Re: Failing assertions in indxpath.c, placeholder.c and brin_minmax.c

<div dir="ltr"><br /><div class="gmail_extra"><br /><div class="gmail_quote">On Sun, Jul 26, 2015 at 10:32 PM, Andreas
Seltenreich<span dir="ltr"><<a href="mailto:seltenreich@gmx.de" target="_blank">seltenreich@gmx.de</a>></span>
wrote:<br/><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span
class="">MichaelPaquier writes:<br /><br /> >> Footnotes:<br /> >> [1]  <a
href="https://github.com/anse1/sqlsmith"rel="noreferrer" target="_blank">https://github.com/anse1/sqlsmith</a><br />
><br/> > This is really interesting stuff. I think that it would be possible to<br /> > extract self-contained
testcases from your tool and those queries to<br /> > reproduce the failures. It is written that this tools connects
toa<br /> > database to retrieve the schema, what is it exactly in the case of<br /> > those failures?<br /><br
/></span>Iused the database "regression" that pg_regress leaves behind when you<br /> remove the --temp-install from
it'sdefault invocation through make<br /> check.  Sorry about not being explicit about that.<br /><br /> So, dropping
oneof the queries into src/test/regress/sql/smith.sql and<br /> invoking<br /><br />     make check
EXTRA_TESTS=smith<br/><br /> was all that was needed to integrate them.  I was then able to perform<br /> "git bisect
run"on this command.  Er, plus consing the expected output<br /> file.<br /><br /> I'm using the regression db a lot
whenhacking on sqlsmith, as it<br /> contains much more nasty things than your average database.<span
class="HOEnZb"><fontcolor="#888888"><br /></font></span></blockquote></div><br /></div><div class="gmail_extra">Ah, OK.
Thanks.The code is licensed as GPL, has a dependency on libpqxx and is written in C++, so it cannot be integrated into
coreas a test module in this state, but I think that it would be definitely worth having something like that in the
codetree that runs on the buildfarm. We could have caught up those problems earlier.  Now I imagine that this is a
costlyrun, so we had better have a switch to control if it is run or not, like a configure option or a flag.
Thoughts?<br/>-- <br /><div class="gmail_signature">Michael<br /></div></div></div>