[OPEN-ILS-GENERAL] metabib.browse_entry unused cleanup?

[Mike Rylander]

Josh,

It's not causing any issues in any instance that EOLI oversees.  It's
a unique string listing based on a B+Tree index, so until you get past
several quintillion entries you should be fine.  The one benefit of
removing orphaned entries would be for the off chance that the terms
in the string are collectively entirely unique and you've disabled the
visibility test for autosuggest, where you might get a browse
suggestion that would lead to a dead end.

If you do clear them out, make sure you vacuum and analyze the table
(/not/ vacuum full, mind) afterwords.

HTH,

--
Mike Rylander
 | President
 | Equinox Open Library Initiative
 | phone:  1-877-OPEN-ILS (673-6457)
 | email:  miker at equinoxinitiative.org
 | web:  http://equinoxinitiative.org


On Mon, Apr 2, 2018 at 4:10 PM, Josh Stompro <stomproj at exchange.larl.org> wrote:
> Hello, I’ve been changing around some indexing rules and applied the fix to
> the full stop/rda relator code normalization issue and now just noticed that
> all the no longer used browse entries are sticking around in
> metabib.browse_entry.
>
>
>
> There are 175K entries out of 950K that are no longer used according to
> metabib.browse_entry_def_map and metabib.browse_entry_simple_heading_map.
>
>
>
> Has anyone found this to be an issue and removed those unused entries?  I
> don’t think this is causing any issues, it just seems strange to hold on to
> that data.  Any typo that was ever entered in a browse field and corrected
> is recorded there.
>
>
>
> Thanks
>
> Josh
>
>
>
> Lake Agassiz Regional Library - Moorhead MN larl.org
>
> Josh Stompro     | Office 218.233.3757 EXT-139
>
> LARL IT Director | Cell 218.790.2110
>
>