contained ZEO server installation. > > It only works when ZODB and its dependencies are installed in > in the standard Python path (typically site-pa

[ZODB-Dev] Death to _p_independent

2010-04-22 Thread Jim Fulton
https://bugs.launchpad.net/zodb/+bug/124613 proposed getting rid of _p_independent for BTree length objects. The _p_independent is a relic of not having MVCC. While not all storages have MVCC, the ones that matter do. BTree length objects are the only objects that use _p_independent. I propos

[ZODB-Dev] mkzeoinstance

2010-04-22 Thread Jim Fulton
ZODB includes a script named mkzeoinstance that tries to create a somewhat self-contained ZEO server installation. It only works when ZODB and its dependencies are installed in in the standard Python path (typically site-packages). It only works on unix-like systems. It has no tests. This is an