On Thu, Oct 9, 2008 at 6:01 PM, Toby Johnson <[EMAIL PROTECTED]> wrote: > Alexander Gavrilov wrote: > > On Thu, Oct 9, 2008 at 9:38 AM, Toby Johnson <[EMAIL PROTECTED]> wrote: > > > Hi Alexander, sorry it took me a while to get back... the patch looks fine > to me on the face of it, but then again I'm not really familiar with the > differences between the two versions of SQLite. Is the speed of using either > method comparable? Are you familiar with whether the SQLite3 Perl module is > available on Windows via ActiveState? (I haven't looked yet). > > > Yes, SQLite 3 is available as DBD::SQLite. But I didn't see any noticeable > differences in performance: switching the DB to a ramdisk is a lot > more effective. > So the only argument for switching is that SQLite2 is sort of obsolete. On > the > downside, it may bring hidden bugs, like the one in that hunk which adds > '+0' to > a request to make it work. > > > Well, if you really felt up to it you could abstract the database stuff out > into version-specific modules, then detect which was installed at runtime. > What platform are you currently running on again?
I'm currently using a Fedora 9 x86_64 system with 8GB RAM. The previous one was a 32-bit system with only 512MB. On an unrelated topic: which combination of actions on the same path does SVN allow within one commit in the dump file? E.g. git allows any sequence of commands in its fast-import format, but remembers only the final state. I suspect that SVN is more restrictive. Alexander _______________________________________________ vss2svn-users mailing list Project homepage: http://www.pumacode.org/projects/vss2svn/ Subscribe/Unsubscribe/Admin: http://lists.pumacode.org/mailman/listinfo/vss2svn-users-lists.pumacode.org Mailing list web interface (with searchable archives): http://dir.gmane.org/gmane.comp.version-control.subversion.vss2svn.user
