On Dec 18, 2009, at 4:14 PM, gharris999 wrote: > > andyg;496875 Wrote: >> On Dec 18, 2009, at 3:39 PM, gharris999 wrote: >> >>> >>> If I'm running 7.5 embedded (on Fedora 12) from the svn code (svn >>> 29654), should the Settings->Information page report the SBS version >> as >>> "Version: 7.5.0 - rTRUNK @ UNKNOWN" ? I.e. should the Embedded >> branch >>> be reporting itself as TRUNK? >> >> That's normal. >> >>> >>> In another post, Andy states that the embedded branch now has >>> "In-process scanning (no scanner.pl)". Yet when I initiate a "scan >> for >>> new", a separate scanner process is launched: >> >> Probably because that function is broken in the web UI. Enable >> auto-rescan and then just leave it alone, and see if it does the right >> thing. > Thanks, and yes, you're right. Once I re-enabled "Automatically detect > changes", the separate scanner.pl process no longer got spawned. > > FYI: my "scan for new" was initiated via the CLI, not the webUI. > > Is this by design? I.e. if "Automatically detect changes" is disabled, > the separate scanner.pl gets used? Or is this a loose end that you'll > tidy up once your bigger fish come out of the fryer? While I understand > that the in-process scanner is important for TinySC, I actually kind of > like the separate scanner on FatSC. And I'm not seeing any speed > advantage to the in-process scanning on my system.
Yes we'll tidy it up later. Right now all focus is on TinySC. It's not for speed, it's to save memory and allow you to use your library while it's being scanned (when using SQLite). Speed in-process is slightly slower than a dedicated scanner.pl but not that much slower. > For now, should the separate & in-process scanners produce the same > result? Yes, exactly the same. _______________________________________________ beta mailing list beta@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/beta
