https://bugzilla.wikimedia.org/show_bug.cgi?id=48625

Marc A. Pelletier <m...@uberbox.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|---                         |WONTFIX

--- Comment #22 from Marc A. Pelletier <m...@uberbox.org> ---
It is pointless and expensive, and can be done in a line or two of PHP or perl;
and probably no more in pretty much any other language.  Do you not realize
that the query above causes an index lookup and row fetch in the database for
every single page?

Indeed, given that query, you more than triple the resources required to run
the query, as well as increase the size of the result set by sending redundant
data at every row.

You /have/ an alternative:  use one line of code at the beginning of your
script to fetch the namespaces info, and look the name up when outputting it.

Does that mean small changes in scripts that used to rely on offloading their
work to the database?  Yes.  It's certainly a much smaller change than was
required when we added edit tokens, for instance.  APIs and interface evolve. 
Obsolete junk is pruned all the time.  This is one of those times.

Reopening this bug will not change those facts, nor will it magically alter how
databases work to make it sane to push what is a trivial hash lookup in the
application into an expensive join at the database.  So please don't; I'd
rather not have to start cracking down on disputes in bugzilla (which isn't the
right venue for this anyways).

-- 
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
Wikibugs-l mailing list
Wikibugs-l@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikibugs-l

Reply via email to