Re: timeout implementation issues

Bruce Momjian wrote:<br /><blockquote cite="mid200204080317.g383Hj511314@candle.pha.pa.us" type="cite"><pre
wrap="">HiroshiInoue wrote:<br /></pre><blockquote type="cite"><pre wrap="">Bruce Momjian wrote:<br /></pre><blockquote
type="cite"><blockquotetype="cite"><blockquote type="cite"><pre wrap="">I guess it's a matter of definition: Do you
considerSET variables<br />database state or session metadata?  I think some are this and some are<br />that.  I'm not
surehow to draw the line, but throwing everything from one<br />category into the other isn't my favorite solution.<br
/></pre></blockquote><prewrap="">You seem to be suggesting that we should make a variable-by-variable<br />decision
aboutwhether SET variables roll back on ABORT or not.  I think<br />that way madness lies; we could spend forever
debatingwhich vars are<br />which, and then who will remember without consulting the documentation?<br /><br />I feel
weshould just do it.  Yeah, there might be some corner cases<br />where it's not the ideal behavior; but you haven't
convincedme that<br />there are more cases where it's bad than where it's good. You sure<br />haven't convinced me that
it'sworth making SET's behavior<br />nigh-unpredictable-without-a-manual, which is what per-variable behavior<br
/>wouldbe.<br /></pre></blockquote><pre wrap="">I am with Tom on this one.  (Nice to see he is now arguing on my
side.)<br/></pre></blockquote><pre wrap="">I vote against you. If a variable is local to the session, you<br />can
changeit as you like without bothering any other user(session).<br />Automatic resetting of the varibales is rather
confusingto me.<br /></pre></blockquote><pre wrap=""><br />I don't see how this relates to other users.  All SET
commandsthat can<br />be changed in psql are per backend, as far as I remember.</pre></blockquote> Per backend or per
session?<br/><blockquote cite="mid200204080317.g383Hj511314@candle.pha.pa.us" type="cite"><pre wrap=""><br /><br
/></pre></blockquote><br/><br />