Re: ExecEvalVar does not have appropriate ExprState? - Mailing list pgsql-hackers

Thanks for the confirmation. I am happy to have understood some basics of PG. <br /><br />Also, I think the similar
standsfor the Const, Param,<br /> CoerceToDomain and CaseTest nodes, right? They too cast the pointer to another type.
<br/><br />Thanks for the help Heikki. (I could better call you HL, if you dont mind :P )<br /><br />Regards,<br
/>Vaibhav<br/><br /><br /><div class="gmail_quote">On Mon, Mar 28, 2011 at 1:05 PM, Heikki Linnakangas <span
dir="ltr"><<ahref="mailto:heikki.linnakangas@enterprisedb.com">heikki.linnakangas@enterprisedb.com</a>></span>
wrote:<br/><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div
class="im">On28.03.2011 10:02, Vaibhav Kaushal wrote:<br /><blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px#ccc solid;padding-left:1ex"> Hi,<br /><br /> I see the ExecInitExpr says something like this:<br
/>------------<br /> case T_Var:<br />        state = (ExprState *) makeNode(ExprState);<br />      
 state->evalfunc= ExecEvalVar;<br /> ---------------<br /><br /><br /> But the ExecEvalVar function definition
says:<br/><br /> ------------------------<br /> Var                *variable = (Var *) exprstate->expr;<br />      
 TupleTableSlot*slot;<br />        AttrNumber      attnum;<br /><br />        if (isDone)<br />                *isDone
=ExprSingleResult;<br /><br />        /* Get the input slot and attribute number we want */<br />        switch
(variable->varno)<br/>        {<br />                case INNER:                             /* get the tuple from
theinner node */<br /> -------------------------<br /><br /><br /> Since ExprState ->  expr in its final form would
finallycontain only:<br /><br /> NodeTag         type;<br /><br /> I think that the pointer being cast in the
ExecEvalVaris actually<br /> already a form of Var which is passed around as Expr for sake of<br /> function call.<br
/></blockquote><br/></div> Right, exprstate->expr is a Var in ExecEvalVar.<div class="im"><br /><br /><blockquote
class="gmail_quote"style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> So, was the node in the Expr
treefor the corresponding ExprState node<br /> of the ExprState tree actually a 'Var'?<br /></blockquote><br /></div>
Yes.<divclass="im"><br /><br /><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc
solid;padding-left:1ex">Also, I have tried, but failed to fidn the exact place where the Expr<br /> tree is created.
Justpointing me to the file / function which does this<br /> would be of great help.<br /></blockquote><br /></div> A
rawexpression tree is created in the grammar, src/backend/parser/gram.y. It is then transformed in parse analysis phase
tothe form the planner accepts, in transformExpr(). The planner can do some further transformations, like replacing
immutablefunction calls with Consts.<br /><font color="#888888"><br /> -- <br />  Heikki Linnakangas<br />
 EnterpriseDB  <a href="http://www.enterprisedb.com" target="_blank">http://www.enterprisedb.com</a><br
/></font></blockquote></div><br/>